Contents · – a presentation of the ITU-T T.120 series recommendations, Trond Ulseth..... 95 Charging of ATM services, ... In today’s society investment in competence is said

Contents

Guest editorial, Tove Kristiansen ................................. 1

Telecommunications in distance education– An introduction, Tove Kristiansen ............................. 2

Distance education – a response to social needs or technology in search of application? Erling Ljoså ....... 9

Field trials – the need for user participation intelecommunications development, Tove Kristiansen .. 15

Cultural and educational barriers to tele-education– Experiences from the use of videoconferencing, Tove Kristiansen ......................................................... 20

Distance education in the electronic classroom, Bjørn Hestnes and John Willy Bakke ......................... 22

Something in the air – Norwegian experiences with telelecturing in flexible education, Gunnar Grepperud ..................................................... 29

Telecom and Open Learning – a challenge toinstitutional cooperation, Harald Haugen and Bodil Ask .................................... 34

From electronic mail to Internet – where does it take us? Torstein Rekkedal ................ 38

Design guidelines for WWW-based courseenvironments, Anke Eekma and Betty Collis .............. 44

Learning in collaborative virtual environments– Impressions from a trial using the Dovre framework, Ola Ødegård and Karl Anders Øygard ...................... 51

Distance education in Norwegian manufacturingindustry and the education sector – Currentsituation and future needs, Tom Erik Julsrud ............. 59

The globalisation of education, Robin Mason ............ 69

The future of learning, Anthony W. Bates .................. 82

Telektronikk

Volume 92 No. 3/4 - 1996ISSN 0085-7130

Editor:Ola EspvikTel. + 47 63 84 88 83

Status section editor:Endre SkoltTel. + 47 63 84 87 11

Editorial assistant:Gunhild LukeTel. + 47 63 84 86 52

Editorial office:TelektronikkTelenor AS, Telenor Research & DevelopmentP.O. Box 83N-2007 Kjeller, Norwaye-mail: [email protected]

Editorial board:Ole P Håkonsen, Senior Executive Vice PresidentOddvar Hesjedal, Vice President, Research & DevelopmentBjørn Løken, Director

Graphic design:Design Consult AS

Layout and illustrations:Gunhild Luke, Britt Kjus, Åse AardalTelenor Research & Development

Feature

Status

Kaleidoscope

Special

International research and standardization activities in telecommunication: Introduction, Endre Skolt ..... 125

COST 219: Future telecommunications and tele-informatics facilities for disabled people and elderly, Per Helmersen .......................................................... 126

COST 235: Radiowave propagation effects on next generation fixed-services telecommunication systems, Agne Nordbotten ........................................ 128

COST 231: Evolution of land mobile radio(including personal) communications, Per Hjalmar Lehne and Rune Harald Rækken ......... 131

50 years of radio links in Norway. The transition from the telecommunication Stone Age, Knut Endresen .. 141

A presentation of the authors ................................... 168

Protocols for multimedia multiparty conferencing– a presentation of the ITU-T T.120 series recommendations, Trond Ulseth ................................ 95

Charging of ATM services, Ragnar Andreassen,Øyvind Breivik, Einar Edvardsen, Thor Eskedal,Nina Kloster and Olav Østerbø ............................... 105

The frequency assignment algorithm used in themobile network planning tool MOBINETT,Ralph Lorentzen ....................................................... 118

Modern information andtelecommunications technologyis about to revolutionize the fieldof distance education. Only a fewyears ago, all distribution oflearning material and communi-cation between teacher andstudents was totally based onpostal services. Today, we see amultitude of media and commu-nication technologies being used.Even though mail is still themedium most used in distanceeducation both nationally andinternationally, the use oftelecommunications is spreadingrapidly and is expected tobecome dominating in our part ofthe world early in the next cen-tury. That is why we have chosento call this issue of Telektronikk“Tele-education” and not dis-tance education.

What we witness in Norway, asin many other countries, is thatthe differences between distanceeducation and ordinary education are slowly being wiped out.New technology is being brought into the educational settings,opening up for a globalisation of education. Distances do notmatter any more. Students may collaborate with students inother countries just as easily as with students next door. Expertsmay give lectures to all parts of the world without having totravel. Employees may study from home, in their work place or

at a local study centre. They maychoose whether to study at theirown pace using asynchronousmedia, to follow a fixed sched-ule being connected to theirteacher and fellow studentsthrough synchronous media, orusing a mixture of the two.

This independence of time andplace admits the learner to amuch larger degree than before,the freedom to choose a learningmodel which suits the individualbest. We are witnessing thedevelopment of a more individu-alized system of distance educa-tion, which we may call learningon demand. This respresents achallenge to all established edu-cational institutions, which willhave to become more learneroriented than they are today.

The flexibility of modern com-munications technologies, is ofgreat importance to the growing

demand for continuous and life long learning in the work force.In today’s society investment in competence is said to be themajor factor for companies to survive in the increasing competi-tion. This need for continuous updating of the employees’ com-petence, can greatly improve in efficiency by means of tele-edu-cation.

1

Guest editorialB Y T O V E K R I S T I A N S E N

Telektronikk 3/4.1996

2

Telecommunications in distance education– An introduction

B Y T O V E K R I S T I A N S E N


1 What is distance

education?

Distance education is teaching wherethe teacher and pupil(s)/student(s) areremote in terms of space and/or time.Technical aids are used to disseminatestudy materials and to provide genuinetwo-way communication, as a meansof supporting the learning process.(St.meld. (Government White Paper)No. 43, 1988–89, p.88.)

The history of distance education startedsome one hundred years ago. The pur-

pose was to offer the opportunity to learnto remote regions and groups of the pop-ulation with limited access to ordinaryeducational services. For many years cor-respondence schools were the only onesto offer this kind of courses. During thelast ten years, however, other educationalenvironments have shown an increasedinterest in distance education.1 This isdue in particular to two aspects: societalchanges and technological innovation.

Demographic changes, industrial re-structuring and the pressure exerted on usthroughout our lifetimes to continuallyupdate what we know and to acquire newskills means that our need for continuingand refresher education is greater thanever. Improving the skills of employeesis an increasingly important investmentarea for all types of companies. The pro-vision of continuing and refresher educa-tion for the work force has to a greaterextent become the responsibility of theindividual employer. At the same timethere is a greater demand for improvedefficiency and better exploitation ofresources. This includes the savings ontravel and course budgets and the valuefor the company of investments made inits employees’ education.

This increased need for learning meansthe education sector faces new chall-enges. New groups of students with spe-cific educational needs are emerging.These groups consist primarily of adultsin employment, with families and socialties which restrict their mobility. Learn-ing must take place where they are, in theworkplace, in the home or at a localstudy centre.

In this context the development of tech-nology is important. On the one hand,technological developments help speedup the social changes which create theneed for continual learning. On the otherhand, the technology provides newopportunities for this learning to takeplace without the student needing totravel to an educational establishment.Via the telecommunication system,teaching sessions can be distributed tothe student, no matter where he or she is

located. As compared to postal commu-nication, the student is able to enjoy a farmore direct form of two-way communi-cation with the tutor, and with other dis-tance students. Depending on what thesituation of the students are, what theirneeds are, and what subject is to betaught, the course may be offered viasynchronous or asynchronous media,based on audio, text or video communi-cation. Very often a mixture of mediawill be used, combined with ordinaryclassroom instruction and group work.As the technological development goesfurther, the number of possible combina-tions will grow and the individual studentwill to a larger extent have the possibilityto choose media and learning models thatare best suited to meet his or her needs.

2 The “generations” of

distance education

The development in distance educationmay historically be divided into variousgenerations, according to what kinds oftechnological devices that have beenused at various times to establish two-way communication between teacher andstudents.

One way of presenting this developmentis the following model of three genera-tions:

First generation was characterised by theuse of mail only. The contact was basedon written communication transported bythe postal system and ran between theteacher and the individual student (ref.Figure 1).

Second generation used broadcasting(radio and television) and audio andvideo cassettes distributed by post toreach the students with teaching material.The two-way interaction was still mainlymail-based, but supplemented by the useof the telephone (ref. Figure 2).

Third generation saw the introduction oftools that opened up for a greater varietyand flexibility of interaction. Electronicmail and computer based conferencingrepresented asynchronous communica-tion that proved to be stimulating andmotivating to both teachers and students.An important aspect was that it allowedfor students to interact with each other.This generation also saw the introductionof videoconferences and videophones(ref. Figure 3).

Teacher Students

Teacher Students

RADIOTV

(Audioand videocassettes)

Figure 1 First generation of distance education

Figure 2 Second generation of distance education

1 A more thorough presentation of thehistorical development of distanceeducation is given in the article “Dis-tance education – response to socialneeds or technology in search of appli-cation?” in this issue of Telektronikk.

3Telektronikk 3/4.1996

Following these three generations, afourth generation is about to take shape.This is the generation where all commu-nication has turned digital, multimediatechnology is being introduced into themarket, and the interactivity and flexi-bility of both technology and learningand teaching models are much improved.

3 Telenor’s engagement

in tele-education

Telenor (Norwegian Telecom) has sincethe mid 1980s been working in the fieldof distance education and has during aten year period run a large number offield trials in this area.2 These trials havehad two objectives. The first was to testpreviously established services to gainbetter insight into the advantages and dis-advantages they offer when used for dis-tance education purposes. The secondobjective was to devise new telecommu-nication services which could be used forthis purpose, by testing new equipment inconditions specific to distance education.By doing so we have been able to gatherinformation from users which is of greatvalue to us when developing these ser-vices, material which not only helps tech-nological development but which alsosheds considerable light on the educa-tional and communicational aspects oftele-based distance education.

The field trials have partly been initiatedby Telenor, partly by other establish-ments. Some of the trials have been jointprojects. Others have been carried outwithout Telenor’s direct involvement,though Telenor has been responsible forthe assessment of these. Telenor hasemphasised the need for a thorough eval-uation of all the trials the company hasbeen involved in, and the reports pro-duced contain a considerable amount ofuser findings.

There are many circumstances that canaffect the outcome of a trial. Key aspectsin every distance teaching situation areeconomic, educational and organisationalconsiderations. The extent to which agiven medium is suited to a specific dis-tance education context will thereforedepend on a number of factors and theway these factors interact. The technol-ogy will, in other words, be only one ofmany aspects that must be evaluated

when selecting a suitable communicationaid for distance education purposes.However, the technology is important initself, because it will to a great extentdetermine the form of communication. Itopens the way for certain possibilitieswhile simultaneously setting limitations.In the following we will present some ofthe experiments Telenor R&D has beeninvolved in and experiences gained fromthem.

4 Audio-based

communication

The telephone is internationally a muchused communication aid in distance edu-cation. A number of countries have manyyears of experience in utilising telephoneconferences as a way of teaching andgathering groups of students for dis-cussion purposes. In Norway the tele-phone has primarily been used by corre-spondence schools as a means of directcontact between the tutor and individualstudents. Open arrangements allowingstudents to call the tutor/school when theneed arises have shown that it is difficultto get the student to call on his or herown initiative. However, there is plentyof evidence indicating that direct contactover the telephone is an important meansof motivating and supporting the studentin a period of private study. The corre-spondence schools have therefore gradu-ally incorporated telephone contact withthe students as part of the work-load ofcorrespondence school tutors.

Telephone conferences

In telephone conferencing it is possibleto link up several locations simultane-ously to form a group “meeting-place”over the telephone. In this way contactcan easily be made between tutors andstudents who live a long way from oneanother. Telephone conferencing can beused to spread information, to maintainsocial contact and for professional activi-ties.

One of the first trials Telenor R&D tookpart in, during the winter of 1988–89,was testing telephone conferencing as asupplementary service in adult education.The trial formed part of a project at ahigh school in the northern part of Nor-way, where an adult education course inhistory was started in the autumn of1988. A total of 30 people applied to takepart in the course, which was based on a

correspondence course. Since they livedfar apart, there was no point in arrangingcombined teaching, as is often used inNorway, i.e. local classes are held onceevery week or fortnight as a supplementto the correspondence course. In this pro-ject, as an alternative, weekly classes ofone hour duration were broadcast onlocal radio.

In addition, telephone conferences wereused to bring the course participantscloser to one another and to create a“classroom environment” despite thegreat distances involved. This workedvery well. Most of the participants hadnever heard of telephone conferencingbefore the trial commenced, and therewas little enthusiasm among the partici-pants for testing this facility. After tryingtelephone conferencing a couple oftimes, however, the participants were sopleased with the benefits gained that theywished to continue using it.

Interviews with the participants showedthat telephone conferencing was import-ant both from a social and an educationalpoint of view. Seen in terms of socialbenefits, it would appear that telephone

2 There is a separate article in this issueof Telektronikk on “Field trials”.

Figure 3 Third generation of distance education

4 Telektronikk 3/4.1996

Another important advantage of tele-phone conferences is their flexibility.They can be established between indi-vidual students or groups of students, andthe location of the participants is of noconsequence. Within the field of distanceeducation, telephone conferencing is con-sidered to have great growth potential,considering the widespread distributionof the telephone, its user-friendliness andlow level of cost.

5 Visual communications

Visual media used for distance educationpurposes can be divided into two maincategories: one-way and two-way com-municating. Television is a one-waycommunicating visual medium, whichhas played an important role as a mediumfor distributing educational programs tothe population both in Norway and othercountries. It may be transferred via thenational or private broadcasting net-works, local cable or satellite. One-waycommunicating media provide fewopportunities for feedback from studentto tutor. Two-way communication can beprovided, however, by combining tele-vision teaching with other media, e.g.telephones, fax machines, and/or com-puters. In Norway, we have made someexperiences with satellite distribution oflectures combined with telephone con-ferences.3 We have also made som ex-periments with videotelephony in thisconnection. They have so far been limit-ed, but very promising. Most of the stu-dents reported the use of videotelephonesto be an improvement as compared totelephone conferences, mainly becausethey experienced a closer contact withtheir teacher.

At Telenor R&D we have to a largeextent been engaged in experimentingwith two-way visual communication inthe telecommunication networks, i.e.videoconferencing and videotelephony.This is due to the fact that videoconfe-rencing was introduced as a public ser-vice in the Norwegian telecommunica-tions network in 1983, and that TelenorR&D at an even earlier stage, in the late1970s, initiated studies in the field of

image compression techniques whicheventually, in 1991, resulted in a Nor-wegian made videotelephone (producedby the Norwegian company TandbergA/S) (ref. Figure 4).

The transfer of broadcast quality real-time images requires a great deal oftelecommunication network capacity. Byusing coding techniques to digitise andcompress the information contained inthe image, the transfer capacity requiredcan be reduced. While the transfer of animage of broadcast quality requires acapacity equivalent to approximately1000 analogue telephone lines, video-telephony can be transmitted on only onesingle digital telephone line (64 kbit/s)(ref. Figure 5).

Reducing the transfer rate does result in aloss of quality. In distance education,videoconferences and videotelephony arenonetheless interesting services. By uti-lising two-way visual communicatingmedia it is possible to produce a teachingsituation which in many ways resemblesthat found in traditional classrooms:

• The teacher and students can see oneother. This gives them a feeling ofproximity and direct contact, whichhas a motivating effect.

• It is possible to visualise the teaching by

- showing text, illustrations andobjects with the aid of a documentcamera

- producing text and figures on ablackboard, on paper or a personalcomputer

• It is possible to show movement whichcan be of benefit when

- giving instruction in subjects thatinvolve motion, e.g. drama, musicand physiotherapy

- demonstrating work operations andlaboratory exercises.

The similarity to classroom teachingmakes many teachers feel comfortableusing this technology because it allowsthem to act in much the same way as theyare used to.

It should be recognised, however, thatdistance education no matter what tech-nology is being used, is a totally differentmethod for both teaching and learning.Using this medium therefore needs care-ful preparations from both teacher andstudent.

Figure 4 The Tandberg Vision videophone

3 Some Norwegian experiences fromsatellite distribution are presented inthe article “Something in the air –Norwegian experiences with telelectur-ing in flexible education” in this issueof Telektronikk.

conferencing motivated the students to“meet” others similarly situated withwhom they could discuss problems theyhad in common. In educational terms,telephone conferencing gave the studentsthe opportunity to discuss those parts ofthe syllabus that had not been coveredsufficiently in the radio broadcasts and totalk about topics which they found diffi-cult.

The course participants achieved verygood examination results, and the coursedrop-out rate was low. Although this tellsus nothing about the significance of tele-phone conferences in isolation, it doesindicate that the tested combination ofmedia worked well for the participants.

The model chosen in this project is con-sidered to be of great interest for distanceeducation, because it is based on simple,inexpensive and easily accessible formsof media with which the users were al-ready familiar. Both radio and telephonesare media we may assume can be foundin every Norwegian household. Radiobroadcasts also have the advantage thatthey can be recorded and listened to at alater juncture for revision purposes. Al-though telephone conferences demand acertain amount of conversation disciplinewhich will be unfamiliar to most peoplewho are only used to normal telephoneconversations, this is something that themajority of individuals quickly becomeaccustomed to.


Telenor R&D has been engaged in anumber of field trials using videoconfer-encing and videotelephony in distanceeducation. It all started with a projectcalled “Telematics in the North” launch-ed in 1988, in which videoconferencingwas defined as an area of investment.Telenor R&D played an active part inthis, in two areas in particular: tele-medicine and distance education. Withina short period of time, a number of loca-tions had installed videoconferencingequipment. There was an especially largeamount of activity in Finnmark (the mostnortherly part of the country), where thegreat distances between participants andlack of teachers meant that many en-vironments found this form of distanceeducation very interesting. The impactthis form of teaching could have on ruralareas and small communities was thoughtto be of great value.

Most distance education trials involvingvideoconferencing have been betweentwo locations, so-called point-to-pointconnections, where a teacher in one loca-tion has taught students in another. Insome cases a group of students have beenpresent in the same location as the teach-er, at the same time as the course hasbeen followed by a group of studentsover the telecommunications network. Inthe course of time technical develop-ments enabled the simultaneous trans-mission of a teaching session from oneeducational establishment to severallocations, and this has been utilised in anumber of contexts. Eventually, technol-ogy moved even one stage further. In thesummer of 1991 it became possible tolink up several locations in one fullmulti-point conference, that is, all loca-tions became able to see and hear oneanother (ref. Figure 6). Telenor R&D has

participated in trials involving all thesedifferent models.

The very first trial involving videocon-ferences took place in a fishing village onthe northern coast. A newly started ele-mentary course in fishery studies at thehigh school lacked qualified teachers in arange of subjects. This was compensatedfor by organising for the transmission ofdirect teaching over the telecommunica-tions network. In addition, employees inthe local fish processing industry got theopportunity of further education throughlectures given to them on the screen.

Soon other examples followed. Regionalcollege centres saw videoconferencing asa tool for reaching new categories ofstudents and students unable to attendcourses at campus. The Norwegian mili-tary service used videoconferences fordistributing courses to locations spreadall over the country. Telenor used it forinternal training of employees. Hospitalsused it for further education of doctorsand nurses.

With the introduction of videotelephony,two-way visual communication hasbecome interesting to a larger amount of

Previous situation

1telephone-

line

1000telephone-

lines

Present situation

Video andaudio codec

1 digitaltelephone

line

TANDBERG Vision TANDBERG Vision

Video andaudio codec

Figure 5

Point-to-point

Point-to-multipoint

Multipoint

Figure 6 Models for use of video-conferences


educational institutions, due to the lowercost and better accessibility as the digitalnetwork is being built out – ISDN (Inte-grated Services Digital Network). Today,there is quite a large number of institu-tions using videotelephony for educa-tional purposes in Norway.

It is used both for transferring lectures,supervision, consultation, habilitation,and discussion groups. The technicalinstallations differ from case to case,depending on the number of participantsand the subject being taught. At TelenorR&D we have made certain efforts todevelop solutions to satisfy the needs ofboth teachers and students as they havebeen expressed in field trials we havebeen involved in.4

Experiences made from all the projectsthat have been using videoconferences orvideotelephony are very much the samewhen it comes to the communicativeaspects. Both students and teachers thinkit important to see each other. They findthat “meeting on the screen” makes themfeel close. Especially the students reportthat they feel they get to know the teach-er quite well. Nevertheless, as with allkinds of distance education courses, it isrecommended that teacher and studentsmeet each other before a course starts.

All the Norwegian trials indicate thatboth teacher and students quicklybecome accustomed to the technologyand that they easily learn how to operateit. It is worth noting that the quality ofthe sound is felt to be far more crucial tosuccessful communication than the qual-ity of the images.

As all the teaching takes place on ascreen, it becomes more structured thanin the classroom. It demands a highdegree of concentration from both teach-er and students, and even though itallows for continuous interaction, experi-ences show that spontaneous interactionis hard to achieve. The dialogue has to beplanned by the teacher, and he or shemust to a large extent invite the studentsto interact. Pedagogically, this is wherethe greatest challenge lies in furtherdeveloping the use of videoconferencingand videotelephony in distance educa-

tion, to find good methods for utilisingthe possibility for two-way communica-tion.

6 Text-based

communication

The most commonly used form of textcommunication in distance education,besides hand-written letters, is computer-based communication which allows forelectronic messages, letters and contribu-tions to be transferred at high speed.There are also systems which enable thetransferral of hand-written messages viathe telecommunications network, elec-tronic digitising pads and electronicboards. At Telenor we have in co-opera-tion with the university of Oslo develop-ed an electronic classroom, combiningthe use of electronic boards with audioand live video communication.5 Also thefax machine has a great potential in dis-tance education. Telenor has tested allthese forms of text-based communica-tion. Here we would like to focus oncomputer mediated communication,which is expected to grow rapidly in theyears to come.

Computer mediated communi-cation

The use of computer mediated communi-cation in distance education provides theopportunity for two-way communicationregardless of time (asynchronous com-munication) and place. Tutors and stu-dents are able to work whenever andwherever they feel like it. Since it is aflexible and readily available form ofcommunication, computer mediatedcommunication meets many of the re-quirements of the traditional distanceeducation institutions. People’s access toa computer is nonetheless still limited, al-though the rapid rise in the number ofcomputer terminals both in the workplaceand at home makes computer mediatedcommunication an increasingly interest-ing alternative in distance education.

During the last couple of years, we haveseen a dramatic growth in the use of theInternet, and this is thought to becomethe dominating medium for computermediated communication in distance

education in the years to come. Manydistance education institutions do, how-ever, already have long and valuableexperience in the use of various comput-er conferencing systems. The varioussystems may differ in terms of user inter-face, menu system and organisation ofconferences. Common to all of them isthe fact that they include three princi-pally different forms of communication:

• Electronic mail. One-to-one commu-nication: messages, comments, ques-tions, information and answers can besent as electronic mail from one personto one or more addressed recipients.

• Electronic bulletin boards. One-to-many communication: information issent electronically from one sender asan open message to a larger number ofunaddressed recipients.

• Electronic conferences. Many-to-many communication: messages, ques-tions and comments are sent electron-ically from one conference participantto a common conference, a forumestablished for a limited number ofparticipants. The message can be an-swered by one or more of the otherparticipants. The conference thusresembles a meeting place for studentsand tutors where communicationoccurs freely among all the partici-pants.

The metaphor often used for computerconferences is the school building itself.As a student/participant you have to beadmitted. When you are “inside”, youmay choose whether to read your mail, toread a notice on the board, to be socialand chat with others in the “café”, takepart in a thematic discussion that is opento everybody or, if you are a “member”,discuss with a smaller circuit of people“behind closed doors”.

Experiences drawn from the use of com-puter conferences so far clearly show thatthe number of students who access theconference merely to read messages farexceed those who log on to the system tosend comments. The degree of activityamong students also seem to be stronglyrelated to the activity of the individualtutor, i.e. the more active the tutor, themore active his or her students.

In Norway it was the largest of the tradi-tional correspondence schools that firstbegan to utilise computer conferences asan integral part of distance education.Computer conferences proved well-suit-

5 This is presented in the article “Dis-tance education in the distant class-room” in this issue of Telektronikk.

4 Examples of this is given in the article“Field trials – the need for user partici-pation in telecommunications develop-ment” in this issue of Telektronikk.


ed to their established structure and theservice could be offered to studentsthroughout the country. Telenor R&Dsupported in the late 1980s trials involv-ing computer conferences at two of ourlargest distance education institutions(NKS and NKI). We also tested the useof computer conferences as part of a jointNordic education programme for re-searchers, in co-operation with the Uni-versity of Oslo.

There is now a large number of educa-tional institutions in Norway which areusing computer mediated communicationas an integral part of their courses. Somehave courses that are based completelyon electronic distribution and communi-cation, and the use of Internet is rapidlygrowing.6 The introduction of electronicnetworks has also led to the collaborationbetween institutions and we see newways of organising studies that offergreater openness and flexibility to thestudents.7

As the penetration of computers intohomes gets higher, and the demand formore flexible ways of learning grows inconjunction with a greater need for up-dating the competence level of the workforce, the use of electronic based coursesis expected to “explode”.

7 The Internet

As mentioned above, the use of the Inter-net for distribution and communication indistance education is growing rapidly.This is a challenge to course designerswho need to learn new methods for pre-senting a course in a new context. Someenvironments have started working onguidelines for designing WWW-basedcourses.8

The fascinating thing about the Internetis that it combines all the previously sep-

arate modes of communication; text,sound, images, graphics and live video(although still of some limited quality). Itgives you direct access to all the digitallystored information in the world. You maycommunicate as easy with persons inyour own neighbourhood as with personsliving at the other side of the earth. Youmay also make your own presentationsthat can be read and responded to by any-body anywhere. This is obviously a chal-lenging technology to all kinds of educa-tional institutions.

At Telenor R&D we have so far littleexperience ourselves with the Internet indistance education. We have, however,

since the summer of 1996, been engagedin the utilisation of the Internet inschools. In conjunction with that, wehave developed learning material forteachers and IT-co-ordinators at schools.This has been offered as free videos to allschools, and as courses given locally intwenty regions of the country. The course“Internet in education” is about to betransformed into a WWW-based course,which is economically supported by theNorwegian Ministry of Education.

Our engagement in schools is part ofTelenor’s initiative to provide teachersand pupils with limitless communication,a project called “The electronic school

6 One example is NKI, see the article“From electronic mail to Internet –where does it take us?” in this issue ofTelektronikk.

7 See the article “Telecom and OpenLearning – a challenge to institutionalcooperation” in this issue of Telektron-ikk.

8 See the article “Design guidelines forWWW-based course environments” inthis issue of Telektronikk.

path” containing the delivery of commu-nication lines and equipment, a Web-sitefor schools9, the above mentioned educa-tional offering and research.

In the field of research we have signed atreaty of agreement with the Ministry ofEducation on collaboration from 1996 to1999. The treaty is meant to support theMinistry’s plan for implementing infor-mation technology into Norwegianschools, the aim being to make all teach-ers and pupils into personal users of IT.

We have, as part of our research in thisfield, started a joint project with BritishTelecom Labs, connecting two schools inNorway with two schools in England.The intention is to let pupils from bothcountries communicate and collaborateby using videotelephones and the Inter-net.

This project is an example of two import-ant trends in today’s education:

• The internationalisation of teachingand learning10

• The merging of distance educationwith classroom teaching.

8 The future

As new technology is brought into theclassrooms from primary schoolsupwards, there will no longer be a cleardistinction between ordinary and distanceeducation. Collaborative work may takeplace between pupils and teachers acrossnational borders. Experts from all overthe world may be brought into the class-room for consultation, and guest lecturesmay be given from wherever the lectureris situated. Thus, we will see a mergingof distance and classroom teaching.

We will also see a great variety of tech-nologies being used. Telecommunica-tions will become more and more domi-nant, although they will still be used inconjunction with local gatherings andphysically distributed media (books, let-ters, cassettes, CD-ROMs). The mergingof various forms of communication as wehave seen in the Internet, will also appear

in other media. We already have mobiletelephones that at the same time are tinypersonal computers. WEB-TV is about tobe introduced, giving you access to theInternet from your ordinary TV set.

Within a few years we will also have thepossibility of meeting in virtual space,trespassing the screen and entering theartificially created world itself. There, wewill interact with the representations ofother humans. At Telenor R&D we aredeveloping VR solutions that may proveto be of great interest to distance educa-tion in the future.11

Whether we like it or not, the technologi-cal development is a major driving forcein the development of education andmust be taken seriously. Technologyitself cannot, however, solve the need forlifelong learning in tomorrow’s society.Much work has to be done at the educa-tional institutions in terms of pedagogicaland organisational development. Onlywhen technology, pedagogics and organi-sation are developed in conjunction canour educational institutions manage tomeet the challenges of tomorrow’sdemand for flexible and open learning.


9 The address to our Web-site is:http://www.skoleveien.telenor.no.

10Further reflections on this are given inthe article “The globalisation of educa-tion” in this issue of Telektronikk.

11An example of this is given in the art-icle “Learning in collaborative virtualenvironments – impressions from atrial using the Dovre framework” inthis issue of Telektronikk.

One of the Norwegian painter EdvardMunch’s pictures, ‘The Plough Team’,shows a man who is out ploughing. Hehas two sturdy horses drawing hisplough, one is black, the other white.They are harnessed together, but in thepicture it appears as though each is pull-ing in slightly different directions.

The development of technology is indeeda part of the development of a society,but even so we can sometimes feel thattechnology is an uncontrollable memberof our ‘ploughing team’. It is therefore arelevant question to ask: is it society’sreal need which is decisive in the type ofdistance learning that is on offer? Or isdistance learning to some extent a tech-nology driven invention in search of rele-vant application?

This question of the relationship betweentechnological and general social develop-ment in connection with distance learn-ing is indeed tightly knitted. There arefew, if any, social researchers who havebeen interested in this subject, so thereare few research or literary sources avail-able to lean on. I shall try to open up thetopic in a few different ways – by look-ing at historical trends, some typicalexamples and a few personal reflections.

The first technological

stage – education by post

Bernhard R. Tynes started a furniturefactory in 1927 in Sykkylven in Sunn-møre, when he was 20 years old. In1930 he expanded and built a newfactory. When he was only 13 he hadbegun to make wooden buttons, and at16, ski poles. He was one of the found-ers of the ‘industrial fairytale’ in Sun-nmøre.

In 1934 Tynes took a correspondencecourse in technical drawing, freehanddesign and business accounting. Hisfirst technical education was from acarpentry college in Molde, but hegradually became aware of a need forfurther education in design as part ofhis job as manager of an expandingfurniture factory ... Tynes himselfdesigned the furniture which was to beput into production, and the factory inSykkylven became a model company.[2]

More than sixty years ago we can alreadysee that correspondence courses wereused to increase competitivity in smallNorwegian companies. Indeed, corre-

spondence courses have been in use insuch connection for over eighty yearsnow, in Norway. Distance education, inother words, is not a recent phenomenonbased on new technology. It has been inwidespread use over large parts of theglobe for the last century or more. Evenso, from the very beginning, distanceeducation has been affected by generaltechnological developments. Pitman, theoldest correspondence college in Europe,was started in England as soon as thenational postal system had been estab-lished. Indeed, the Post Office communi-cation system was a prerequisite for thegrowth of correspondence colleges. Thissystem still works extremely well, at anyrate in Norway. It is cheap, effective andreasonably fast, and covers as near100 % of the population as possible.

When Ernst G. Mortensen ventured tostart NKS (the Norwegian Correspond-ence College) in 1914, the basis for hissuccess was not new technology. Hehimself had no more education than a 6months business course after a basic edu-cation, but he was desperate to start hisown business. He took the initiative in apreviously neglected area in Norwegianeducation, namely business education.The two first courses – ‘Double entrybookkeeping’ and ‘Business correspond-ence’ – filled a need for practical skills ina rich variety of establishments in manysmall local communities. He was luckywith his timing, for work was plentifulduring the First World War. Mortensen

wrote in his advertisements and schooltimetable, “We live in restless times.Never before has work pressure been sointense as now, and never before hasdemand for skills been so high”, andindeed he was correct. The need forworking skills and education wasacknowledged to an increasing degree,and the public at large was thirsty forknowledge.

The Norwegian school system was in noposition to meet this need. Certainly,there were public schools and businessschools in the larger cities, but very fewpeople had either the resources or possi-bility of travelling from home to attendthese courses. Loss of earnings, dearmaintenance and school fees made it dif-ficult for most. At the same time knowl-edge was an important prerequisite forhigher paid work and social advance-ment. The two course offerings fromNKS cost the dizzy sum of 40 kroner, ata time when a normal skilled workman’swage was about 0.50 kroner per hour.Even if that was expensive, school supp-ort costs were much higher.

At that time business education was thepoor relation in the Norwegian schoolsystem. The industrialisation and devel-opment of business at the turn of the cen-tury created a large demand for training.Business education gradually came toconsist of a range of short courses, fromthree months and upwards. At first it wasmainly private, without state or local

9

Distance education – a response to social needs or technology in search of application?B Y E R L I N G L J O S Å


authority support. In 1907, for the firsttime, there was a demand for some levelof education or experience to gain a busi-ness certificate, and gradually there alsodeveloped a public demand for specificcourse contents. The Central Bureau forStatistics wrote, after reviewing thestatistics for 1909:

“The available information shows thatthe overwhelming majority of busi-nessmen have no specialist education,and that on the whole they are recruit-ed from the lower ranks of society.”

Ernst G. Mortensen was successful withbusiness correspondence courses, andfrom 1917 the range of courses offeredwas greatly increased, including techni-cal and agricultural courses. In 1920 thecollege offered different courses andcombinations for businessmen, shopassistants, bookkeepers, secretaries,council treasurers, bricklayers, carpent-ers, electricians, machinists and civil ser-vants in the customs, postal and telegraphservices. In addition, a course was deve-loped to help in preparation for theLower Secondary School examination,for those interested in raising their gen-eral standard of education. These courseswere very popular with teachers – so-called seminarists – who wished to gainqualifications to teach at a higher level inthe school system.

Obviously, the correspondence coursesErnst G. Mortensen established in Nor-way were not technology inspired. Theyhad their origin in a social need for edu-cation and knowledge development,closely connected to practical work andindustry. The geographical aspect, whichis often stressed, was only one aspect ofthis social adaptation.

Correspondence courses built on a wellunderstood technology, but even so, rep-resented a pedagogic development thatwas truly radical. They were received onthe one hand with praise, but on the otherin the educational establishment, withscepticism, even though the ‘seminarists’seized the opportunity for further educa-tion. It took eight years from the founda-tion of NKS before this style of educa-tion was even mentioned in the currenteducation press.

Technology and distance

education

The postage stamp and a coherent pricemechanism for post was one of the hist-orical prerequisites for the growth ofmodern distance education. Anothertechnological connection was the print-ing and book trades. Rightly so, therewere a number of correspondence col-leges which grew from normal schools,and individual creative operations withstencil machines and similar apparatus.But very often we find a connection witha publishing house. In Germany, GustavLangenscheidt (later a well-known pub-lishing name!) worked on a self-studycourse with a Frenchman, Toussaint, asearly as 1856, and a book dealer calledHachfeld produced a course in technicalskills from 1895 [7]. One of the largestand best known American correspond-ence colleges, International Correspond-ence Schools, was founded around 1890by a newspaper editor in Scranton, Penn-sylvania, as a result of a series of articleson worker protection in the mine indus-try.

Newspapers have had only sporadicimpact on distance education, but on theother hand, connection to a publishinghouse has been common and lasting.Development and production of corre-spondence training material has a techno-logical relation to both book and maga-zine production. Ernst G. Mortensen is agood example of this. He was engaged inbook publishing and printing, and estab-lished himself as one of the major maga-zine publishers in Norway in the 1930s.

New media and technologies graduallycreated new opportunities for offeringeducation. The gramophone record madeinroads in language courses at Langen-scheidt at least as early as 1906 [7], andobviously was a great stride forward inimproving the practising of correct pro-nunciation. Later, the record was natural-ly replaced by tape and then by cassettes.These are media which integrate easilywith correspondence courses.

Broadcasting on the other hand is a con-siderably different medium. The firstattempt at schools broadcasting by radiocame in the 1920s. The radio lecture wasa popular programme format, and wasearly used to supplement correspondencecourses, e.g. in Germany, Canada andAustralia. Broadcasting has often beenconcession controlled and consequently

reserved for larger, national players. Thisis particularly so with television, whichbegan in the mid-1940s. Here again edu-cation was an important element. Forexample, Chicago TV College in around1960, was an important forerunner of theOpen University. Right up to presenttimes, broadcasting has a particular posi-tion in distance education, either asschools broadcasting, or in combinationwith correspondence courses and studycircles [4]. In Norway, we experiencedthe first co-operative project in languageeducation between NRK radio and a cor-respondence college in 1949, and since1960, adult education via radio and tele-vision has been a continuous offering.NFU, the State Institution of DistanceEducation, was founded especially toorganise this type of co-operative app-roach, in 1979.

The telephone had been in existence longbefore it was used for distance education.The first attempts were made as early as1939, but its use became more wide-spread from around 1960, especially inNorth America [9]. The most commonuse today is in telephone conferencing.Even though the telephone has a place indistance education, it has never reallytaken off in the sense of having a centralposition. Even though it can obviouslyfunction well in telephone conferencing,it has its limitation as an audio basedmedium.

Video was a new wonder-child, muchdiscussed in the early 1970s. Specialcompanies were set up to develop andmarket educational videos. Today, wefind little trace of them. However, videocassettes have acquired a foothold ineducation. We find both programmes ofbroadcast quality for a large market, andsimple productions with recordings oflectures or a short introduction as a basisfor group work. As a rule video is inte-grated with a more comprehensive edu-cational project. The length of time it hastaken for video to take hold has obvious-ly some connection with price levels andhow widespread video players havebecome.

In the mid-1970s experiments began withuse of satellite-carried distance educa-tion in Canada and the USA. In somecases it was combined with a reversechannel over the telephone network, anda new form of distance education hadbeen born. In contrast to the normal tele-vision it was possible to ring in duringthe broadcast, ask questions, or join in


discussions. The media was more inter-active, and this interactive use of broad-casting has become quite popular, espe-cially in North America [6]. In othercountries, including developing coun-tries, satellite is most often used only tocarry television programmes, withoutinteraction. In Europe, trans-nationaleducational broadcasting via satellite wasinitiated around 1990, with programmeinput from educational institutions andfree transmission time. Although fundingvanished, transmissions continued at alower level, and a number of nationalsystems also exist. We also have somecommercial, quasi-educational channels.In Norway, NORNET represents anattempt to introduce the more interactivestyle of distance education.

In parallel with satellite use, the tele-vision companies began to send picturesover the telephone network. It is possibleto send pictures both ways, and multiplesites can be connected in video confe-rences. For a long time this was fairlyexpensive, but with steadily improvingcompression technology this develop-ment has led to the video telephone, atthe same time as the ISDN network hasprovided a higher transmission rate. Thismeans that cost is no longer the chiefproblem with this technology, but evenso, there are limits when it comes toorganising many participants and manysites simultaneously [10], [11].

Computers represent a technology that,so to say, has revolutionised the whole ofsociety since the Second World War.Obviously, it has also been adopted forthe administration of distance education.It has taken a longer time before IT madeinroads for educational purposes. Wemay distinguish between two mainforms:

• CBL – Computer Based Learning

• CMC – Computer Mediated Commu-nication.

Computer based learning has never had alarge application in distance education,for many good reasons. These includecost, lack of standardisation, little flexi-bility, difficult access to equipment, etc.In the 1970s, NKS developed a pro-gramme for correction and commentingof student replies, which can perhaps beconsidered in this category. It has notbeen maintained.

Use of computer mediated communica-tion appears to have a larger application.

The first conference systems began to beused 10–15 years ago. American institu-tions again were the first, but there hasalso been Norwegian activity and con-siderable development [8], [12]. NKIintroduced its own system in 1987, andNKS launched its ‘Electronic College’two years later. The Ministry of Educa-tion had its own ambitious project in thisfield, and several higher education insti-tutions became involved. The currentNITOL project, with the colleges atTrondheim, Stord and Agder as centralpartners, is in some respects an outcomeof the initiative of the Ministry.

Over the last few years it has becomeclear that the Internet stands out as a sortof common highway for most of whathappens in computer mediated communi-cation. The World Wide Web is just afew years old, but today it is the standarddomain in the Internet world. In this way,de facto standards make it easier todevelop distance education as well.Already today, hundreds of distancecourses have found their way onto theinternational market through Internet.Also Norwegian institutions, amongthem NKS, prepare new and exciting ini-tiatives in this area.

What we are now seeing is that many ofthe technologies I have discussed hereare in the process of blending together.Databases, telephony, broadcasting,printing technology, etc. – they will allcome to be based more and more on thesame digital technology. Multimediareplaces the individual media we knowtoday. That will have considerable andimportant consequences for distance edu-cation.

Which training is suitable

for social development?

What do we really mean when we saythat education and training should besuited to the development of a society?Can we say that distance education as aform fills this need, and in particular, thatdistance education in Norway is suited tothe social developments we see? It is eas-ier to ask than to answer, but even so, Iwill try.

Just before Christmas 1995, the EU com-mission issued a so-called White book onEducation [5]. That document pointed tomany important challenges and aimswhich are relevant to education in a mod-ern society. Generally, we can say that

education and training should serve threemain aims: personal development, socialintegration and enablement for work andsocial life. How can we relate this to thecurrent social developments?

The White book points to the multiplicityof factors that affect education’s place insocial developments, and there are threechief factors that drive more in-depthchanges than most. These are:

• Future growth of the informationsociety

• Internationalisation of the economy

• The effects from scientific and techno-logical inventions.

The social changes connected with IT areoften compared with the effects of theIndustrial Revolution. I am certain thiscan be disputed, but I am also certain thatthe effects are considerable. First andforemost, IT has changed work processesand the way we organise production ofgoods and services. Mass production isgiving way to smaller, tailored runs.Work is being organised more flexibly,with teamwork and companies in a net-work. Job contents are more varied anddemanding, at the same time as peoplebecome more dependent on each other ina complex system of functions. Muchroutine work is disappearing, and in itsplace more skill intensive work andworkplaces are growing up.

If we see this in relation to education, itmeans obviously that the content of edu-cation should be re-focused, so that itcovers the new skills needs. In addition, atechnological convergence is takingplace between education and the work-place. That will be meaningless if educa-tion continues to use a classroom andblackboard with a teacher at a lectern andstudents with desks in a row – if educa-tion should prepare us for life in a mod-ern society. Computer tools and commu-nications media ought to be natural aids,and training work ought to adopt projectwork, independent responsibility and co-operation on how to utilise the informa-tion resources available in our informa-tion society.

The second important factor the Whitebook points to, is the internationalisationof the economy. We saw a clear illustra-tion of that during the recent workshopstrike in Norway. When the media talkedabout the problems of a long strike, itwas not the national consequences that


were emphasised, but those for the carindustry in Germany and Sweden, andthe erosion of confidence in Norwegianindustry internationally. We are all awareof the increasing difficulties we have indriving a national economic policy in amarket which is more and more open tointernational competition. In addition, weshould remember the cultural effects ofthe globalisation of society.

Again this development has considerableconsequences for the education sector. Itis not only the contents that are import-ant, even more important is the con-siderable emphasis now being made – inall countries – that the education sectormust contribute to enhance a nation’s orregion’s competitive strength. Mosteconomists believe that one of the mostimportant competitive factors is a popu-lation’s skill level. It is therefore believedthat effort in education can also con-tribute to reducing employment prob-lems, which has become a bugbear forour society. It is not enough in this con-nection to strengthen basic education inthe normal education system, access totraining and skills development must beavailable to all age groups.

The third factor mentioned, is currentscientific and technological inventions.Here we meet a paradox, fifty to a hund-red years ago most people regardedscientific progress as positive and prom-ising. Today, we gain new knowledge atan increasing tempo, but at the same timeit is increasingly regarded as a threat.Look at the environmental problems andthe discussions on genetics. Much pointsto science and technology being split offfrom cultural development and basic eth-ical thinking. If we are to get everythingin the right proportions before the nextmillennium, it is clear that the educationsector has an important role to play.Above all, it cannot just offer more andmore knowledge. It must contribute morethan previously to creation of under-standing for relationships, perspectiveand duties, across the traditional disci-plines and cultures.

From these three factors we can formu-late some general requirements for aneducational and training system which issuitable for social development:

• Adopt available technology and avail-able information resources

• Use workstyles which promote com-munication and co-operation

• Contribute to the fulfilment of skillneeds in working life

• Provide access for all to a system oflifelong learning

• Link knowledge in working life andsocial conditions

• Give cross-disciplinary and cross-cul-tural understanding

• Strengthen a whole, social and ethicalperspective.

Does distance education

satisfy these demands?

When asking if distance education satis-fies these needs, I have to reply, in partsyes, in parts no, and partly that ‘itdepends ...’. If we start with the last twodemands, there is no conclusive answer.I am afraid that distance education hasnot taken up the challenge, any morethan ordinary education, of providing acomplete understanding and perspective.What we positively can say, is that dis-tance education is often used by peoplewho wish to gain a greater breadth intheir education. An engineer will studymanagement, a secretary marketing, asocial worker reads a little law, etc. Inthis way, we contribute to a strengthen-ing of cross-disciplinary breadth ofknowledge among the participants. Aneffective system for mature and furthereducation must not only provide updatingof skills from basic education. It isequally important to develop breadth andcombine skills from various skills areas.

The great advantage that distance educa-tion has, lies in the middle of the needslist. Distance education was createdexactly for the purpose of opening accessto knowledge and education for newgroups. It is considerably ahead in beingan established system for lifelong train-ing and updating and renewal of workrelated competence. As a system it isflexible, efficient and reasonably low indemands on resources. An example:Since 1989 between five and six thous-and nursery assistants have completed abasic training for their work throughNKS, without a single classroom beingbuilt, or a new full-time teaching positionbeing created. That would obviously nothave been possible with a traditionaltraining organisation.

The two needs at the top of the list bringus back again to our main problem: dis-tance education, technology and society. Iwould like to look at each of them in turn.

Adopt available technology andavailable information resources

In my opinion, education that is to be instep with social development, must useavailable technology and informationresources. This does not mean, at anyrate in a negative sense, that it should becontrolled by technological development.This negative feeling in the expression isconnected to two dangers. The first isthat people can be tempted to adopt tech-nology too early, i.e. that technology canbe available but not to the actual users ofan educational offering. This leads to thepossibility that choice of technology doesnot help to provide good education, butcan be a new barrier that prevents usersfrom participating in education. You caneasily confirm that by looking at the hist-orical progression. Many technologieshave been adopted, but it has often takena long time from the first adoption beforethey reached widespread use and wereuseful in connection with education.

NKS has over the last ten to fifteen yearshad a systematic, running evaluation andtesting of new technological possibilities.In the course of this process we have,amongst other things, made a strategicchoice – to invest in visual media, pri-marily television and video. There aremany reasons for that choice, but one ofthe most important is that these twomedia are available to all Norwegians –and not only available – people alsoknow for the most part how to work theequipment.

The second strategic choice we havemade, is to invest in data communication.This is much more problematic asregards availability. Most people haveindeed met and used a computer at work,and many also at home. But it is still aminority that has free access to commu-nication. The most recent figures I knowfor connections to the Internet, are 21 %,of which 7 % are home users [1]. Evenso, this is several hundred thousand peo-ple, and the numbers are increasingrapidly. We have chosen data communi-cation as an investment area, because weare convinced that shortly, this form ofcommunication will be normal in mostworkplaces, and it is gradually appearingto be easy and user friendly.

The second danger which causes technol-ogy to be viewed with suspicion, is thatwe use a technology which is inappropri-ate to our goal, or we use it in an unsatis-factory manner. There is nothing unusual


in this, and it is not always the fault ofthe technology. There are times when thetechnology can be the prime mover, andsometimes the driving force, because it iseasier to get money for a technically so-phisticated experiment than for one thatis less exciting, so to say. Perhaps it istechnologists and people who are espe-cially fascinated by technology who bothdevelop and evaluate projects. It is un-fortunately harder to pull in sceptics andthe less capable, and projects can there-fore easily be biased towards technology.

But even the non-technical and peda-gogues can make wrong evaluations ofwhat technology is best for. When it isnew, indeed there is no-one who canaccurately judge, and so there must be aproportion of trial and error before wereach firm conclusions. There is reasonfor stronger criticism when we some-times see that we have not learned fromour errors, but continue on the wrongtracks. This can be blamed on being in arut, the financial position or organisation-al inertia.

One of the best known experts on tech-nology and distance education, TonyBates at the University of BritishColumbia in Vancouver, writes that thereare two main patterns in today’s distanceeducation. Representatives of one streammaintain that the ideal method of educa-tion is group teaching in a classroom orlecture theatre, and that the nearer dis-tance education can come to that idealthe better, and so technology should beused that best imitates that ideal. Theother tradition is based on developmentof a teaching tool, preferably in the shapeof printed texts and/or television pro-grammes, which can be supplementedwith different forms of communicationbetween pupil and teacher. New technol-ogy can often be used as a supplement tothis basic model [3].

It is not difficult to recognise these twotraditions in Norwegian distance educa-tion. Tony Bates says – and I believe heis right – that both these methods ofusing technology are inappropriate andmis-directed. Technological developmenthas reached a point where we shouldbegin to develop new educational mod-els. The problem is not that we are con-trolled by technology, but that we let ourpast experience come in the way of gooduse of the technology that exists.

Adopt work styles thatbuild up to communicationand co-operation

The second demand in connection withdevelopment of the information society,is for workstyles that reflect the way inwhich we have begun to organise our-selves at the workplace. Organisation ofprojects, building of networks, co-opera-tion and communication are some of thebuzzwords. I believe that it is right to saythat distance education has not come sofar, it is still largely characterised by one-way communication and communicationbetween teacher and the individual stu-dent. The picture has been influencedsomewhat by extended use of combinededucation styles, where groupwork andclass teaching are regular elements. Evenso, I do not feel we have come farenough.

What is of interest, is that changes in theorganisation of work processes are close-ly connected with technological frame-works and tools of different kinds. Thisalso means that the compulsion to be inthe same location falls to pieces. Theextent of home working increases, andcompanies and projects are held togetherindependently of geography. This oughtto mean that the same technology andtools that enable a workstyle, should beable to be used in a similar way in dis-tance education. In this way, we willhave a style of distance education thatmirrors real life, and contributes to build-ing up co-operative skills. In this sense,technology can be a bridge-builder be-tween the training system and the work-place.

This thought pattern has been the basisfor NKS investing in data communica-tion since 1989, especially in the use ofcomputer conferencing systems. Even so,we have only been partially successful.This autumn, therefore, we are tryingsomething new, in a different style.Firstly, we begin with the Internet, usingstandard tools. Secondly, we develop anew series, partly in a new style, ofcourse offerings, which are closely con-nected to central training needs in theworkplace, both public and private. Weoffer subjects such as practical projectmanagement, systematic improvementmanagement and strategic competenceplanning. Thirdly, we try to build in morereal-life examples, exercises and pro-jects, that require participants to co-oper-ate on tasks as a part of learning. As faras we are concerned that will be a new

style of distance education, and we willcertainly gain much experience in theprocess that will help us to improve.

Conclusion

Finally, I want to remind you of the pic-ture of the ‘Plough Team’ of social andtechnological development. Distanceeducation should be a plough that opensup for new life and encourages growth inthe furrows of development. It is ob-viously pulled by strong forces. There-fore, it also requires strong hands to steerthe plough if it is not to deviate.

References

1 39 prosent av norske husstander harPC. Oslo, Aftenposten, 29.10.96.

2 Amdam, R P, Bjarnar, O. NKS : enbedrift i norsk skole. Oslo, NKS-For-laget, 1989.

3 Bates, A W. Technology, open learn-ing and distance education. London,Routledge, 1995.

4 Cain, J. Mass media : educationaltelevision and radio. In: The systemof distance education. Papers to the10th ICCE International Conference.Ljoså, E (ed.). Malmö, ICCE/Her-mods, 1975.

5 Commission of the European Com-munities. White paper on educationand training : teaching and learning: towards the learning society. Brus-sels, 1995.

6 Daniel, J S. Satellites in distance edu-cation : Canadian experiments on theHermes satellite. In: Correspondenceeducation : dynamic and diversified,vol. 1 : The Advance Papers, 11thWorld Conference. Wentworth, R B(ed.). London, ICCE/RRC, 1978.

7 DIFF. Zur Geschichte des Fernstudi-ums. Eine Ausstellung des DeutschenInstituts für Fernstudien an der Uni-versität Tübingen, Tübingen 15. Junibis 11. Juli 1992. Tübingen, DIFF,1992.

8 Feenberg, A. CMC in executive edu-cation : the WBSI experience. In:Proceedings. Nordisk konferanse omfjernundervisning, opplæring ogdataformidlet kommunikasjon. Fjuk,


A Jensen, A E (eds.). Oslo, NKS/USIT, 1991.

9 Flinck, R. The telephone as aninstructional aid in distance educa-tion. A survey of the literature. Lund,Department of Education, Universityof Lund, 1975. (Pedagogical reports,1975, 1.)

10 Kristiansen, T (ed.). A window to thefuture : the videophone experience inNorway. Kjeller, Norwegian Tele-com Research Department, 1991.

11 Kristiansen, T. Five years of re-search into the use of telecommuni-cations in distance education.Kjeller, Norwegian TelecomResearch Department, 1993. (TFreport R 29/93.)

12 Rekkedal, T, Søby, M. Fjernun-dervisningsmodeller, datakonferanserog nordiske utviklingstrekk. In: Pro-ceedings. Nordisk konferanse omfjernundervisning, opplæring ogdataformidlet kommunikasjon. Fjuk,A, Jensen, A E (eds.). Oslo, NKS/USIT, 1991.


Developing telecommunications for usein distance education is a process ofadjusting technical devices to diversesocial and pedagogical settings. In thisprocess it is important that scientistsand users communicate to make surethat the services are developed to meetthe user needs. The field trial is amethod that allows for direct interac-tion between co-operating users, re-searchers and industrial partners. Anexample of a development processbased on this co-operating triangle isthe development of a Norwegian ISDNvideotelephone and its adjustment to aspecific distance education situation.

Technology must be

studied in its use

The technological development is nodoubt a major driving force in the devel-opment of distance education, and mustbe taken seriously. Knowledge about thevarious new technologies that are intro-duced is necessary in order to furtherdevelop the potentials of distance educa-tion.

Focusing too narrowly on technology,however, may be disastrous. We knowthat technology is but one of many issuesthat must be taken into considerationwhen studying distance education. Tech-nology can only be understood whenstudied in its use. To some people newtechnology can be exciting, to others itcan be scaring and alienating. It canbring people together, but it can alsoleave people in isolation. It can be a help-ful tool and it can be a strait-jacket. Inother words, there is very little we cansay about any technology in distanceeducation without considering the socialand pedagogical setting it is going to beused in.

There are several aspects which have tobe analysed in every particular distanceeducation situation if the use of telecom-munications is to be successful.

Attention must be paid to:

• The man-machine interface

• Pedagogical and communicativeaspects

• The physical surroundings

• The organisational structure

• The social infrastructure.

The users and their needs

The most important question to answerwhen considering what technology tochoose and how to use it, is “who are theusers and what are their needs?”. Theproblem is that people are often unawareof what their needs or preferences are.

An example to illustrate this, is a projectthat was run in the northern part of Nor-way some years ago, where a furthereducation course was offered by the localupper secondary school. The course wasbased upon a correspondence course, theintention being to give support throughlectures given by a local teacher. Thismodel of combined methods is muchused in Norway. Typically, the localteaching will be given once a week oronce a fortnight. In this case, the studentslived scattered over such a large area thatit was not possible to gather them on aregular basis to ordinary classroomteaching. As an alternative the teacherdecided to broadcast his lectures throughthe local radio once a week.

The question arose; what about the socialaspects of learning, and what about thetwo-way communication between stu-dents and teacher? As an attempt to meetthese needs the project leader thought it agood idea to arrange telephone confer-ences in conjunction with the radio lec-tures. That would give the students theopportunity to ask questions and com-ment on today’s lecture and at the sametime give them the feeling of being part

of a larger group of students. Now, thiswas not needs uttered by the students. Asa matter of fact, many of them were notparticularly enthusiastic about the idea ofmeeting each other in a telephone confer-ence. Many of them had no experiencewith telephone conferences and othersthought it to be just a meaningless add-on. Nevertheless, they agreed to try it,and what happened was that they foundthese telephone conferences to be so use-ful that they ended up expecting it to be anatural part of later courses they register-ed for.1

What this example illustrates is that peo-ple often have to experience the technol-ogy to become aware of their own needs,or rather, what needs the technology canfulfil. For instance, none of us had anyneed for a fax machine twenty years ago,but now that it is there, and we havebecome used to it, we can hardly managewithout it. Thus, it is true to a certainextent, that new technology creates newneeds. At the same time it is importantthat new technical devices and telecom-munication services are developed tomeet these and already existing needs. Toachieve this, it is necessary that scientistsand users communicate. Telenor R&Dhas seen field trials as a convenientmethod in this aspect.

15

Field trials – the need for user participation in telecommunications developmentB Y T O V E K R I S T I A N S E N


1 The project is presented in the article“Telecommunications in distance edu-cation” in this issue of Telektronikk.

FIELD TRIAL

Technology Needs

Solutions

Researcher User

Figure 1 Field trials

Field trials

Field trials as a scientific methodacknowledges both the researcher and theusers as participants in the developmentprocess. The desirable communicationbetween researcher and user is achieved.It is a method which allows the research-er an opportunity to study the introduc-tion of modern technology in a socialsetting very closely, whereas the usersget direct influence upon the shaping oftechnological and pedagogical solutions.

Telenor R&D has over the past ten yearsbeen running a number of field trials inthe area of distance education. Throughthe field trials we have learnt from theusers what they have experienced ashelpful in the learning process, whatwere the barriers, what needed improve-ment and what should be dropped.Through this dialogue with the users wehave been able to refine and furtherdevelop our services, that is, both theequipment and the rules for use. Whereappropriate we have involved our indus-trial partners in the development process.Much of our development work is basedon a co-operating triangle, consisting ofour research institute, the industry andthe users (ref. Figure 2).

The initiative to new or further develop-ment work may come from either of thethree partners. The users may express aneed for some specific communication

tools in a particular distance educationsituation. Researchers at our institutemay predict the need for some telecom-munication tool and invite users and in-dustry to explore its potential, or theindustry may come up with some newpiece of equipment they want to test tosee if there is a market for it.

In order to highlight the importance ofuser participation, the story of the devel-opment of the ISDN videotelephone willbe examined in some detail, including thestory of how it was put into use and ad-justed to a specific distance education sit-uation in one of our field trials.

The development of ISDN

videotelephone

The development of the Norwegianvideotelephone was initiated at TelenorR&D [1]. Some of our scientists startedstudies in the field of image compressiontechniques already in the late 1970s. Themain objective was to investigate thepossibilities for transmitting live colourimages with a bit rate of a few hundredkilobits per second. This would highlyreduce the requirement for transmissioncapacity in the telecommunication net-work. At that time, there was alreadyequipment available on the market fortransmitting digital video on 2 Megabitper second channels. Compared to thetransmission of ordinary analogue videoas we know it from television, much wasalready achieved. Whereas the transmis-sion of television signals requires acapacity equivalent to 1000 analoguetelephone channels, the transmission ofdigital video on 2 Megabit per secondchannels requires a capacity equivalent to30 analogue telephone channels. By 1984our scientists were convinced that itwould be possible to transmit live colourimages at 64 kilobit per second. Thatwould make it possible to transmit videoon only one digital channel. At that time,very few people really believed that itwould be possible to achieve the necess-ary compression ratio with acceptableimage quality.

Nevertheless, in 1986 we signed the firstdevelopment contract with the Norwe-gian company Tandberg A/S. During thefollowing two years, a Norwegian low bitrate codec was developed, which made itpossible to transmit live colour imagesover a single 64 kbit/s digital connection.It was first presented at the Telecom ’87in Geneva, Switzerland, and both the

cost-performance ratio and the physicalsize of the equipment were major break-throughs at that time.

As soon as the codec was ready for pre-sentation we started our first field trials.This was not an ISDN codec. It had to beused on dedicated 64 kbit/s digital con-nections. It must be admitted that therewere certain limitations to the quality ofsound and image presentations of thisfirst Norwegian-made video codec. Nev-ertheless, the quality proved to be suffi-ciently good for the videotelephone toappear as a useful tool for various cate-gories of users.

Experiments were carried out in the areasof

• Distance education• Remote supervision• Sign language• Remote surveillance.

These experiments gave us some veryvaluable experiences that were chan-nelled back to the scientist developers atour research institute and at Tandberg.Encouraged by its technical success andseveral successful field trials, togetherwith the emergence of ISDN, our re-search institute decided at the end of1988 to continue work in the field ofaudio-visual telecommunication, concen-trating especially on the videotelephoneservice. Work that had recently beenstarted in international standardisationbodies also played an important role inthis decision. The aim of the ISDNvideotelephone project was to establish apublic videotelephone service on ISDNon a trial basis from the beginning of1993, which was achieved.

In the development of this terminal thescientists paid much attention to theexperiences that had been made in thefield trials. There was a clear need foroperating procedures and controls to beas simple and easy to use as possible.From the beginning it was decided toincorporate the human factor aspect aspart of the project to ensure a high degreeof usability. Much effort was put intodesigning a terminal that would be justone piece of equipment with a minimumneed for connecting cables and wires,that would be user-friendly in terms ofinstallation and use. It should be as easyto use as a normal telephone. The termi-nal has now been industrialised and isbeing sold by Tandberg Telecom.


Telenor R&D

ndustry Users

Figure 2 Co-operating triangle

The early field trials proved that in areaslike for example distance education,there is a need for connecting externaldevices to the terminal. When commu-nicating with a group of people you needto connect an extra camera, a larger mon-itor, and perhaps also extra microphonesand loudspeakers. These options areimplemented in the terminal. Thevideotelephone is primarily designed forperson-to-person communication. Whenused in distance education, adjustmentshave to be made, both technical andorganisational ones. One example of thatis the field trial that was carried out inone of the largest timber districts in Nor-way, the Trysil project. This project is agood example of how our co-operatingtriangle with users and industry works insuch a development process.

The Trysil project

In Trysil the local secondary schoolaimed to become a competence centre fordifferent target groups in the district. Avideotelephone was installed at theschool, thus enabling the school to offerdistance education courses to adults whowould otherwise have had difficulty inattending any kind of further education.Instead of having to travel to a remoteschool, these adults could attend coursesat the local school in the afternoons.There qualified teachers from other partsof the country would appear on thescreen and communicate with them. Inthis project, the videotelephone equip-ment was used for three different courses[2], [3].

Communicating at a distance, throughmedia, represents a new experience tomost teachers. In addition to masteringthe technology, there are several factorsin such a situation that may place a strainon the teacher. It is important that theteachers be allowed to concentrate onteaching and establishing a dialogue withthe students. Effort was therefore madein this project that operation of the tech-nology be as simple as possible. We hadin earlier field trials seen this need forintegrated, user-friendly solutions.

To avoid a chaos of cables and wires anda multitude of buttons to operate, all theequipment was put into a cabinet, whichwas designed at our request by a smallNorwegian firm called NovaKom. Thiswas an integrated solution, where all thecables were hidden and all the equipmentwas easily operated with the aid of a few

buttons. Feedback given by the teachersat an early stage of the project, led to fur-ther refinement of the design and func-tions of the cabinet until it had becomethe user-friendly solution we had intend-ed it to be (ref. Figure 3).

In all three courses, the distant teacherwas communicating with a group oflearners, the size of the group varyingfrom five to twelve persons. This made itnecessary to enable the students to seethe teacher and the presented materialclearly by installing a large screen. A 46-inch screen was erected, featuring theability to display a small picture withinthe larger picture. This gave the studentsthe opportunity to see their teacher in asmall video window while the rest of thescreen displayed a written presentationmade on the computer. It may be arguedthat seeing the teacher’s face is of minorimportance in the learning process. Thepsychological value of this “face in thewindow” should not be underestimated,however. Experiences show that it givesa feeling of “human touch” which provesto be motivating to the students.

A large screen is in itself no guaranteefor clear presentations. Earlier experi-ences had taught us that much attentionmust be paid to how the material is pre-sented. Due to the reduced quality of thecoded video signal, it is crucial that theteachers do not put too much informationinto their presentations. It takes somepractising to prepare good presentations.The teachers in this project were advised

to spend some time practising on theirown before starting their teaching, thusensuring that the presentations would bereadable to the students.

To establish a dialogue between teacherand students, earlier field trials hadproved the audio connection to be moreimportant than the video connection.Having to use a microphone to addressthe teacher turned out to be felt as abarrier by some students. In other fieldtrials we had seen the interaction behampered because too many studentswere sharing one microphone or becausethe students would have to move to reacha microphone. Based on these exper-iences, special effort was made in theTrysil project to give the students easyaccess to the microphones by installingone microphone for every two persons.By placing the microphones immediatelyin front of the learners, the physicalbarrier of addressing the teacher wasreduced to a minimum (ref. Figure 4).

An attempt was also made to minimisethe psychological barrier by allowing thestudents to meet their teacher in personbefore the course started. This has provento be of vital importance for achieving agood dialogue in all kinds of distanceeducation, and is highly recommendedirrespective of the type of technologybeing used.

At the teacher’s end an image of thewhole group of students was displayedon a small monitor. Due to the quality of


Figure 3

this first generation codec, it was imposs-ible for the teachers to recognise individ-ual faces on the screen. This was feltunsatisfactory to the teachers. They soonexpressed a need for recognising theindividual students with whom they werecommunicating. At the same time theyfound it important to be able to see thewhole group in order to have a feeling ofreaching them all and to know that theywere all actually present. To fulfil thisdual wish, extra cameras were installedand connected to the microphones so thatevery time a student pressed the micro-phone button, the camera focusing onhim/her would be activated. As soon asthe button was released, the camerashowing the whole group would be acti-vated. This solution proved to be muchmore satisfactory to the teachers. Theyfound it easier to communicate with thestudents than before. The equipment wasimplemented by the same company thathad designed the cabinet, NovaKom, andis another example of how the users’needs were identified by our researchers,who in co-operation with an industrialpartner found a technical solution to meetthose needs.

Developing a “multimedia

cabinet”

While running the Trysil project, an en-gineering high school that planned to

Figure 4


offer a distance education course toTrysil, announced a need for simultane-ous computer communication alongsidethe video and audio communication. Thisgave the impetus to a further develop-ment of the technical set up used inTrysil. Again, NovaKom was engaged byour research institute to come up with asolution that would satisfy this need.

NovaKom had already designed a cabinetfor use in telemedicine, where this ideaof integration was realised. Both sent andreceived video images were displayed atthe screen of a computer, thus enablingthe users to operate from one screeninstead of three as it was before.

This solution fulfilled the wish for simul-taneous audio, video and computer com-munication and was adapted to our newversion of the cabinet tailored for dis-tance education. Our aim was once againto make operation of the equipment assimple as possible. A short menu wasdesigned with only a few functions tochoose from. The applications wereplaced under Windows and easily operat-ed by a mouse. To simplify operation ofthe equipment to the students, the controlof local and remote cameras was giventhe teacher.

The engineering high school expressed aneed for the teachers of computer pro-gramming to communicate on an individ-

ual basis with every student. To makethis possible, the teacher was given acontrol panel of the class on his/herscreen. The students were equipped with“press-the-button-to-speak” micro-phones. When a student pressed thebutton, a cross would appear on theteacher’s control panel at the computerscreen. By clicking at that cross, theteacher would open up that student’smicrophone. At the same time, this stu-dent’s mouse and keyboard would beactivated, to allow the teacher and stu-dent to communicate person-to-person.Headsets were used in order to avoid dis-turbing the other students.

This cabinet, the “Teleproff” (Figure 5),has been used for distance educationgiven by the engineering high school atGjøvik to engineers and building con-structors in Trysil [4]. Again we can seehow participation from the users hasbeen instrumental in the development oftechnical devices especially designed fordistance education.

The need for co-operation

The field trial in Trysil was based on asituation where a teacher at one locationmet with his “class” at a distance. The“class” was rather small, and the dia-logue played a central part in the teach-ing/learning process. The equipmentdeveloped was especially designed forthis and resembling situations.

Other situations may call for differenttechnical and organisational solutions. Insome cases, the transmission of lecturesmay be offered as a supplement to self-study courses where a large number ofindividual students spread all over thecountry are mainly working on their own.In other cases, the transmission of lec-tures held by highly qualified specialistsmay be the main issue. The lecture maybe given to students at the same locationas the lecturer at the same time as it istransmitted to other locations, or it maybe transmitted to distant students only.Interaction may be part of the session orit may not.

Various situations call for various solu-tions, both technically, communicatively,pedagogically and organisationally. It hasbeen argued in this article that it may bedifficult to predict what will be a goodsolution. To find that out, different mod-els must be tried out in real social set-tings. Users must be invited to make their

experiences and come forward with theirjudgements. New technology is not syn-onymous with new technical devices.New technology is defined by how thetechnical devices be used. Thus, thedevelopment of new technology cannotbe considered finished before the rulesfor how it is to be used are developed.This is a process that takes time, and it isimportant that the users are involved init.

To sum up, the development of technic-ally functional solutions is not sufficientto make good models for distance educa-tion. In addition, effort must be put intodeveloping the pedagogical, communi-cative and organisational aspects of dis-tance education. To succeed in this, tech-nicians, researchers, pedagogues andorganisers of distance education mustcooperate. Only in this way can wedevelop telecommunication solutions anddistance education models to meet theneeds of the users.

References

1 Kristiansen, T (ed.). A window to thefuture. The videotelephone experi-ences in Norway. Kjeller, NorwegianTelecom Research Department,1991. ISBN 82-423-0158-1.

2 Bårdseng, S E. En pedagogisk vur-dering av Trysil-prosjektet. Forsøkmed bildetelefon i fjernundervisning.Elverum, Elverum lærerhøgskole,1991.

3 Bårdseng, S E. En pedagogisk vur-dering av Trysil-prosjektet. Delrap-port 2. Elverum, Elverum lær-erhøgskole, 1992.

4 Hetland, P. Det gjenskapte klasserom: utformingen av teknologi for fjer-nundervisning. Evaluering av fjern-undervisning via Teleprof C200.Elverum, Høgskolen i Hedmark,1995. (Rapport 8.1995.)


Figure 5

Pho

to: R

une

Lisl

erud

/SA

MF

OT

O

20

Cultural and educational barriers to tele-education– Experiences from the use of videoconferencing

B Y T O V E K R I S T I A N S E N


In this article I will give some exam-ples on how cultural heritage and edu-cational attitudes can become barriersto exploratory use of new technologyin distance education and thereby alsopossible barriers to efficient learning.My examples are drawn from experi-ences made in experiments with video-conferencing.

Videoconferencing

In Norway we have during the last tenyears run a number of experiments withvideoconferencing in distance education,at first using 2 Mbit/s connections andlater 64 – 128 kbit/s (videotelephony).The experiments have been run at vari-ous educational levels from uppersecondary schools to colleges and univer-sities. Telenor R&D has been engaged inmany of these experiments.

There are two aspects to videoconferenc-ing which I would like to point out:

1 The resemblance between videocon-ferencing technology and televisiontechnology.

2 The similarity between educational sit-uations established through this tech-nology and ordinary classroom teach-ing.

Teachers and students alike tend to getassociations partly to television andpartly to the classroom situation whenfirst confronted with this technology.These associations make videoconferenc-ing a technology easy to adapt to. At thesame time, they tend to set the standardfor how this technology is used, and to acertain extent that proves to be a barrierto an exploratory and potentially excit-ingly new way of utilising this technol-ogy in distance education.

Resemblance to television

The associations to television tend tomake teachers act as though they weretelevision newsreaders, whereas the stu-dents lean themselves comfortably to theback, expecting to be entertained. Inother words, both teachers and studentsget the idea of a medium where inform-ation is given one way from a sender to apassive receiver. And indeed, this is tele-vision technology, using cameras, moni-tors, microphones and loudspeakers forthe distribution of a message. But video-conferencing is by no means television.

Videoconferencing is two-waycommunication

Contrary to television, videoconferencingallows the receivers to respond to themessage transferred. Videoconferencingis not a one-way medium, but a two-waymedium. Only when this is understood,can its potential be fully explored in dis-tance education.

Whether a teacher has the conception ofvideoconferencing as a medium for one-way or two-way communication makesthe whole difference when it comes totaking on a teacher-oriented or a student-oriented approach. In order to achieve astudent-oriented educational situation itis necessary to establish a dialogue be-tween teacher and students at the verybeginning. When communication is firstestablished, both parts tend to forget theirassociations to television. The dialogue isnot established by itself, though. It mustbe initiated by the teacher and needscareful planning to be successful.

As the students’ association to televisionmay become a psychological barrier tolearning, it must be attacked by theteacher at once, inviting them to a dia-logue, letting them immediately experi-ence that this is actually very differentfrom television and that they are expect-ed to be active in class.

Similarity to classroom

education

The other aspect referred to, the similar-ity to classroom education, is positive inthe sense that both students and teacherseasily adapt to the situation. Neverthe-less, this similarity too, can in somerespects prove to become a barrier to anefficient use of the medium and thelearning potential. I will give someexamples.

In most cases, videoconferencing hasbeen used for ordinary lecturing, in otherwords, for one-way communication. Thestudents are invited to pose questionsmuch the same way as in a lecture at theuniversity, i.e. at the end of a 45 minuteslong one-way delivery of information.Thus, the lecturer’s association to ordi-nary lecturing, sets the standard for howto teach at a distance.

This is NOT utilising the medium’spotential for two-way communication.Used in this manner, it could be arguedthat the lecture might just as well havebeen video-taped and sent the remotestudents by post. This would allow themto look at the tape at their own pace, stopand rewind, have discussions whereappropriate, and pose questions and com-ments to the lecturer afterwards, on thetelephone, by fax or electronic mail.

Using videoconferencing for one-waylecturing can of course be justified. The

Figure 1 Similarity to watching television (The Trysil Project)


point I want to make here is that our cul-tural and educational heritage tends tomake us copy our traditional way ofteaching when starting distance educa-tion, thus preventing us from exploringits potentials and thereby representing abarrier to development.

Potential for activating thestudents

Distance education through videoconfer-encing could be run quite differently. In-stead of the teacher lecturing all the time,the remote students could make presenta-tions from their site. They ordinarilyhave the same facilities for presentingvisuals as the teacher and can easily drawand write or show preproduced figuresand illustrations. Few teachers seem tohave the fantasy to let the students dothis, or rather, it does not fit into theirteacher-oriented pedagogical view, sothey do not even consider it a possibility.

This medium does allow a more student-oriented focus, though. Not only can thestudents make presentations for theteacher, but also for fellow studentsplaced at various sites around the coun-try. Some adult students are content withgetting one-way lecturing on the screen,but some are by no means satisfied withthat. I have met many adult distant stu-dents who were frustrated because thelectures had been given in a traditionalmanner, with very little room for dia-logue. Many adult students also getannoyed when teachers fail to make ref-erences to their job experiences and helpthem relate new-gained theoreticalknowledge to their daily job situation.They want that to a much larger extentthan most teachers seem to be aware of.

Videoconferencing/videotelephony mayalso be used for direct interaction be-tween students. This has been done in arather untraditional experiment in Nor-way with great success. At two of ourteachers’ colleges located in the westernand northern parts of Norway the leadersof the drama sections decided to usevideoconferencing as a tool for co-opera-tion. As part of the training they let theirstudents at each college meet on screen,acting various roles in the same screen-play. This worked out very well. Teach-ers and students were all very excitedabout it, and found it stimulating to meetcolleagues and students at another col-lege in this way.

A tool for developing new

pedagogical methods

The resemblance to television technologyand the similarity to classroom teachingmakes videoconferencing easy to adaptto for both teachers and students. Thesetwo aspects may, however, also becomebarriers to an exploratory use of thistechnology. It should be recognised thatteaching and learning through videocon-ferencing differs from any other kind ofteaching and learning, and has to beexplored on its own premises.

As argued in this article, videoconferenc-ing may be implemented in distance edu-cation, not just for copying traditionalteacher-oriented lecturing, but as a toolfor developing new pedagogical meth-ods. To achieve this, the medium’spotential for two-way communicationmust be utilised to a much larger extentthan what has been done so far, andteachers must take on a more student-oriented pedagogical approach.

22

Distance education in the electronic classroomB Y B J Ø R N H E S T N E S A N D J O H N W I L L Y B A K K E


The electronic classroom was develop-ed for distance education purposes.A research team called MultiTeam/Munin was established to find out howan ordinary classroom could be ex-

tended to reach students at other loca-tions. One of the main issues for theproject team was to build a classroomwhich was easy to use for lecturersexperienced with regular classroom

situations but not with distance educa-tion. They should be able to start usingthe electronic classroom without hav-ing to revise their teaching material orchange their teaching style. When lec-

Tromsø

Trondheim

Oslo

Bergen

Kjeller

Tavle 1..... ........................ .... ....... ........... ..

Overhead

Overhead

UNIK

OverheadUNIK

Overhead

Tavle 1..... ........................ .... ....... ........... ..

Overhead

Overhead

UNIK

OverheadUNIK

Overhead

TANDBERG Vision

OverheadUNIK

Overhead

Tavle 1..... ........................ .... ....... ........... ..

Overhead

Overhead

UNIK

Figure 1 The infrastructureused for the project


turers became familiar with the class-room technology, they were invited totry out new pedagogical principles andmore advanced presentation tech-niques step by step. To gain relevantevaluation results from regular use,the classrooms were placed at two dif-ferent locations at the University ofOslo. Description of the classroom con-cept and the evaluation results aredescribed in this paper.

1 Introduction

As the first organisation in Norway,Telenor R&D started in 1977 experi-ments with LM Ericsson’s picturephone[1]. Many research projects continuedwhich resulted in the first public trialvideo conference service in 1983 [2].Later on, this service grew to sixty videoconference rooms all over Norway. Dis-tance education was one of the main usesfor the service and started in 1987 be-tween Båtsfjord, Vardø and Tromsø [3].There were also connections to theworld-wide video conference service.The amount of use was not as high asanticipated, which led to a market ana-lysis carried out in Norway in 1991 [4].

The market analysis was the biggest everin Norway concerning video conferenc-ing and its conclusions were:

• Every year 10 million meetings areperformed in Norway where partici-pants travel to the meeting.

• In 2.2 millions of these meetings, theparticipants themselves were of theopinion that the meetings could havebeen substituted by a video conference.This potential correlates with otheranalysis outside Norway, [5] and [6].

• In 1.5 millions of the meetings, it wasalso a requirement to see “what is talk-ed about”, in addition to see “who istalking”. “What is talked about” iseither presented on a blackboard or anoverhead projector.

• A large part of these meetings dealtwith education in one way or another.

This new knowledge of the use led to thestart of the MultiTeam project in 1991.Later on, a joint project was initiatedwith participants from UNIK (Centre forTechnology at Kjeller, University ofOslo), USIT (Centre for InformationTechnology Services, University of Oslo)and T R&D (Telenor Research andDevelopment). The project was called

MultiTeam at UNIK and T R&D, andMunin at USIT. The joint project wasmanaged from T R&D. Some of the pro-ject members also participated in a Euro-pean project called ESPRIT MICE(European Strategic Programme forResearch on Information Technology,Multimedia Integrated Conferencing forEurope) [7].

The main infrastructure used in this pro-ject was the Supernet, which was a Nor-wegian packet switched 34 Mbit/s net-work connecting the four universities inNorway and T TR&D. In addition, themore narrowbanded IP-network was used(Figure 1).

2 The objective

The objective of the project was to pro-vide lecturing at one of the Universitysites in Norway available for students atother sites, both inside and outside Nor-way. Therefore, a concept called “Elec-tronic classroom” was developed. One ofthe main concerns of the project groupwas to make it as easy as possible for thelecturers to start using this new form ofeducation. They should be able to bringwith them their regular teaching materialinto the classroom without having tochange and adapt it. In addition, theyshould be able to use their usual teachingstyle, for example moving around andbeing “on the air” all the time. Thereshould be no need for helpers to operatethe classroom, so a high level of automa-tion was built in. Therefore, there wereno camera people there, no script girlsand no producer. The only people werethe lecturer and the students.

3 The development of the

classroom

The idea was to design a classroomwhich incorporated video, audio andwhiteboard information. Previous experi-ences from distance education and videoconferencing clearly indicated the im-portance of having a shared educationalmaterial. It was also considered import-ant to design a classroom not radicallydifferent from an ordinary classroom;one that could be used for regular class-room teaching, group work and distanceguidance.

The three main media used in the elec-tronic classroom were:

• Audio – used to hear who is talking

• Video – used to see who is talking

• Data – used to see what is talked about(on the electronic whiteboard).

Figure 2 shows a typical classroom situa-tion. Microphones are mounted in theceiling, evenly distributed in order tocapture the voices of all participants. Thelecturer used a cordless microphone so asto be able to move around freely. Onecamera had focus on the lecturer and wasplaced at the back of the room. This cam-era could handle the flickering of awhiteboard which for most of the timewould be behind the teacher. Two cam-eras were also placed at the front of theclassroom with focus on the students. Avideo-switch selected the camera corre-sponding to the microphone with theloudest input-signal. A trolley with twoTV-monitors was placed at the front ofthe classroom and one at the back of theroom. The two cameras at the front werefor the students, so they could be seen bythe remote students. The one at the backwas for the lecturer. When he was look-ing at the local participants, he wouldalso see the remote students on the moni-tor at the back. Each trolley had onemonitor for incoming video (covering theremote site) and one for outgoing video.A whiteboard was available for the lec-turer for writing and displaying his teach-ing material. To operate the whiteboard,a light-pen was used as a writing, point-ing and erasing device.

At the remote site, there was anotherclassroom with audio, video and white-board information. Thereby, the remotestudents could hear the lecturer and thestudents and see the lecturer’s move-ments and body language. Behind thelecturer they could see where he wasabout to operate on the whiteboard, butthe video channel had too low resolutionto provide full whiteboard information.To see the whiteboard information, theyhad to look at the electronic whiteboard.

4 The audio

Audio was the most important part of amultimedia conference. If the audio waspoor or lost, it could not be compensatedby good quality on both video and data.Without acceptable audio, the conferencemay as well be terminated. Our experi-ence shows clearly that the importance ofthe audio hardly can be overestimated.

Minimal requirements to the audio sys-tem was that audio signals from the lec-


turer should be heard by everyone, andthat all the other participants should alsobe heard when they took the floor – simi-lar to a regular classroom situation.

The microphones were mounted in amatrix formation in the ceiling of theclassroom in order to minimise noisefrom tables, chairs, etc.

The audio in the classrooms had a band-width of 7 kHz, which was sufficient forspeech.

Classroom, UNIK

Overhead

Overhead

UiO

Overhead

UNIK

Overhead

UiO

Overhead

UNIK

Overhead

Classroom, UiO

Overhead

UiO

Overhead

UNIK

Overhead

Figure 2 A typical electronic classroom situation


The ideal solution for audio connectionwas to have an open, two-way communi-cation channel. To overcome the prob-lems of reverberation and echo, a systemfor echo cancelling was implemented.Audio from one site was transmitted tothe loudspeaker(s) on the remote site,where it was picked up by the micro-phone(s) and transmitted back again. Bysubtracting the incoming signal from theoutgoing signal, this undesired echo wasattenuated. A problem, however, was thatthe signal from the loudspeakers waschanged by the acoustic properties of theroom. Available technology claimed tobe able to “subtract” up to 40 dB (deci-bel) of the echo. This implied that thecancelling capacity was not sufficient inall situations. To eliminate the echo com-pletely, the loudspeaker volume wasreduced when 40 dB was not enough.

5 The video

The video system provided moving pic-ture from the remote site. The studentscould see the lecturer, where he waswriting on the whiteboard, as well as hisgestures and body language. One cameraplaced on top of the trolley at the back ofthe classroom covered the lecturer, andtwo cameras placed at the front of theroom covered the students.

The video system turned out to be quitecomplicated in the wish to meet therequirements of a high degree of automa-tion as there was no “production team”available for the lectures. Since therewere several cameras, a switching systemactivated the most appropriate camera.This was accomplished by a video switchwhich monitored the microphone levelsand selected the camera corresponding tothe microphone with the loudest audiosignal. The camera in use had autofocus,was tuned for the best zooming angle andhad auto-iris. The camera at the back ofthe classroom, which covered the lectur-ing stage, had a specific tripod. A motorcontrolled the pan and tilt function. Thecordless microphone transmitted, in addi-tion to the audio signal, also control sig-nals to the camera tripod. This was thereason the camera followed the speakerso he was in the picture all the time.

6 The data (whiteboard

information)

At each site, there was at least one white-board displaying the written content of

the lecture. The lecturer might use theelectronic whiteboard for writing and dis-playing his presentation material. Fromthe literature we know these applicationsas “shared workspace”, “global win-dow”, “electronic whiteboard” or “sharedwindow”. In this project, the whiteboardmedium was considered the second mostimportant medium (the most importantwas the audio).

The lecturer could load his transparenciesinto the system before the lecture, andselect transparencies by pointing andclicking on a scrollbar. When a transpar-ency was displayed, the lecturer couldwrite and draw comments on it, using thelight-pen. There was also an eraser func-tion that allowed him to erase parts of thecontent.

Figure 3 A lecturing situation

Figure 4 The principle of the whiteboard


The whiteboard was in principle an ordi-nary computer screen. Since CRT-tech-nology did not permit monitors largerthan 21” without a big increase in cost, avideo cannon was used, allowing a 100”whiteboard (2 metres wide and 1.5metres high).

The working area of the whiteboardcould contain the same amount of infor-mation as an A4-document or transpar-ency (950 x 950 pixels). (An overloadedtext document or a drawing with manydetails was usually not a good solution,seen from a pedagogical point of view.)With the high resolution whiteboard, itwas possible for the lecturer to use exist-ing educational material prepared for theregular courses. This maid it easier forthe lecturer to commence using thewhiteboard.

For writing on the whiteboard, a light-pen was introduced. At one site, the posi-tion of the pen was determined by itsiteration with the scanning mechanism ofthe video cannon. At the other site, thewhiteboard was in principle a large pen-pad or digitising board.

The light-pen had three buttons, like athree button mouse. The write-functionwas activated when the pen was pushedtowards the surface of the screen, where-

upon pixels were echoed to the white-board(s). Pressing one of the buttons onthe side of the pen would produce a cur-sor on all the whiteboards. Each site hadtheir own colour on the cursor so the par-ticipants could see which site was point-ing at a given time. Pressing the otherbutton activated an eraser function. Thebasic functions of the pen were thereforesimilar to the standard equipment of ablackboard: Chalk, pointer, and sponge.

As Figure 3 shows, the whiteboard wasbased on a back projected video cannon.The input to the cannon was a computersignal meant for a data screen. To reducethe depth requirements in the roombehind the whiteboard, a mirror wasused. The screen of the whiteboard wassemi-transparent, for projection andwriting.

7 Experiences from the

users

In any project, the verdict of the end-users is the ultimate measure of success.Therefore, great effort was taken in orderto evaluate the experiences of lecturerand students. Being used for a series oflectures at the University, the appropri-ateness of the classroom for educationalpurposes was evaluated.

7.1 Experiences from thestudents

Two group interviews were undertakenwith the students. In addition, theyanswered a questionnaire after eachclass. Members of the project team alsoattended most of the lectures, therebybeing able to observe the reactions of thestudents.

The attendance of the students was moti-vated by the content of the course only,not by the fact that the course was givenin the electronic classroom (an exercisein new technology might be a motivationfor students in informatics). The studentsalso had to pass an examination, anotherfact making them attentive critics of theclassroom.

The overall evaluation from the studentswas positive, referring to the classroomas an environment for learning. Therewere also several critical responses, serv-ing as inspiration for further developmentof both the pedagogical approach and theclassroom. One reaction was that the lec-turer should have had more experiencewith the classroom before the coursestarted, reflecting the fact that the class-room was completed just before the startof the course.

Regarding the technology in the class-room, the students seemed to be impress-ed by the quality of the whiteboard,whereas the quality of the sound was thesource of several critical remarks, con-firming the experiences of the projectteam.

There were some mixed remarks regard-ing the video channel. The students at theremote site appreciated the video signalof the teacher, whereas members of bothgroups of students seemed to be distract-ed by the self-view function, makingthem uncomfortable with taking thefloor. Sometimes the self-view monitorswere even turned off.

One important issue in distance educa-tion and conferencing was the feeling ofpresence of the other party or parties. Atfirst, the students at the remote site foundthe lecturer distant, not just in the literalsense. A great improvement for theremote students was accomplished whenhe visited their site, thereby becoming areal person, not just a small figure on thescreen. This was in accordance withcommon knowledge from video confer-encing, that meeting in person was valu-Figure 5 Lecturer with the light-pen at the electronic whiteboard


able for the quality of later interaction viaelectronic media. Contrary to our initialhypothesis, the local students were themore critical of the classroom. Halfwayin the course, all the students at theremote site voted for the experiment tocontinue, whereas a large minority of thelocal students voted for a termination ofthe project.

Explanations provided by the studentswere partly that they felt pity for theremote students, partly that the operationof the electronic classroom introduced anelement of distraction in the lecture.

A further substantiation of their argumentmay be developed, considering the al-ternatives of the two groups of students.The alternative for the students at theremote site was either to commute to theother branch of the university, or not toattend the course at all, whereas the al-ternative for the local students was aseries of lectures without experiments,interruptions or distractions.

7.2 Experience from thelecturer

The overall evaluation from the lecturerwas also positive, and his several criticalremarks were very constructive, and animportant input to the development pro-cess. One obvious reaction was to thefact that the classroom was completedjust a few days before the course started,making it impossible to get familiar withthe classroom.

The lecturer found the whiteboard easyto operate, and the functions availablefairly well implemented. There were sev-eral additional features he would haveimplemented in the whiteboard, like thepossibility to zoom in on certain areas ofthe whiteboard, or to display animation.

It was a disadvantage, according to thelecturer, that the system appeared some-what rigid, because it worked best whenthe set of transparencies was loadedbefore the lecture. He found that thisproperty of the whiteboard made it diffi-cult to improvise during the lectures.

He also had some critical remarks to theaudio system and the light-pen, whileemphasising the overall positive impres-sion of the classroom.

The lecturer complained about a lack ofcontact with the students in the remote

classroom, and he found it difficult tofeel their presence. He also had to invitethem directly to raise questions and par-ticipate in the discussions.

8 Experiences from the

operation of the class-

rooms

The development of the classrooms wasan on-going process, also being basedupon experiences gathered during theproject. In the development process, sev-eral lecturers from the University of Oslogave valuable advice.

8.1 The electronic whiteboard

There were several problems with theoperation of the pen, partly due to driverand hardware problems, partly becausethe pen-technology at one of the sitesrequired that the pen was held orthogon-ally to the whiteboard, causing someproblems for writing and drawing on thewhiteboard. Another problem with thislight-pen was that it was activated about2 feet from the whiteboard, often produc-ing some unintended results. The conclu-sion from the project team and the lec-turers was that this pen was suitable as apointing and clicking device (similar to a

mouse), but not as a device for freehandwriting and drawing. The pen-technologydeployed at the other site proved to be abetter solution for writing and drawing.

8.1.1 Changes and improvements

The first version of the whiteboard hadthe menu system on the right hand sideof the whiteboard. After a few lessons,the lecturer found that it would be moreconvenient to have the functions on theleft hand side, where he was usuallystanding during the lectures. A new lay-out was implemented with the buttonsplaced on the left hand side, making thelecturer feel more comfortable.

8.1.2 Scanner

To provide better flexibility and allow alarger room for improvisation, the lec-turer wanted some kind of overheadfunction so that he could show new trans-parencies not previously stored in thesystem. One alternative was to use a fastscanner, but after searching the marketno satisfactory product was found.

Figure 6 Seen from the lecturer. Remote students on the upper monitor, self-view on the lower


8.1.3 Document camera

A document camera presenting imageson the whiteboard was another alterna-tive proposed by the lecturer. As experi-enced from the video conference servicein Norway these cameras did not have asufficiently high resolution for the class-room. The document camera was install-ed, however, as a backup facility.

8.2 Audio

The quality of the audio represented oneof the major problems during the firstmonths of the project. The packet switch-ed network could rearrange packetsreceived in the wrong sequence, but therewere no mechanisms to compensate forpacket losses or damaged packages. Thislatter aspect made some seminars on thenarrowbanded parts of Internet very hardto follow.

The quality of the audio was continuous-ly improved during the project period byincreasing available bandwidth in theaudio channel. Further steps have to betaken in order to obtain an even betterquality.

8.2.1 Backup facility

In addition to the audio channel, a tele-phone conferencing unit was installed inthe classrooms as a backup system. Thenit would still be possible to carry on withthe education in case the audio channelshould fail.

8.3 Video

The whiteboard being back-lit was thesource of some problems with the light-ing. With no extra light, the lecturerappeared as a silhouette for the studentsat the remote site, whereas lamps direct-ed towards the whiteboard resulted inreduced readability at the local site.Lighting from various angles was triedout, the solution being that the lecturerwas illuminated from a lamp mountedfrom the ceiling, about 1 metre in front ofthe whiteboard.

Automatic camera control in the form ofan inexpensive “follow me” camera wasalso tested, with little success. A moreexpensive equipment was tried out a littlelater with good results.

The camera on the lecturer provided anoverview of the whiteboard and the lec-turer, making it somewhat difficult forthe remote students to observe the ges-tures of the lecturer, let alone catch hiseye. This in turn caused the lecturer toseem somewhat distant. In addition to thehardware codec, there was a softwarecodec as a backup system.

Each lecture was videotaped, whichopened the possibility for a later reviewof the session for both students and lec-turer.

8.4 Operation of the classroom

It was considered very important tomaintain a high stability during the lec-tures since this was a regular universitycourse. This in turn called for a supportservice for the lecturer. Fortunately, oneof the students attending the course wasalso a participant in the MultiTeam pro-ject. He assisted the lecturer in setting upthe equipment, and took care of problemsoccurring during the sessions. Similarly,at the other site, there was a memberfrom the MultiTeam/Munin project pre-sent, to ensure the proper functioning ofthe classroom.

The Supernet performed well during thecourse. There have been some minorproblems with routers at two occasions.

9 Implications for further

studies

The development process and the opera-tional phase provided valuable insightsregarding user aspects and technologicalaspects of the electronic classroom, whilealso pointing out several areas for furtherresearch and development.

One set of questions refer to the way theusers experienced their participation in adistributed classroom. The results fromthis project shows clearly that perceiveddistance is a very important topic.

The students at the two sites perceivedthe series of lectures in quite differentways. (It is here interesting to note thatthe students at the remote site were morepositive towards the project, even thoughthey were complaining about how remotethe lecturer was.) The lecturer also madecomments about the distance to theremote students, and that he had to takespecial measures in order to integrate

them into the discussions. There wereimportant challenges in overcoming thisfeeling of distance.

Two hardly surprising results from theproject was the importance of the audio,and that the whiteboard never could getlarge enough – both important topics forfurther development.

References

1 Jacobæus, C, Blomberg, H. LM Eric-sson 100 år, Band III : Teleteknisktskapande 1876 – 1976. Van NostrandReinhold, 1992.

2 Nordahl, C K. Fjernsynsmøter : en nyprøvetjeneste i Norge. Telektronikk,79, (3/4), 1983.

3 Kristiansen, T. Fjernundervisning ifiskerifag : forsøk med toveislyd/bilde kommunikasjon over 2Mbit/s samband. Kjeller, NorwegianTelecom Research, 1989. (TF report28/89.)

4 Aardal, A et al. Få fart på Norge medtelekonferanser! Kjeller, NorwegianTelecom Research, 1992. (TF reportR 26/92.)

5 Kollen, J H, Garwood, J. Travel/communication tradeoffs : the poten-tial for substitution among businesstravelers. Bell, 1975. (Bell CanadaReport.)

6 Egido, C. Teleconferencing as a tech-nology to support cooperative work :its possibilities and limitations. In:Intellectual Teamwork : Social andTechnological Foundations of Co-operative Work, p. 351–371. Gale-gher, J, Kraut, R E, Egido, C (eds.).Lawrence Erlbaum Associates, 1990.

7 Kirstein, P T, Handley, M J, Sasse,M A. Piloting of multimedia integrat-ed communications for Europeanresearchers (MICE). In: Proceedingsfrom INET, 1993.

8 Clayman, S, Hestnes, B, Kirstein, P.The interworking of Internet andISDN network for multimedia con-ferencing. Information services &use, 15, (2), 1995. ISSN 0167-5265.

1 Telelecturing in Norway

– the NORNET-project

In a collection of articles, “Open andDistanced Learning Today” (edited by F.Lockwood, 1995), D. Keegan presents anevaluation of an Irish satellite-based pro-ject from 1993–94. Keegan assumes thatthis is one of the first, if not the first, uni-versity-accredited virtual classroomcourse by satellite in Europe besides pro-jects like Europace and Eurostep whichhe characterises as mainly technologicalexperiments [1].

Without putting too much of an emphasison this, Keegan has in fact made a mis-take in his assumption about “who is firstin Europe”. Already in the autumn of1985, Telenor in Norway, started its firstexperiment with satellite broadcasting aspart of an educational course. A smallgroup of students were able to followpart of a French course in their ownhome, in front of their own television [2].Since 1990, we have in Norway beenable to achieve a relatively extensiveexperience with satellite-based distanceeducation, mainly through the so-calledNORNET-project. During the periodfrom January 1990 to January 1996 morethan 10,000 adult students have beengiven an educational opportunity.Through the utilization of 360 receivingstations (consisting of 85 study centres,and upper secondary schools that havebeen equipped to receive satellite broad-casts (transmissions)), 24 different edu-cational courses have been given, andone has cooperated with educationalestablishments at upper secondary level,university and college level and privatecentres of competence. The number ofparticipating county councils in NOR-NET has increased from 3 to 11, and thecouncils have also appointed coordinat-ors to work with this enterprise. The Uni-versity of Bergen’s Media Centre(UMS), which has had the main respons-ibility for producing and sending the tel-electures, has up until today produced400 hours of telelectures.

NORNET (The Norwegian Network forDistance Education) was established as aproject in 1989. The aim was threefold:

• Distribute further education to the pop-ulation or vocational groups that other-wise would have difficulty in takingpart in the normal courses given byeducational establishments

• Offer industry, commerce and manage-ment a distribution ‘channel’ to give

information and labour-market trainingto its users

• Develop an infrastructure to promoteknowledge.

Teaching by satellite is (and will be)mainly given as telelecturing, based ontraditional methods. Lectures are sentdirectly from the studio, and codedthrough Telenor’s satellite to each recipi-ent mainly at a local study centre. Theform of presentation used, especially thelecture part, is very similar to plenarylectures that are given at universities andcolleges. Most of the time is “one-way-lecturing” with a certain amount of timegiven to enable students to ask questionsand give comments.

The oral presentation is supported bywritten presentations in the form of pre-prepared texts, computer-designedposters, flip-over material and preparedpapers. Most telelectures sent from thestudios are supplemented by conversa-tions, guest lecturers and are sometimesillustrated by video sequences.

To give participants the opportunity oftalking to the lecturer, either by askingquestions or giving comments, time hasbeen reserved during each transmissionfor tele(phone)conferencing at previouslychosen study centres. The students areinformed about this and prepare their roleas “public askers of questions or com-ment makers” and in this way representthe rest of the participants. The othergroups have the possibility of sendingquestions and comments to the lecturervia fax and/or telephone.

The students are organised in subjectgroups and work together in a variety ofways between broadcasts. The subject isworked with in local groups, individualstudy and written correspondence. Thegroup work is partly followed up by thecentral course supplier / educationalinstitution.

NORNET is based on both direct trans-mission and full television quality.Through the use of satellite technology itis possible to reach all parts of the count-ry at the same time, and this gives thepossibility of having an unlimited num-ber of participants. NORNET’s transmis-sions are transferred through the broad-casting standard MAC. This means thatthe broadcasts are coded and can only bereceived by registered addresses with thecorrect receiving equipment.

The actual instruction takes place in theTV studio at the University of Bergen orat the University of Oslo and is transmit-ted through the radio link system toTelenor’s earth station at Nittedal, wherethe programme is coded and transmittedup to the satellite ‘Intelsat 5’. Everyreceiving station is equipped with a satel-lite dish, a decoder and a coding card intheir television.

It is necessary to build up a close cooper-ation with the individual local authori-ties, as each authority is responsible forthe local coordination of the enterprise.In each of the participants’ local commu-nity a coordinator or facilitator has beenappointed. This person is responsible forgiving information, registering courseparticipants, preparing the local studycentre so that it can receive the broadcastand is also the point of contact with theproject leader in Bergen. The coordinatorfunction plays a key role in the project.

2 The challenge of in-

creased knowledge –

the NORNET-evaluation

It is often said of distance education thatit is an education without any limits orboundaries. This statement can be inter-preted in many different ways. In the firstinstance it gives a picture of an educa-tional form which can give possibilitiesthat exceed geographical and timeboundaries. In its formulation it alsoimplies that education will become amore and more international phenomen-on. The traditional frontiers are no longera barrier for educational suppliers. Paral-lel with the development of education asa market product, Norway will also morefrequently experience offers of educa-tional supplies from different countries.

If we reverse the order of the statement,“education without any limits”, we canalso see that distance education is aboundary breaker, meaning that it chal-lenges our accustomed conception aboutteaching and learning. Through a flexibleeducational programme one can see out-lines of a qualitatively different form ofeducational organisation, where the tradi-tional mass production paradigm is re-placed by a more flexible and much moreindividualized approach. “There’s some-thing in the air”, something new and rad-ically different, which is about to pushitself forward into the traditional educa-tion system. It points towards solutions

29

Something in the air – Norwegian experiences with telelecturing in flexible education

B Y G U N N A R G R E P P E R U D


which are way ahead of the present-day’seducation system. It is with this perspect-ive in mind that one should analyse anddiscuss distance education.

Within the NORNET-project a numberof minor evaluations have been executedin connection with individual courseoffers [3]. These evaluations havebrought to light the fact that satellite-based distance education is in fact manythings. Even though telelecturing, dis-tributed via satellite, has been a commondenominator, the evaluations have shownthat there is a diversity within the basicorganisation, as well as in the preparationof the actual teaching.

The evaluation reports which are nowavailable (only in Norwegian), vary insize, quality and also have different lev-els of ambition. They differ from simplecourse evaluation to results given byappointed external evaluators. A com-mon feature is that they have been writ-ten within a somewhat short period oftime. Because of that one has neither hadthe time nor the possibility to go in depththeoretically or empirically. The method-ological approach has mainly been theuse of rather limited questionnaires. It isalso interesting to note that the Nor-wegian reports and evaluation haverarely built upon each others’ experienc-es, which makes it difficult to developspecific knowledge in this complicatedfield of study.

The most comprehensive evaluation ofNORNET was made in 1993. 580 stu-dents, out of a total of 881, were inter-viewed via telephone about their experi-ences with television-based education.For three of the four educational offersthat were involved, 90 % of the totalnumber of participating students wereinterviewed. The majority of local co-ordinators (45 of 60) and teachers (30 of45) were also interviewed, partly throughthe use of questionnaires, partly by tele-phone and partly in private meetings.Also representatives from the educationalinstitutions that had given the courses viaNORNET were interviewed.

The evaluation and points of view givenby the three involved groups have beensummarized, commented upon and anal-ysed in a preliminary report with theworking title, “Increased Knowledge isalso a Challenge” (from now on referredto as ‘the NORNET-study’) [4]. Experi-ences from NORNET are also comparedwith national and international evaluation

and research in the field of satellite-baseddistance education.

The main conclusions in the NORNETstudy coincides with the research whichhas been done on satellite-based distanceeducation in several countries so far:

• It is technically possible, also in Nor-way, with its fjords, mountains andscattered population, to give an educa-tional offer where lectures and teach-ing are distributed via satellite.

• Nothing indicates that the academicquality or/and the educational level arein any way damaged by this kind oflecturing.

• Local study centres are important inmany ways; as receiver stations, ascommon ground both socially and edu-cationally, as marketing operators andas a “listening post” to register anynecessary requirements.

• The satellite model contributes to “addi-tional knowledge for more people”.

The NORNET-study focuses on a num-ber of situations connected to organisa-tion and teaching. We will present theparticipants’ and teachers’ points of viewin some of these areas.

3 What do the students

experience?

The NORNET-study focuses on four dif-ferent courses, a smaller course about ECeconomy, a further education course atupper secondary level about Welfare Ser-vices for the Elderly, two qualifying edu-cational courses at university and collegelevel covering Public Administration (5credits) and Working Life Psychology(10 credits). The largest group of stu-dents (42.5 %) were between the ages of41 and 50. The average age of the stu-dents following the NORNET-courseswas higher than of those following norm-al distance education courses that areotherwise offered in Norway. In three outof four courses there was a majority ofwomen. In the course covering WelfareServices for the Elderly the dominance ofwomen students was a natural phe-nomenon, because the course wasdirected at the occupational group, whichis predominately made up of women. Thecourses were primarily taken as furthereducation, except for one offer, that ofPublic Administration, which was taken(especially by women) as a first univer-sity subject.

Two-thirds of the students maintainedthat satellite teaching was their onlychance of being able to study. Incorpo-rate in the answer and a decisive factor,quite naturally, was that they didn’t needto travel a long way, leave home or takeleave of absence from work, in order tostudy. In Norwegian and Scandinaviancontext, it is interesting to notice a devel-opment which points out that geograph-ical distance is only one of many reasonsfor developing studies outside education-al institutions. In the NORNET-study wefind that people living right next to aneducational institution, prefer to studythrough distance education. Withoutopening up old debates about the conceptof distance education, it should be point-ed out that it is about time we demandeda new term to cover what we mean bydistance education.

The NORNET-study also shows thatmany students are neither aware of whicheducational possibilities exist on themarket, nor have they done anything toprovide themselves with this knowledge.One of the courses offered by NORNETwas also, at the same time, available as acorrespondence course, produced byNKS Distance Education together withthe College of Lillehammer. The courseoffered was marketed over a longerperiod and in different ways throughoutNorway. Even so, few, if any, studentsattending the NORNET-project seemedto be aware of it. One of the main con-clusions, therefore, is that it is crucial inconnection with any future recruiting ofstudents to flexible educational courses,that the information and motivation aboutthe courses take place as closely to theusers as possible.

When students are asked to give a fullevaluation of the total course package(by giving an estimated value on a scalefrom 1–7), “the average grade” given is4.7, that is a little over average. The eval-uation varies from subject to subject,mainly due to the quality of the lectures,partly due to the way the course is struc-tured and what is stressed in the study.While students on one course werereasonably satisfied with the lectures,students on another course felt that theyhad too few. While students of one sub-ject felt that two-way communication bytelephone was to a certain extent a wasteof time, students studying another subjectfelt that this form of communication wasextremely valuable.


Students were also asked if they wouldfollow another course with a similarstructure. The answers were mainlyencouraging, 50.9 % said that they woulddefinitely think about repeating the expe-rience, 31.5 % said that they might, and aclear 13 % of the participants were quitedefinite that they would not be taking asimilar kind of course again. There are anumber of conditions connected towhether one would take such a courseagain or not. The NORNET-study showsthat the amount of work involved, theextent to which one has been a studentbefore and the distance of time sincebeing a student or pupil, play a decisiverole. It appears that women to a lesserdegree were interested in taking part infurther courses organised in this way.

Other Norwegian evaluations of satellite-based distance education is more positivewith regard to the evaluation of the con-tents and organisation. In Vaagland’sevaluation from 1990, 90 % of the stu-dents say that they would not mindattending similar courses again, and over96 % thought that form of distance educa-tion was practical [5]. Lorensen, and oth-ers, reported that all the students thatcompleted the questionnaire recommend-ed that one should continue teaching inthis way [6]. Even in the project that hasreceived the most negative remarks inconnection with content and form, it wasfound that the participants had not beenscared away by their more negative expe-riences. One could quite well continue towork in this way under specific circum-stances [7].

Not surprisingly, the use of satellites andtelelecturing received great attention bothamong students and in the communities.Telelectures sent directly via televisionwas something new, and involved thesetting up of a large technical and organ-izational apparatus. Quite a lot of invest-ments were made to supply the necessaryequipment and the media focused uponthis as it was new and exciting (but alsocaused problems).

In one of NORNET’s minor evaluations itis stated that the focus on satellite andtelevision created an unfortunate situationby taking the attention away from themain intention, which was to try out an“in-service-training model” for teachers inupper secondary schools. Instead of con-centrating on their own processes of learn-ing, the teachers were only concernedabout how the lectures were dispatched.

4 What then are students

concerned with in

connection with their

evaluation of tele-

lectures?

Both because satellite transmissionplayed a central role, but also becausetelelecturing was primarily the basic ele-ment in the educational programme, itwas of special interest for the NORNET-study that this aspect was evaluated. Bothstudents and teachers were thereforeasked to evaluate different aspects of tel-electuring.

Even though one finds some variationbetween the groups of students, subjectsand gender, the main conclusion was thatthe students were satisfied with this formof delivery. Compared with other Nor-wegian evaluations in this field theevaluation results given in the NORNET-study were very positive. The same canbe said if one compares the distance edu-cation students’ evaluation with evalu-ations given by students studying oncampus at universities and colleges.

The lectures were, as far as the studentswere concerned, well planned, especiallywith regard to the contents they present-ed. Another way of putting it is that parti-cipants experienced that the lecturersknew what they were talking about, andthe majority felt that the lecturers gavethem a better understanding of the con-tents both by stressing the central pointsin the study, by presenting new andexciting material and by helping studentssolve their given tasks.

There is more criticism with regard topresentation, and about a third of the stu-dents said that the lecturers could havebeen better at organising their materialmore pedagogically. It was pointed out,especially, that the lecturers kept up sucha tempo that it was difficult to keep upwith writing notes, and that it was diffi-cult to follow everything that was said.This was regarded as somewhat of adilemma by the lecturers themselves, andone never actually found out how oneshould compensate for not having stu-dents in the studio with them. Many dif-ferent tricks were tried out. One of theteachers, for example, said that he usedthe camera-man as if he was a student,and talked to him.

The students were relatively positive tothe lecturers’ use of supportive materialduring their lectures, whether it was com-puter designed posters, flip-overs or useof the document-camera. But a numberof students would have liked to havebeen sent the visual material beforehand.

The fact that telelecturing distributed viasatellite has functioned well is confirmedby a number of investigations in a numberof countries. Wong, for example, reachesa similar conclusion in her evaluation ofsatellite-based distance education at theUniversity of Saskatchewan [8]. In hisevaluation from 1995 (see above) D. Kee-gan asked students on two different occa-sions to evaluate whether these kinds ofcourses were able to sustain the academiclevel of excellency. Mid-way evaluationresults gave 86 % positive answers, andthe final results were the same. Only 1 %felt that it was a disadvantage not to haveface-to-face contact [9].

The positive reactions do not totally repre-sent a united front. Collins and Murphyshow results where participants are inagreement that the lecture form does notsuit satellite distribution. One becomesquite quickly passive and bored. Thisexplanation is primarily based on the factthat the teacher is in control and domi-nates the situation, as well as seeming towant to present too much material in tooshort a time [10]. A natural conclusion tothis is that bad teaching becomes twice asbad when presented on television!

With regard to the contact betweenteachers and students the latter are notquite as positive in their evaluation. Thisis primarily due to the technical problemsin connection with tele-conferences, andto problems concerned with audibility. Itwas the technical problems that causedthe students to give a rather negative“grading” here compared with the evalu-ation given to the lectures.

From a didactical point of view tele-con-ferencing was not particularly successfuleither. Three conditions are deciding fact-ors for this conclusion: The situation wasperhaps not put to good enough use, nei-ther students nor teachers had been giveninformation about various possibilities asto use, and one had too high ambitions forwhat actually was possible. Therefore,two-way communication was rather anartificial element, and represented moreof an illusion about conversation than aconversation in its own rights. The com-


munication became more like a conceptof narrative in the actual programme.

The didactical problem with two-waycommunication is also a visible conceptin other research and evaluation work.Wong, in a similar way as Collins andMurphy, points out that students wereobviously rather reluctant about speakingdirectly over the air. They were clearlyinfluenced by the context in which com-munication was taking place, and theresult of this was that the local assistantteachers had to take responsibility forasking questions on behalf of the stu-dents. In line with Vaagland’s observa-tions one found that students were con-cerned about what the other groups sawand did. Partly, one competed to becomea visible participant in the studio, andthose who didn’t manage to get to thefore had a tendency to lose interest [11].

In an evaluation of multi-point videoconferencing G. Holden verifies Wong,Collins and Murphy’s opinion. Eventhough a number of students say theywant to ask questions, they don’t do so.Partly because the teacher dominates,partly because they dare not. G. Holden’soverall evaluation is that it is the organ-isation of teaching and not the mediumthat is responsible for this passivity. Anumber of those who didn’t participateactively in the interaction sat on the doc-ument camera and then left, and veryoften the teacher did not even notice thatthey had left the studio [12].

5 What do the lecturers

experience?

After the first year of satellite-based dis-tance education Wong concludes in herstudy that much of the success is due tothe fact that one had managed to recruitexperienced and charismatic lecturers forthis role [13]. Even though one shouldnot perhaps be quite so generous withone’s praise when mentioning lecturersin the NORNET-project, the students didsay that they thought they should becomplimented for their work. In spite ofthe fact that teachers were just left to sinkor swim, the students have mainly evalu-ated their contribution as positive, eventhough there are some differences to befound from subject to subject.

Nearly all the teachers who participatedwere new beginners with regard to tel-electuring. They were mainly theacademic staff who in their normal role

use (more or less) traditional lecturingmethods. They were placed in quite a dif-ferent context, in a television studiowhere they were to be responsible to acamera, and see no students.

When they look back on their experiencethe teachers are nevertheless very positive.

They do not think that telelecturing is asgood as normal lectures on campus. Buthaving the opportunity to give lectures ina new medium was regarded, in itself, asa positive experience. 70 % of them saidthat it was positive, while only 1 (out of30) said that it had been a very negativeexperience.

When asked what they regarded as beingmost positive about telelecturing theanswers were mainly divided into twoparts. Firstly, it was pointed out that byusing this organisational form, one wasable to reach more people (at the sametime). Secondly, it was pointed out thatby working in this way one also learntand developed oneself (“It demandedmuch more thought about what oneshould present” / “a challenge” / “a newway of teaching” / “learn other forms”).

On the negative side the lack of directcontact and possibility of a dialogue withthe students were the things mainlyremarked upon. Some also mentioned thetechnical problems and the generaluncertainty of teaching via this medium,“I’ve been worried about the situation fora long time”, one of the teachers said.

With regard to this latter remark it isinteresting to note that three of the teach-ers actually said that they didn’t experi-ence any difference between satellite lec-turing and their normal lecturing situa-tion. This can indicate, without making apoint out of it, that some teachers evenon campus apparently have little or nocontact with their students. But theanswers can also be interpreted as mean-ing that they were not influenced by thesituation, even though the surroundingswere different.

The teachers are not so equally satisfiedwith satellite lecturing as the studentsseem to be. Even so, the majority don’tmind giving it another go. 26 said yes, 1said maybe, and 3 said no. Almost half ofthe teachers find that form of flexibleeducation very useful, while the otherhalf say that it is “usable”.

Why are teachers so positive? When theyare asked to give reasons, they give anumber of different answers:

• It is part of the future• The possibility of reaching many• Become well-known (a star) • Challenging/exciting• Technically informative• Different working methods, new challenges• Like lecturing• No reason for not trying• It must be possible to do better.

This evaluation seems to coincide withother evaluations given by those thatteach on flexible education studies.

The following is taken from Dahlén andLundgren’s summary of their evaluationcomments [14].

Teachers take it for granted that dis-tance education requires another kindof planning and gives less room forimprovisation than normal teachingsituations. On the whole, distance edu-cation demands that teachers give farmore of their time.

Teachers emphasise distance educa-tion and the distance students in amuch more positive way. They believethat contact between them and thesestudents functions well, that the dis-tance students are motivated, inde-pendent, full of initiative, have criticalideas and take more responsibility fortheir studies.

A more or less compulsory question toteachers involved in flexible education iswhether the contents attain just as high alevel of excellence as normal studies. Ofthe 10 teachers connected to the NOR-NET study, who gave an answer to this,eight of them say that compared with thenormal course the students’ results aremuch better or better. Only one says thatthey are somewhat poorer. Even thoughthe number of answers are few, they ver-ify the evaluation from other similarcourse offers. A summary of 12 qualify-ing distance educational courses in 1993shows that in 9 of the projects the resultswere evaluated as good or very good, 2were evaluated as average and 1 wasevaluated as weak. For 3 of the 12 courseoffers it was shown that the results werebetter than those of an ordinary study,and for 6 projects that they were equallyas good [15].

There is therefore no specific reason toworry that flexible education course


results are qualitatively weaker thanthose of on-campus studies. Moreover,the examination results show that one isfaced with a hard-working and motivatedgroup of students.

Experiences with distance education inNorwegian higher education has shownthat teachers themselves profit fromworking with distance education compar-ed with teaching on the campus-basedcourses. Did the teachers participating inthe NORNET-studies feel the same?

7 of the 30 teachers say that they haven’tlearnt anything new compared to theirnormal teaching. Otherwise, the follow-ing comments were made:

• The importance of good planning,structure and clarity

• No matter what the setting a need toget away from one speaker/monologue

• Prepare a more complete lecture thanis usual in an auditorium

• More aware of one’s performance

• Put more stress on details

• Self-assurance, more routine

• Significance of how one looks (!)

• Rely more on the manuscript than onthe performance part.

Which advice does one have to thosewho plan to give telelectures?

• Jump right out into it, it will probablybe all right

• Plan well – not too much material

• Try and get some variation

• Take it easy

• Find a relaxed approachable form. Usesimple language – not too much sub-ject jargon

• Try to imagine there is an audience –use body language

• Activate the students in their groups /use your imagination

• Divide the teaching into smaller por-tions.

6 The beginning of some-

thing new?

The NORNET-project cannot be regard-ed as standing for new approaches in thefield of didactics. One has, mainly, keptto the ‘known’ and tried to do this as well

as possible. In connection with telelectur-ing one has primarily contributed toimproving the traditional lecturing withina television setting.

Even though one would have liked to seegreater technical and didactical develop-ment than was shown in the project, onemust, as a main conclusion to the projectas far as it had come in 1994, say that itwas positive.

The results show that flexible educationbased on satellite-based presentationseems to have found a form which func-tions in relation to different user groups.This conclusion is also supported byother internal and external investigations,besides NORNET, in Norway andabroad.

Perhaps the most important conclusion isthat this study conforms with other re-search and evaluation work in this field,and establishes the fact that educationalresults and the quality of the studies arein no way behind those which one findsin the ordinary education system. Thismeans that if one continues to developflexible education courses one also hasthe opportunity in the future of establish-ing a qualitatively better education andsystem of learning. As a clear distinctionfrom the ordinary education system,which has about reached its maximumpotential, flexible education is just in itsbeginner stages. Perhaps we have reach-ed the time when one should begin withmore radical changes?

References

1 Keegan, D. Teaching and learning bysatellite in a European virtual class-room. In: Open and distance learningtoday. Lockwood, I F (ed.). London,Routledge, 1995.

2 Blom, D, Kristiansen, T. The Jev-naker project : two experiments withlive interactive television. Kjeller,Norwegian Telecom Research, 1988.(TF report R 58/88.)

3 Vaagland, O. Satellittoverført fleksi-bel utdanning i Hordaland : vurder-ing av et forsøksprosjekt. Bergen,University of Bergen, Institutt formassekommunikasjon, 1990.

4 Grepperud, G. Økt kunnskap er ogsåen utfordring. Tromsø, UNIKOM,University of Tromsø, 1996. (Work-ing draft.)


5 Vaagland, O. Økt kunnskap er ogsåen utfordring. Tromsø, UNIKOM,University of Tromsø, 1996. (Work-ing draft.)

6 Lorensen, M et al. Fleksibel utdan-ning via satellitt : et pilotprosjekt forembetsstudieundervisning i syke-pleievitenskap. Oslo, University ofOslo, Institutt for sykepleievitenskap,1991. (UiO, 1/91.)

7 Nordkvelle, Y, Fritze, Y. Vurder-ingsrapport for “Prosjektarbeid somlæringsvei”, second draft. Lilleham-mer, Høgskolen i Lillehammer, 1995.

8 Wong, A. Televised courses at theUniversity of Saskatchewan : Some-thing old, something new. Saskatoon,Canada, The University ofSaskatchewan, Division of Extensionand Community Relations, July,1989. (Evaluation Report.)

9 Keegan. Televised courses at theUniversity of Saskatchewan : Some-thing old, something new. Saskatoon,Canada, The University ofSaskatchewan, Division of Extensionand Community Relations, July,1989, 114–115. (Evaluation Report.)

10 Collins, V, Murphy, P. A new adultstudent : learning by interactive satel-lite. In: Developing Distance Educa-tion. Konferanserapport fra ICDEsverdenskonferanse i Oslo, 1988.

11 Wong. A new adult student : learningby interactive satellite. In: Develop-ing Distance Education. Konfer-anserapport fra ICDEs verdenskon-feranse i Oslo, 1988.

12 Holden, G. Multipunkt videokonfer-anse : evaluering av fjernundervis-ning via video, p. 45 and p. 66. Oslo,SEFU, 1994.

13 Wong. Holden, G. Multipunktvideokonferanse : evaluering av fjer-nundervisning via video, p. 45 and p.66. Oslo, SEFU, 1994.

14 Dahlén, S, Lundgren, H. Sjuksköter-ska på distans. Uppsala, UppsalaUniversitet, Pedagogiska Institu-tionen, 1995.

15 Grepperud, G. Fra prosjekt til per-manens. Tromsø, SOFF, 1993. (Rap-port 4.)

34

Telecom and Open Learning – a challenge to institutional cooperationB Y H A R A L D H A U G E N A N D B O D I L A S K


Introduction

1996 was declared the European year forlife long learning, stating the fact thatthere is an increasing demand in allwalks of life for updating and refreshingof existing knowledge and skills in orderto keep abreast with professional andsocial requirements. With significantrevisions of methods as well as in contentof professions, with tight working sched-ules and continuous challenges by newtasks and requirements, most profession-als tend to postpone the extra effort toembark on traditional continued educa-tion or in-service training programmes.

New information and communicationtechnology (ICT) has for a couple ofdecades posed increasing challenges tothe working force. A few adventurers andenthusiasts, on their own or by initiativefrom their employers, threw themselvesinto the technical whirlpool in the early1980s. Gradually, they were masteringdifferent operating systems, obscure userinterfaces and arbitrary pieces of soft-ware available for their purposes. Im-proved user friendliness as well as escal-ating capacity/price ratio and popularisa-tion of the technology have all contribut-ed to an increasing application of ICT atall levels.

The combination of life long learningand ICT naturally lead to programmesfor provision of in-service trainingthrough the use of computers and elec-tronic networks. Both trans-Europeanand different national initiatives havebeen launched in order to exploit the pos-sibilities incorporated in the new technol-ogy. Several projects are in progress,gathering experience and developing newmethodologies to suit both course provid-ers and learners. The electronic learningenvironment may be a challenge to thetraditional auditorium, classroom andhotel based seminars.

Cooperation between course providinginstitutions seems to be a fruitful trackfor saving efforts and resources. Throughdifferent projects in the field of ICTbased ODL one has found that partners inproduction of learning material benefit alot from exchange of electronic lessons,examples, programmes, etc. They tend tosave much on development costs,administrative systems and infrastruc-ture, and at the same time increase thequality of final products and offers to thetarget groups.

Future role of

professionals

Life in the information and knowledgesociety puts new demands on profession-al roles and performances. Not only doesthe technology itself change the charac-teristics of traditional work, it also shiftsthe required skills for certain operationsinto completely different levels of under-standing or detailed knowledge. In onedirection the visible results of efficientworking in a store is changed from thatof manually noting and adding numbersto that of pressing the correct buttons ona calculator or cash register – or movingthe bar codes past a photo cell or similardevice. On the other hand, the engineerdoes not need the same drawing skills asbefore when constructing a house or aship, but must control the drawing facili-ties of a computer, requiring a higherlevel of understanding and insight intopractical consequences of certain deci-sions.

In teaching as well as in company train-ing programmes the role of the presenteror trainer is gradually changing from thatof being a knowledge provider and a wellof solutions to all problems, into being aguide or adviser for learners to find andchoose from the information and instruc-tional material available. In a societywhere information and our access to itare growing exponentially, it is becomingincreasingly important to assist learnersin acquiring skills of information handl-ing. An important part of this is todevelop a critical and reflective way ofsorting, discriminating and interpretingthe information provided. They must alsobe able to turn the acquired informationinto knowledge and understanding oftheir immediate surroundings and thechallenges facing them.

IT and telecommunication

Particular requirements are related tonew technology as part of new profes-sional roles. ICT is not only a tool and asource of information, it is also in its ownright an important part of workmanshipand required skills of future citizens. It iswell accepted that a clerk expecting tomaintain his work in the future mustknow how to apply word processing, andthat an accountant will be obsolete if hedoes not know how to handle computerbased accounting systems. Maybe it isalso accepted that a journalist or a historyteacher must know how to search in

databases for interesting extensions to hislessons on the present situation in theMiddle East. But is it realised as equallyimportant for a social science/biologyteacher or an environmental adviser tomanage system dynamics and simulationmodels in order to understand orcommunicate the challenges of local orglobal pollution?

Telecommunication seems to play anincreasingly important role in future soci-ety. Capacity, bandwidth and speed oftransfer for large amounts of digitiseddata, including sound and graphics, iscovering greater parts of the world everymonth. From being a separate technologyit is now accelerating its integration intocomputer and other information tech-nologies, changing the PC from a stand-alone device into a small node in aworld-wide network of information.Large databases are included in the samenetwork and professionals of all kindsare meeting in the virtual space to discusstheir problems and points of interest.

Open and Distance

Learning (ODL) as a tool to

cope with new demands

The ever increasing demand for updatedskills and knowledge in different fieldsmakes it impossible to cover it allthrough traditional education or trainingsystems. Companies and public institu-tions have long traditions in arrangementof courses in-house or at hotels, etc.when new skills or principles were to beintroduced into their activities. No creditsor formal qualifications were attached tomost of these courses. It was part of thein-service training offered by theemployer and a requirement to be up-dated on the task to be performed.

An alternative for raising the qualifica-tions of the staff was to let a selected fewattend formal education programmes,acquiring new degrees or diplomas.Expenses related to such programmescould be borne by the employer, by theprofessional her-/himself or as a jointagreement between the two parts. Costscould be considerable for study pro-grammes running over several years inorder to obtain a degree, even in caseswhere the education or training was pro-vided outside regular working hours.

Distance education has long been a prac-tical way of taking courses and obtaining


qualifications. Specialised institutionslike correspondence schools and openuniversities have provided thousands andmillions of citizens with interestingoffers of initial and further education. Inthe age of new ICT it is natural to applythe technology for distribution of learn-ing material. As a consequence of thisthe importance of distance in the learningprocess is changing. It is now becomingmore essential to talk about open learn-ing, possibly combined with learning at adistance into open and distance learning,ODL.

The definition of ODL varies accordingto organisation, structure and possibilitiesincluded. In this paper we will stick tothe interpretation where the openness isparticularly directed to the open access oflearning material through electronic net-works. The distance of the students is notof the same importance to us; they maywell be regular students inside the insti-tution, staff members or individuals inanother part of Europe. The importantprinciple is that university courses, learn-ing material as well as exercises andexaminations are available to the learnersin their part of the world. With certainrestrictions the services should also beavailable to them when it is required, i.e.asynchronous distribution and availabil-ity is the general rule.

Further and in-career training throughODL therefore makes it more easilyavailable, costs and regulated workinghours can be saved and the traditionaluniversities and colleges may be openedto the public for life long learning pro-grammes. The effort for traditional uni-versities in order to change some of theirtraditional mass lectures and demonstra-tions into transferable, ICT-based learn-ing material is quite a scary exercise. Itmay not even be possible or desirable inmany cases, in particular subjects or top-ics. The latter is certainly the case forsome of the special aspects, e.g. in train-ing of nurses or in teacher education,where important values can hardly bedetached from the human touch. Thereare, however, lots of topics and coursesrelated to teacher and nurses training thatmay successfully be offered through ICTbased ODL, and where positive resultsare already observed.

Challenges to institutions

The enormous efforts and resourcesneeded for providing ODL material tonew target groups of learners simplyrequires that parallel work and develop-ment of products should be avoided. Allpartners will benefit from cooperationinstead of competition, exchange ofmaterial instead of protection, specialisa-tion of competence instead of coveringall fields at every institution, sharing ofduties instead of performing doublework. In particular, the learners will ben-efit from the products and facilitiesresulting from such cooperation.

There must be regulations and rules toplay by, however. Models for coopera-tion between academic institutions forthe purpose of ODL have been tried outin several European projects, providingvaluable experience as a basis for revi-sion of models and development of newways for providing ODL to the optimalbenefit for both students, staff and theinstitutions as such. Most valuable resultsand experiences have been acquiredwhen both academic institutions andreceiving organisations have been work-ing together in the planning and develop-ment of ODL courses and material.Examples of such activities will be pro-vided in the following sections.

Institutional cooperation

Cooperation between academic institu-tions seems to be an educational policy,nationally as well as internationally,including the establishment of an educa-tional cooperative network between cen-tres of higher education and research.Several countries throughout Europestate in their White papers or other politi-cal documents principles and guidelinesfor cooperation and sharing of expertisebetween educational institutions. As anexample of how cooperation betweeninstitutions and exchange of experienceare stressed, the European Commission’sSocrates programme has partnership be-tween educational institutions as one oftheir criteria to allocate funding to pro-jects (Socrates guidelines 1996).

Enterprises throughout the world claimthat there is a huge need for updating ofpersonnel. To meet this need we alreadysee a growing tendency of real coopera-tion between enterprises and the formaleducational system in order to updatetheir employees. Normally, the employer

has the responsibility for updating theirstaff, but by now we more often see acombination of updating the staff byoffering them formal qualifications.Development of necessary curriculum toobtain formal qualification is based oncooperation of enterprises, between thecompanies and the formal educationalsystem, where the companies play a cen-tral role in describing the needs and alsoare the ones to find or provide the know-how.

Introduction of electronic networks, e.g.Internet, increases the functionality andfacilitates practical activities among co-operating institutions. Both academicstaff, learners and the outside learningsociety are supposed to benefit fromdiverse specialisation and expertise, shar-ing of duties between institutions.

Together, the collaborating institutionsmay become able to provide a greatervariety of studies to their learners thanany one of the institutions can manage onits own. Collaboration should emphasisethe importance of being able to utilise thevarious profiles, specialities and profes-sional expertise of the different institu-tions and enterprises. This should in-crease both the width and depth of thelearning environment.

One example of how institutional cooper-ation has been practised is the way theNITOL project has organised their work.(NITOL = Norway-net with IT for OpenLearning). Four academic institutions inNorway have successfully practised theirown model of cooperation since 1994.Much of the success in this pilot projectis probably based on long-standing rela-tions and trust between the key personnelinvolved. The project’s primary goalswere to:

• Gain experience and develop methodsfor ICT based, open, flexible and dis-tance learning at university level

• Facilitate cooperative research anddevelopment between geographicallyseparated academic units and person-nel.

In order to approach these goals partnershad to:

• Establish an open network makingeducation within Norway-net availableto students and other groups andindividual participants from business,schools, administration, etc.


• Gain experience on distribution of edu-cational materials through electronicnetworks

• Test the use of electronic conferencesand mail systems as the basis for groupwork and co-operative learning

• Develop an extensive, dynamic andcreative electronic learning environ-ment based on local and wide areaelectronic networks.

Organising NITOL

Despite the close relationsand friendship between theactive partner representativesin the project, a formal organ-isation had to be establishedin order to be prepared fordisagreements or conflictsthat might arise over econ-omy, responsibility for stu-dents, networks, software,academic property, etc. Tomeet this requirement twoformal contracts were estab-lished, one general agreementof cooperation between thefour institutions, signed bythe heads, and one particularNITOL contract signed by theproject representatives fromeach institution.

The signed contracts aremostly dormant. In a fewcases it has turned out veryuseful to have them as refer-ence and guidelines for dis-putes that arise. The projectagreement is mainly in linewith the project group’s feel-ing of common sense, whilethe institutional contract is animportant document wheninstitutional authorities some-times forget their commit-ments to NITOL.

Technical support and stand-ards have proved to be im-portant issues for an ODLproject based on ICT. Thisalways turns out to be atouchy issue between experts.Among the ‘experts’ in such aproject are tens of teachersand hundreds of students –who all have their own, wellfounded preferences for soft-ware and communication sys-tems. On the other hand, thereare also hundreds of ‘non-experts’ or novices who need

to be given simple solutions that fill theirpresent needs. Among the latter groupare also some of the course providers. Itis therefore of utmost importance that theproject group is able to make firm deci-sions, stick to the solutions given, andthat there are able and committed peopleto support the technical system decidedupon. As new technology develops andbecomes available – outside the project –a careful balance must be found between

existing, stable solutions and introductionof new, improved facilities for the projectservices. Responsibility for these techni-cal decisions lies with the project group,while implementation and support areplaced with one of the partners.

Objects of collaborative

activities

Course material

Together, the collaborating institutionsare able to provide a greater variety ofstudies to their students than any one ofthe institutions can provide on its own.The collaboration emphasises the import-ance of being able to utilise the variousacademic profiles, specialities andprofessional expertise the differentinstitutions stand for. The colleges areable to enhance their curricula at bothundergraduate and graduate levels, thusproviding their students with a greatervariety of subjects than can be offered atany one institution alone. In addition,quite a few students want a more flexibleprogramme that allows them to study atthe time and pace of their own choice,with fewer restrictions on attendance andplace of study. Collaboration aroundflexible and open learning makes thispossible.

Because of the variety of professionalprofiles at the co-operating institutions,close collaboration strengthens the wholeteaching and research environment ateach institution. Experience is thatthrough electronic communication it ispossible to work together in a close andefficient manner in a “virtual” environ-ment of co-operation. This has alreadybeen established between the NITOLinstitutions, and is spreading throughoutthe rest of the IT environment. This is asystem that effectively establishes andutilises contacts across the boundaries ofinstitutions, regions and countries.

Course material is developed along threedifferent paths or models for institutionalcooperation:

The Open Access Model (Figure 1) isone of several models within the NITOLactivity. This was the first one tested,where each institution offered some oftheir traditional courses, revised or re-developed for electronic transmission totheir own students as well as to outsiders.

Commonmodule

Special-ised

module I

Spesi-alised

module II

CourseInstitutions

Figure 2 The Composite Model

Common course "pool"

Differentmodulespresentedin acommoncatalogue

Institution 1

Institution 2

Institution 3

Institution 4

NITOL

Figure 1 The Open Access Model


Visions

The models for ODL through telecom-munication, taking advantage of the insti-tutional cooperation described above,may open up for simpler access and high-er flexibility for all kinds of professionalsto obtain in-service and further educa-tion, as well as taking new formaldegrees, without leaving their regularwork/duties or their families. With thegrowing familiarity of Internet, telecom-munication and similar facilities inhomes and work places, the time maysoon be ripe for higher education mod-ules to be available when and where thelearner finds need or desire for it. Thiswill compose an important aspect in thepolicy and possible success of lifelonglearning.

The beneficiaries of this model weremainly the students, having easier accessand more courses to choose between. Theinstitutions and the professors gainedexperience in dealing with distant stu-dents and the colleagues’ ways of pre-senting learning material. A goldenopportunity to have a look at how col-leagues presented and designed theirteaching material was a fringe benefit.

The Composite Model (Figure 2) ispractised when the institutions composelarger courses by joining two or moresmaller modules developed at differentinstitutions.

The result could be a tailored study forspecial purposes or just an extension of asmall course, which a single institution isnot able to develop alone, due to restric-tions on time, staff, expertise, etc. Thismodel seems to have its strength espe-cially when colleges/institutions aresmall or medium sized, where resourcesand the variety of the staffs’ expertise islimited.

In the composite model, the institutionresponsible for the specific module, alsotakes responsibility for the students’exercises within the actual module. Elab-oration and marking of exercises are alsotried out as a collaborative, compositework, while assessment normally hasbeen dedicated to one institution.

Joint Venture Model (Figure 3) is amodel where the main idea is continuous,real collaboration between the professorsin the development of course work. Thismeans that two or more institutions with-in the network join forces in developingopen, flexible learning materials in theform of lessons. One or more lessonsforms a module, and the final course con-sists of several modules. Mature studentsand professionals jointly contribute to theunderstanding and learning of a topic orsubject. This is not utopia, it is already areality in the joint venture model. As anextension of the model, a virtual, jointventure learning society on the networkdoes not seem far away.

A first evaluation of collaboration,assessing the intentions and the modelsdescribed, identifies characteristics likeinspiration both to students and profess-ors, improvement in the quality as well asan increase in quantity of the coursesoffered. All parties involved seem to besatisfied and inspired to further effortsfor open and flexible learning

Mutual services

An important issue in the part-nership agreement is the mutualrecognition of credits, coursesand modules between the part-ners involved. Students maythus choose modules andcourses from a larger menu,combining it into more variedand specially tailored lines ofstudy. It is, however, still up tothe institution that issues thecertificate to recognise the finalcombination of subjects andcourses.

External funding has played acentral role in the early stages ofNITOL. Applications and pro-posals for funding have turnedout to be more positivelyviewed in cases where a collab-orative organisation has alreadybeen established. The NITOLpartners have therefore in sev-eral cases joined forces in the pro-cess of application, and after-wards shared the benefits obtainedthrough decisions by the projectgroup. The basic principle is sharing offunds acquired in equal parts. Often thegroup agrees to reserve particular sumsfor special services, e.g. technical sup-port, printing of catalogues, administra-tion, etc., and then to share equally theremaining sum.

Marketing

Instead of competing over the same targ-et groups with parallel courses in relatedfields, the NITOL group has consequent-ly, from the start, issued a commoncourse catalogue for all the courses pro-vided by the group. Even though thecourses are provided by individual staffmembers at separate institutions, andmarked this way in the catalogue, thereare great advantages in joint marketing.Distribution of information becomes eas-ier, potential students get a betteroverview, and large scale customers, likethe Navy, can find a more varied offerfor their members.

After only two years the NITOL conceptalready has more meaning to a lot peopleinterested in ODL, than the name of eachof the partner institutions. At the sametime, each institution benefits from be-longing to this group.

Institution I, II & III compose modulesInstitution IV has the main responsibility for thefinal structure of the module and for organisingformalities (assessment, credits etc)

Institution 1

Institution 2

Institution 3

Institution 4

I

II

III

IV

Cooperative development of modules

Figure 3 Joint Venture Model

38

From electronic mail to Internet – where does it take us?B Y T O R S T E I N R E K K E D A L


Introduction

This article discusses the theoreticalbackground and practical reasons behindthe decision of one specific institution,NKI (Norsk Kunnskapsinstitutt), to re-search on and develop distance educationbased on computer mediated communica-tion, and reports experiences from teach-ing through electronic communication fornearly 10 years. Thus, perhaps a morecorrect title of this article could have been‘From correspondence teaching to organ-ising individual and group learning on theInternet’, as our decision to direct ourefforts to develop a distance educationsystem based on electronic communica-tion, was taken as a result of our continu-ous experiments on developing the NKIdistance education system. Computermediated communication was in this con-nection considered to be a medium quali-tatively different from all other techno-logical developments of media for dis-tance education because of its flexibilityand its potential of integrating presenta-tion of learning material in differentforms and capacity for individual, smallgroup and large group communication.

NKI is a multiform teaching institutionoffering full-time and part-time pro-grammes on secondary and tertiary level.NKI courses and programmes are offeredon-campus, by distance education or asdecentralised courses. The distance edu-cation programmes may be supplementedby local face-to-face classes or seminars.NKI Distance Education offers a numberof programmes at university and collegelevel. Some of these are organised in co-operation with Norwegian universities orour own “Polytechnic College” (DenPolytekniske høgskolen). The majority ofthe NKI distance education programmesare secondary level studies preparing forpublic or internal exams.

Distance education and

new media

During the last 25 years distance educa-tion nationally and internationally hasgenerally changed from organising indi-vidual learning based on printed materialand two-way communication via theordinary postal system to multi-medialearning based on an integration of printbased media and new media and commu-nication technology. NKI has followedthese developments closely and tried todevelop and modernise the distance edu-cation system along different lines:

1 Student support andcounselling

The individual adult learner has generallya great need for support to be able tocope with the studies normally in compe-tition with other demands from joband/or family responsibilities. Researchhas shown that distance education sys-tems should develop administrative andteaching support systems to help the stu-dents adapt to the demands of part-timestudies to succeed and complete. In thisconnection we have researched on fol-low-up systems, initial and continuouscounselling, training in distance learningstudy techniques, turn-around time in-volved in two-way communication andspecific training for distance tutors (seee.g. [21]).

2 Media research

The correspondence teaching system ofthe 1960s has gradually been changedinto a multi-media distance teaching sys-tem by the introduction of video-tapes(both video-taped lectures and completelearning programmes) audio-tapes, com-puter software and laboratory kits, aswell as, in some cases, video-conferenc-es, radio- and TV-programmes. Some ofour research projects in this area havebeen carried out in co-operation withother distance teaching institutions andTelenor. Supported by and in co-opera-tion with Telenor we have carried outexperiments on telephone tutoring [22],telefax as a medium for individual tutor-ing and guidance [24], audiographics[25], video, local cable television, satel-lite distribution and video conferencing[7], [8].

3 Computer mediatedcommunication

Computer mediated communication hasbeen one specific priority branch of themedia research since 1986. NKI launch-ed the so-called “EKKO Project” in1986. The acronym, EKKO, denoted“Electronic Combined Education” (inNorwegian the combination of distanceeducation and local face-to-face classes)a name that signified that the computersoftware should constitute a virtualschool or classroom substituting the needfor physical presence in a local class. Theaim was to develop what we called the“Electronic College”, a teaching systemoffering study programmes independentof time and space and facilitating flexible

communication for administrative, socialand teaching/learning purposes. Ourunderstanding was that the developmentswithin computer communication wouldconstitute teaching and learning possibil-ities which really could change distanceeducation dramatically. This articleaccounts mainly for this research, ex-periences and achievements (see e.g.[15], [16], [18], [19]).

Technology and

competitiveness of

distance education

institutions

The rapid developments concerning newmedia and communication technologiesconstitute both new possibilities and pos-sible dangers for distance teaching insti-tutions. The new media create, at least intheory, new possibilities for preparingbetter and more cost-effective learning.At the same time these developmentshave resulted in new types of institutionsentering the market of education, andhave incited traditional schools, collegesand universities to offer media basedteaching. Thus, distance teaching institu-tions and universities have encountered,sometimes unexpected, competition onthe educational market. This situationwas stated in extreme form by TonyBates in an introductory article in anissue of “Open Praxis” focusing on tech-nology under the title: “Hello technol-ogy! Goodbye, distance teaching institu-tions?” [1]. The changes in distance edu-cation caused by the emerging technolo-gies have made some writers describe thenew initiatives as ‘third generation’ dis-tance education [12], the first being cor-respondence education and the secondmulti-media teaching. Bates foresee thatthe technological developments requirenew types of organisations, these newtypes can either develop from distanceeducation institutions, universities andcolleges or new institutions created fromscratch. Until now, it seems that distanceeducation institutions have not changeddramatically concerning the organisationof their teaching and their application ofnew media. For instance, Bates ([2], p.20) states:

“There is more talk than action aboutthe use of technology in distance edu-cation. Even in the most technologi-cally advanced of our member(EADTU) institutions, print, corre-spondence and face-to-face teaching


still predominate. For most Europeandistance learners, these are still theonly media currently that really mat-ter.”

There is reason to believe that this is thesituation also today, even after the explo-sion of interest in the Internet since 1993.According to Bates [2] there are goodreasons why the technological develop-ment has been so slow in the distanceeducation institutions. One is, obviously,that distance teaching institutions havelong traditions and investments bound upin the ‘old’ technologies and the naturalinertia of large institutions acts againstrapid changes. However, there are alsorational reasons for the slow technologi-cal developments. Print, correspondenceand face-to-face are well tried methods.As Bates ([2] p. 21) puts it:

“The use of more advanced technologycan be justified only if it meets one ormore of the following criteria: lowercosts; greater teaching effectiveness;increased accessibility to students.These are proving hard criteria tomeet, so it is not surprising that thereis still major academic and manage-ment resistance to the use of new tech-nologies in most of our member insti-tutions.”

We shall briefly comment on two otherwriters who have discussed the optionsavailable for distance education provid-ers to be able to compete in the new tech-nological environments, one representingthe distance teaching universities, theother representing secondary education.John S. Daniel [4], vice-chancellor of theBritish Open University, concluded in hisanalysis of the competitive advantages ofthe ‘mega-universities’ that:

“... networking students from theirhome computers should reinforce thecompetitive advantage of the mega-universities. Distance education hasalready evolved through two genera-tions, correspondence courses andmulti-media packages. The knowledgemedia (‘the coming together oftelecommunications, television andcomputing is producing a media envi-ronment for distance education that ismore than the sum of its componentelements’ (p. 11)) represent a thirdgeneration of supported open learningthat enriches distance education bygiving students rapid communicationwith the people and learning resourcesof the academic community.” [4]

Margaret Gamlin [5] of the New ZealandCorrespondence School discusses thetransition of distance teaching institu-tions from print and postal-based deliv-ery of traditional correspondence educa-tion to a more immediate interactivetechnology-based delivery. She pointsout that the drive in many countries todayto a competitive education system en-courages conventional providers to usetechnology for innovative delivery ofcourses. However, most current tech-nologies (such as audio and video con-ferencing, including audio graphics) tendto support the replication of the con-ventional classroom – the extended class-room model. She stresses the ‘openness’of the correspondence education modeland stimulated among others by Bates’article [1], she foresees a development inher institution, which builds on this open-ness and flexibility and learner centredfocus, and changes the school’s teachinginto a ‘multi-media’ model.

The above considerations are the reasonsbehind NKI’s decision to go for thedevelopment of computer-based commu-nication and the development of ‘TheElectronic College’.

NKI Electronic College

– 10 years experience of

computer mediated

communication in

distance education

In one of our early papers on computerconferencing titled ‘Computer conferenc-ing – a breakthrough in distance learningor just another technological gadget?’[17] we discussed developments ofmedia and technology and concluded thatcomputer conferencing or computer-mediated communication constituted adevelopment qualitatively different fromall other media with exceptional possi-bilities for developing flexible and opendistance teaching of high quality. After10 years of experiences we believe thatwe may conclude that computer mediatedcommunication in some form willbecome an important technology in dis-tance teaching and learning and overtime change the whole concept of whatdistance education is. On the other hand,this has not at all happened so far.

The virtual school

The basis of our ideas for establishing the‘electronic college’ was largely takenfrom [6] as introducing computer confer-encing as a means to establish a ‘virtualclassroom’ with computer-based commu-nication structures similar to those norm-ally taking place in the normal class-room. The ‘virtual school’ as, we con-ceived our aim, should not only emulatethe classroom activities, but all otherplaces and activities within the schoolsystem.

Morten F. Paulsen, the manager of theEKKO project [14], pointed out the fol-lowing requirements for the virtualschool:

1. It should emulate all the main tasks ofa school: teaching, administrative andsocial.

2. It should be generally available con-cerning geography, technology, econ-omy and student competence.

3. It should be independent of time, i.e.continuously available and acceptasynchronous communication.

4. It should emulate the different needs ofhuman communication, one-to-one,one-to-many and many-to-many.

Some important functions

of computer conferencing

in distance education

Based on NKI experiences and informa-tion from other sources (e.g. [13]) someimportant functions for the computerconferencing system in distance educa-tion were identified:

Distribution of information: Distanceteaching systems have a large need forincreased efficiency in updating anddistributing information to the students,full-time and part-time teachers andadministrators.

Examples: Information about courses,seminars, student associations, examina-tions and updating of learning material.

Two-way communication betweentutor/counsellor/administration andstudent: In most distance teaching sys-tems, submission of assignments for cor-rection and comments is an importantelement. It has been demonstrated thatlong turn-around times may have


destructive effects on course completion[20]. It also takes a long time for studentsto receive answers from their tutors whenthey really encounter problems in theirstudies. To some extent, the telephonehas been applied as a means of communi-cation. Electronic mail is independent ofboth time and space.

Examples: The student may ask ques-tions at any time, without the time delayof mail services. In principle, draft solu-tions may be submitted and commentedon, thus introducing a more flexibleorganisation of tutoring and assessment.If desired, student answers may be madeavailable to other students, before or afterthe submission of their work. Included inthe system can also be on-line computer-scored tests, as a substitute for off-linetesting which we have seen in some dis-tance learning systems. On a higherlevel, the two-way communication mayserve as a guidance for individual studentprojects.

A substitute for face-to-face teaching,introduction of group discussions andproject work: A number of distancelearning systems include the possibilityof face-to-face meetings with tutorsand/or fellow students. For many dis-tance learners, the possibilities of takingpart in such activities are restricted.Some theorists have argued that directteaching may have disruptive results onstudent autonomy and ability for self-study.

Examples: While face-to-face teaching indistance learning systems often seems tohave developed into lecturing/presenta-tion of subject matter, computer confer-encing concerns information exchangeand discussion. Discussions taking placein the classroom can develop into excit-ing experiences of group learning. Thediscussion is time and space independent,the medium seems to foster equality ofstatus between students, and between stu-dents and tutor. Specifically designedgroup learning methods may be applied,such as group submission of assign-ments, group learning and presentations,group seminars and project work.

The public tutorial: Student questionsof a general academic or administrativenature may be accessible to all students,as a question from one student normallywill be of interest to others. Pre-producedcomments on general aspects of a coursecan now be distributed on-line, and thetutor is given an opportunity to expandon the pre-produced learning material.

Peer counselling: As peer counsellingand informal co-operation is a naturalpart of the on-campus activities of anyteaching institution, the possibilities incomputer conferencing are obvious. Ithas been demonstrated that computerconferencing in general may give peerhelp in solving problems – often from an“unknown friend”. In large-scale sys-tems, where hundreds of students arestudying the same subject, peer help maybe of particular importance.

Free-flow discussion: A number of edu-cational conferencing systems haveformally established informal meetingplaces for continuing discussions such asthe “Cafeteria”, or “Local Pub”. Throughthe computer, informal discussions andstudent association activities may beincluded.

The Library: A collective database canbe developed within the conferencingsystem, to facilitate the availability ofrelevant articles, short lectures, etc. to thedistance learner.

Registration, administration, teacherconferences, etc.: Modern distancelearning systems have developed com-plex administrative systems for studentmonitoring. These systems can andshould be integrated with the conferenc-ing facilities.

Development of teaching material: Thesystem may efficiently be used for co-operative development of printed mate-rial – both within the institution and be-tween institutions.

User directory: The system containsinformation on its users, e.g. where onemay find fellow students with commoninterests. Phone numbers and addressescan be made available. The informationactually increases possibilities for directcommunication and via other media.

Initial stages of the EKKO

project

The aim of the project was to: Develop acomputer based conferencing system fordistance learning and apply the systemfor experiments in different contexts togain pedagogical and administrative/organisational experiences within dis-tance education based on computer con-ferencing – in order to install conferenc-ing as a standard option for NKI distancestudents.

The project followed these stages:

1. Introductory search in the field

2. Development of a specific conferenc-ing system on the NKI mini computer,HP 3000

3. Pilot experiments with “on campusstudents”

4. Study visits to institutions in Europeand North America

Home page of the NKI Electronic College


5. Pilot try outs in distance learning

6. Introducing computer conferencing ona larger scale in distance learning.

The First Generation:

1986 – 1993

The first version of EKKO – the com-puter conferencing software emulatingthe ‘Electronic College’ – was designedand implemented during 1986. Duringautumn 1986 we carried out the first pilotexperiments with on-campus students,autumn 1987 the first distance educationcourse was delivered to a small group of4 students. The next semester, spring1987, two additional courses were offer-ed. From the spring semester 1990 NKIDistance Education has been offering acomplete college programme (equivalentto one year of full-time studies) based oncomputer mediated communication. Thisprogramme in ‘Administrative comput-ing’ includes 10 different courses totally.In addition, from the same semester, NKIoffered its distance training programmefor distance tutors through the same sys-tem.

During its most intensive period EKKOserved more than 3000 users, includingon-campus students, prospective stu-dents, active distance students, formerstudents, tutors and administrative staff.The system included an e-mail system,closed and open conferences for adminis-trative, teaching and social purposes, andbulletin boards. During the first genera-tion period the ‘Electronic College’ de-livered more than 1000 courses with anaverage completion rate of above 80 %.

The following were some of the effectsand experiences of the first generation ofthe ‘Electronic College’ at NKI:

From 1990 NKI was one of a few institu-tions world-wide delivering a completestudy programme based on electroniccommunication. Prospective distancetutors could qualify for their workthrough distance education based on com-puter communication. The system wasapplied also for administrative communi-cation and discussions. Full-time staffmembers had been qualified to teachthrough the conferencing system, and apart-time staff of competent computerconferencing tutors has been built up.

Computer conferencing had been appliedin subjects with different didactic solu-

tions; subjects emphasising individualstudy, subjects with emphasis on groupdiscussion and in project work. To beable to exploit the possibilities for dis-cussion, group learning and peer support,most of the initial courses required thestudents to start at the same time and fol-low a fixed common progression duringthe semester towards the exams. Thismade the computer conferencing coursesless ‘open and flexible’ than the ‘corre-spondence type distance courses’. Manystudents express clearly that they prefer –or demand – the freedom known fromun-paced individual study. Thus, courseswhere the students can start wheneverthey wish and study at their own pace hasalso been offered. The students who haveexperienced this individual freedom intheir studies have also been generallypositive, and we see a great challenge indeveloping didactic arrangements com-bining the flexibility of individual un-paced study with the possibilities forsocial interaction in the virtual schoolenvironment.

We also learned during the first genera-tion experiments that teaching via com-puter conferencing often becomes“labour intensive” on the part of thetutor. Originally, we had a hypothesisthat the supposed increase in learningquality in the “virtual school” could becompensated for through less emphasison the development of learning material.So far, our experiences have not support-ed this assumption. Thus, it seems thatinvestments in pre-produced learningmaterial will be approximately the sameas in other large scale systems if the totalquality is to be satisfactory.

The tutors have reported that teaching viaEKKO has been very stimulating, butvery demanding concerning the numberof hours used and the “continuous” atten-tion needed. We see a challenge in devel-oping teaching/learning strategies thatstimulate student-student communicationwithout putting unrealistic demands ontutor resources.

Concerning student participation, NKIexperiences and experiences in other set-tings, such as the British OU [11], provethat conferences may become too smallor too large. The users fall into groups of‘active’, ‘less active’ and ‘passive users’.There must be a sufficient number, or ‘acritical mass’ of active users to make theconferences attractive.

It seems necessary to have an active con-ference moderator in the conference. Thetutor (or another person) must take therole as organiser, active contributor,and/or social integrator. After we form-ally engaged one student as host of the‘on-line cafe’, the social activity and stu-dent satisfaction and motivation seemedto increase considerably.

The teaching/learning and administrativefunctions which can be handled by thesystem, and the quality and quantity ofcontributions depend on whether partici-pation in the system is voluntary orobligatory. Voluntary participation mayresult in low activity which may becomea self-fulfilling circle of diminishinginterest.

In our computer conferencing courses wehave experienced a somewhat highernumber of non-starters than in compar-able correspondence courses, probablydue to initial technical difficulties, andhigher completion rates among the start-ers. Students having completed coursesbased on computer conferencing, achiev-ed better at exams than both correspond-ence students and face-to-face students.We do not have data to make certainwhether these differences are a result ofcharacteristics of the media/methods orby differences in recruitment.

In general, the students report favourableattitudes to computer conferencing as aform of study. They seem to be moreactive in the social conferences than inthe academic conferences. This may be aresult of the fact that most NKI coursesdepend highly on individual work andassignments for submission rather thangroup discussions. Most students statethat the mail system is the most import-ant subsystem of the first generation ofEKKO. This may indicate that the opti-mal way of organising the studies has notyet been found. On the other hand, itseems that many part-time distance learn-ers have to find time efficient studystrategies to be able to cope with thedemands of study beside job and familydemands. This means that many wouldnot give priority to social or academicinteraction relative to individual studyand exercises in preparation for theirexams. These experiences seem to be inline with other findings indicating thatpart-time distance learners often adopttime saving techniques in their approachto learning to survive the demands ofstudy beside job, family and society pres-sures [9], [10].


The Second Generation:

from 1994

NKI considered the first generation ofthe electronic college to be quite a suc-cess – both as a computer system as such,and also concerning how we managed toorganise the distance learning system.We continuously followed other develop-ments in teaching/learning methods andsoftware on the market and examineddifferent products, such as CoSy, Porta-Com and FirstClass, with the aim ofdeveloping an improved ‘second genera-tion’ system. When we had to introducenew solutions because of retirement ofthe old host computer, the requirementswere that the system should

• Be based on standard software

• Provide access to the Internet

• Be attractive to all NKI departmentsand available from any computer net-work within NKI

• Be attractive to possible collaborators

• Be as user friendly as possible.

From January 1, 1994 the new ‘open’Electronic College was introduced. Thissecond generation system was based on aphilosophy of being as open as possibleto other networks and services. Accord-ingly, it was based on Internet, e-mailand the Listserv conferencing system.From the beginning the user interface formodem users was text-based. From Jan-uary 1995 modem users were offeredcommunication software developed forMS-Windows and SLIP (Serial LineInternet Protocol). The second generationstimulated the development of many newcourses and study programmes, such as acomplete programme on InformationNetworks and courses in the use of Inter-net, WWW-presentations, Java-program-ming, etc.

All the courses and programmes develop-ed after the introduction of the secondgeneration system are un-paced and putno limits on times for enrolment. Thissolution has been chosen as a conse-quence of our conclusions from an inter-view survey among EKKO students onrecruitment and study barriers:

“... it is a major challenge to developmethods and organisations in distanceeducation based on computer confer-encing systems which take care of thestudents’ need for autonomy and flexi-bility.” ([23], p 92)

As mentioned, for the second generationsystem NKI decided to use Internet fo-cusing on SLIP, e-mail and ListProcessor(a Unix version of the Listserv confe-rencing system). Listserv offers anadvanced distribution system which maybe explained as a conferencing systembased on e-mail. Such systems can sup-port more users than any other confe-rencing system because they communi-cate with the system via e-mail. The listscan be configured as open or closed, forinstance open to members of the wholeelectronic college (such as the ‘on-linecafe’), a class or a specific group (such asteachers). Lists can further be configuredas one-to-many (information and bulletinboards) or many-to-many (conferences/class discussions) lists. Each list is man-aged by a ‘list owner’ via e-mail. In addi-tion to the lists directly related to teach-ing, social and administrative purposes ofthe electronic college, the system includ-es some international lists: ‘Andrea’, anopen one-to-many list for distance edu-cators in Europe‚ ‘Norwaves’, an openone-to-many list for distribution of‘News from Norway’; and ‘Norweave’,an open many-to-many list for ‘Inter-national friends of Norway’.

The first semester with the new genera-tion was marked by transitions andadjustments to the new system. In someways, the old system was felt more toemulate an ‘image’ of a virtual school.The open system introduced new prob-lems due to lack of standard interpreta-tions among e-mail systems. However,students and staff seem to have adjustedand become familiar with the new sys-tem. The general access to Internetresources outside the NKI realm seems tobe an advantage appreciated by the stu-dents. Access to Internet is offered to allstudents at tertiary level and at the NKITechnical College and consequently,application of Internet resources can beintegrated in some of the study pro-grammes.

Recently, NKI introduced the first threecourses based completely on electronicdistribution and communication, Javaprogramming, WWW-presentations andMulti-Media networks. In these cases thelearning material is available on theInternet/WWW only.

We have just started to adapt the Coursefor Distance Tutors for presentation onthe WorldWideWeb with links to internaland external national and internationalresources.

Some conclusions

We consider the decision to base theelectronic college on Internet, publicavailable software and with ‘open’ solu-tions, to be a sound conclusion. There isreason to believe that more efficient anduser friendly software adapted for teach-ing and learning will be introduced andavailable for our purposes.

Access to Internet and use of Internet areappreciated by students and tutors andopen opportunities for sharing of experi-ences, learning from others and new pos-sibilities for arranging effective learningsituations.

Designing attractive and effective learn-ing is no easy task. To repeat what wehave said many times, there is a long wayto go before we know how to design forlearning adapted to the ever changingpossibilities of computer mediated pre-sentation and communication.

Adapting organisational and administra-tive procedures and systems to the activi-ties of the electronic college is not a triv-ial matter. It may include major changesin organisational systems, working stylesand administrative computer systems.

Computer based communication isdemanding and time consuming both foradministrators, student counsellors andtutors.

There are still many questions thatremain unanswered or that are only par-tially answered. For instance, how willstudents generally accept electronic com-munication in distance education? Today,after nearly 10 years of delivering theNKI computer science programme viacomputer mediated communication andthe same study by correspondence meth-ods, two thirds of the students (whostudy computer science, who have toapply computers and software as part oftheir studies) prefer correspondence edu-cation rather than computer mediatedcommunication. One reason might ofcourse be the differences between thesetwo options concerning flexibility andfreedom.

An important question remains: How canthis technology be applied to represent amore attractive and efficient learning envi-ronment for distance students, and howcan we design the courses and tutoring andadministrative systems to make distance


education based on computer mediatedcommunication really cost-effective?

Sometimes it seems that educators be-lieve that presenting the students to theInternet as a resource of information andmeans of communication solves theproblem of offering effective and effi-cient education at a distance. Our con-clusion is that this is not at all the situa-tion. The challenge for distance teachinginstitutions is to find ways of planning,organising and carrying out teaching inways who help students learn accordingto individual needs independent of timeand place restrictions. This means thatthe medium itself does not solve anyproblems [3]. We must learn how todesign instructional programmes includ-ing student learning activities based onthis new medium to achieve optimal out-comes for different kinds of learners indifferent subjects having different aimsand objectives.

References

1 Bates, T. 1994. Hello, technology!Goodbye, distance teaching institu-tions! Open Praxis, 2, 5–7, 1994.

2 Bates, T. The challenge of technol-ogy for European distance education.In: Media and technology in Euro-pean distance education. Bates, A W(ed.). Heerlen, EADTU, 1990.

3 Clark, R E. Media will never influ-ence learning. Educational technol-ogy, research and development, 42,(2), 21–29, 1994.

4 Daniel, J S. The mega-universitiesand the knowledge media : implica-tions of new technologies for largedistance teaching universities. Mas-ter thesis. Quebec, Concordia Uni-versity, Department of Education,1995.

5 Gamlin, M. Distance learning in tran-sition; the impact of technology : aNew Zealand perspective. Keynoteaddress to EDEN Conference ‘TheOpen Classroom’ Distance Learningand New Technologies in SchoolLevel Education and Training, Oslo,1995.

6 Hilz, S R. The virtual classroom :building the foundations. New Jer-sey, New Jersey Institute of Technol-ogy, 1986.

7 Holden, G. Videokonferanser som delav undervisningsopplegg i fjernun-dervisning. Bekkestua, NKI/SEFU,1992.

8 Holden, G. Multipunkt videokonfer-anse : evaluering av fjernundervis-ning via video. Bekkestua, NKI/SEFU, 1993.

9 Lockwood, F. Activities in self-instructional texts. London, KoganPage, 1992.

10 Marland, P et al. Distance learners’interaction with text while studying.Distance Education, 11, (1), 71–91,1990.

11 Mason, R. Refining the use of com-puter conferencing in distance educa-tion. In: Distance education : devel-opment and access. Croft, M et al.(eds.). Caracas, ICDE, 1990.

12 Nipper, S. Third generation distancelearning and computer conferencing.In: Mindweave : communications,computers and distance education.Mason, R and Kaye, A (eds.). Lon-don, Pergamon, 1989.

13 McCreary, E K and Van Duren, J.Educational applications of computerconferencing. Canadian Journal ofEducational Communication, 16, (2),107–115, 1987.

14 Paulsen, M F. En virtuell skole. DelI. Fundamentet i EKKO-prosjektet.Bekkestua, NKI/SEFU, 1989.

15 Paulsen, M F. EKKO : experiences.In: Media and technology in Euro-pean distance education. Bates, A W(ed.). Heerlen, EADTU, 1990.

16 Paulsen, M F. The NKI ElectronicCollege : five years of computer con-ferencing in distance education. In:Paulsen, M F. From bulletin boardsto electronic universities : distanceeducation, computer-mediated com-munication, and online education.University Park, Pennsylvania, TheAmerican Center for the Study ofDistance Education, 1992.

17 Paulsen, M F, Rekkedal, T. Com-puter conferencing : a breakthroughin distance learning, or just anothertechnological gadget? In: Developingdistance education. Papers submittedto the 14th ICDE World Conference

in Oslo. Sewart, D and Daniel, J S(eds.). Oslo, ICDE, 1988.

18 Paulsen, M F, Rekkedal, T. The elec-tronic college. Selected articles fromthe EKKO project. Bekkestua, NKI/SEFU, 1990.

19 Paulsen, M F, Rekkedal, T. Technol-ogy for adult learning in Norway :including a case study on the NKIelectronic college. Report for theOECD Study B6 on ‘New DeliverySystems and Changing Demand onEducation’, 1996.

20 Rekkedal, T. The written assign-ments in correspondence education :effects of reducing turnaround time.Distance Education, 4, (2), 231–252,1983.

21 Rekkedal, T. Introducing the per-sonal tutor/counsellor in the systemof distance education. Stabekk, NKI,1985.

22 Rekkedal, T. The telephone as amedium for instruction and guidancein distance education. Bekkestua,NKI/ SEFU, 1989.

23 Rekkedal, T. Recruitment and studybarriers in the electronic college. In:Paulsen, M F and Rekkedal, T. Theelectronic college. Selected articlesfrom the EKKO project. Bekkestua,NKI/ SEFU, 1990.

24 Rekkedal, T. Telefax som medium fortoveis kommunikasjon i individuellfjernundervisning. Bekkestua, NKI/SEFU, 1992.

25 Rekkedal, T, Vigander, K. Forsøkmed bruk av telewriter i matematikk-undervisning. Bekkestua, NKI/SEFU, 1990.

44

Design guidelines for WWW-based course environments B Y A N K E E E K M A A N D B E T T Y C O L L I S


WWW-based course environments arerapidly appearing, before there hasbeen time for much theoretical devel-opment with respect to guidelines fortheir design. In this article we describea set of 20 guidelines which could be acontribution to this development. The

guidelines are illustrated through theirapplication to a particular case, theredesign of a WWW-based course“Environmental Management”, now inoperation, with more than 120 learnersin six countries.

WWW-based course

environments

WWW technology is increasingly beingused to support course delivery, for bothface-to-face and distance learners (see [3]for an inventory and analysis). In somecases, the support is a new feature of anexisting course, outside of the courseitself. For example, the WWW is beingused as an information system forcourses, with both one-way broadcast-type communication from institution andinstructor to students as well as integrat-ed access to information related to thespecific learner. The WWW is alreadyfamiliar in the role of “virtual library”,providing hyperlinked access to courseresources, including those specific to thestudy material of a course; those involv-ing multimedia databases of images, textand even audio; those located external tothe course but linked to the course by thecourse designer; and those found by thelearner himself with the use of embeddedsearch tools. In addition, the WWW cannow integrate various forms of communi-cation support to course information andresources, where the communication cantake many forms and modalities. Contin-ual technical developments in the WWWnow allow integration of other digitalmaterials, such as tutorial and simulationprograms as well as various forms ofinteractivity via CGI scripting and Javaaplets. Groupwork and collaboration,both in real and deferred time, are alsonow supported by WWW-based tools.Furthermore, many courses are nowbeing entirely offered through WWWenvironments (for examples, see [4]).

Instructional design for

WWW-based courses

Given all these possibilities, it is not sur-prising that an increasing number ofcourses are including some combinationof these WWW functionalities, and as thecombination increases the WWW sitebecomes more and more the centralmedium for the course. Through oneWWW-based user interface, the learnercan access a wide range of resources andbe able to carry out a wide range of activ-ities. However, the rapid development ofthe WWW technology and the ideasbeing put into practice via the technologyhave not yet been supported by cohesivesets of instructional-design principles toguide their form and logic when they areto be applied to course design. Such prin-

Guideline 1: Be consistent in designing an interface, especially when using text,pictures or icons to instruct the learner.

Guideline 2: Use metaphors that correspond to the user’s daily life, so that he hasconcrete expectations about the result of an action.

Guideline 3: Set as few heading styles and subtitles as are necessary to organisethe content, then use the chosen styles consistently.

Guideline 4: Many basic typographical guidelines for paper lay-out design apply alsoto designing texts for computer screens. The page should look attractive, so make astructured pattern on the page.

Guideline 5: Consider carefully which colours to use and in which situation.

Guideline 6: Because icons and virtual metaphors are not unambiguously under-stood across all computer interfaces and international borders, they should beinvestigated before put into a page.

Guideline 7: Do not confuse the user by putting more than seven (navigational) iconson a page. The preferred number of icons on a page is five, plus or minus two.

Figure 1 Design guidelines for presentation aspects of WWW-based courses

Guideline 8: The user should get a clear overview of the structure of the hyperlinkedsite. This can be accomplished by adding navigational buttons which link to the pre-vious and the next page in the same sequence, as well as a link to a local homepage.

Guideline 9: The user automatically generates a mental model of a hyperlinked site.The user should be helped to make this model a structured one, by adding functionaland graphic continuity between the various components and subsections of the site.

Guideline 10: Menus lose their value if they don’t carry at least four or five links; textor list-based menu pages can easily carry a dozen links without overwhelming theuser or forcing users to scroll through long lists.

Guideline 11: Page design in hyperlinked environments should emphasise the powerof hypermedia links to take full advantage of this medium.

Guideline 12: The user should know what kind of information he can expect whenfollowing a certain link. Therefore, make the name of the link clear, or add some addi-tional information to the link.

Guideline 13: Don’t let the user get lost in the information. A page for a hyperlinkedenvironment should contain no more than two to three 640x480 screens worth ofinformation.

Guideline 14: Keep in mind that some users want to save or print the material ratherthan interacting with it as hyperlinked material from the computer screen. Make aseparate print file or design the pages so that the user (particularly the networkeduser) is able to print pages of the course in a reasonable fashion.

Figure 2 Design guidelines for WWW-based course environments relating to thehyperlinked aspects of those environments


ciples need to reflect at least four streamsof considerations: traditional instruction-al-design insights relating to how learn-ing takes place and can best be organis-ed; design insights from human-com-puter interface and human-computerinteraction research relating to principlesrelating to the usability of systems, in-cluding design considerations relating tolayout and presentation on computerscreens; design considerations specific tohyperlinked learning environments; anddesign aspects unique to the WWW en-vironment itself.

Design guidelines for

WWW-based courses

We have responded to this need for acohesive set of design guidelines forWWW-based courses through a carefulliterature review of the various designaspects noted above [7], and have devel-oped a set of 20 design guidelines as acontribution to this need [6]. We are nowapplying these guidelines in a number ofways, to test their applicability and tosupply insights for their on-going revi-sion. The guidelines themselves can begrouped into four categories: presenta-tion, hypertext, pedagogy, and technol-ogy. We present them briefly here.

Presentation aspects

Presentation aspects relate primarily tothe user interface of WWW-basedcourses, how the course is presented tothe learner (and instructor). Guidelines inthis context are strongly influenced byHCI and user-interface design experi-ence, and relate not only to specific con-siderations such as typography, colour,and the use of images, but also to overallaspects of interface design, such as theapplication of metaphors. From relevantliterature, (see for example [12]), wehave extracted the following seven pre-sentation guidelines as particularly rele-vant for WWW-based course environ-ments (Figure 1).

Design guidelines for hyper-text-related aspects of WWW-based courses

From the literature relating to hypertextsystems, (see for example [11] and [10]),particularly the problems that users mayencounter when navigating such systemsand the problems that learners may facein using such systems as learning environ-ments, we know that making the struc-ture of a hyperlinked system clear to the

Guideline 15: Enable the learner to study whenever and wherever he wants by mak-ing the learning materials accessible at any time and anywhere.

Guideline 16: Learner commitment and motivation can be increased by integratingquestions into the instructional materials.

Guideline 17: The learner should be able to answer questions within the learningmaterial. The answers to the questions should also be available for the learner,preferably after he has answered them.

Guideline 18: The learner should be able to get in contact with other learners, in theform of co-operative working or discussion as well as informal contact.

Guideline 19: The learner should be able to get some form of (local) tutoring.

Figure 3 Design guidelines for WWW-based course environments from theeducational perspective

Guideline 20: Both the learner and the tutor should have access to support servicesto help with technology-related aspects of the course experience.

Figure 4 Design guideline for WWW-based courses related to implementation support

Figure 5 Map of the design guidelines for an individual learning environment

31 65

42 7

presentation

98 1110

hypertext

13

12 14

1615 1918

pedagogy

17

20

technology

Figure 6 Index of the course materials


user is a key consideration. From this andother insights ([9] is particularly recom-mended), we derived the following sevendesign guidelines as particularly relevantto WWW-based courses (Figure 2).

Design aspects relating to ped-agogical aspects

The educational literature offers manydifferent insights into what can be ob-tained relating to the design of WWW-based course environments even thoughthe insights were developed beforeWWW environments existed. Some ofthese insights relate to the individual’sinteraction with learning materials inthemselves; others (particularly from thedistance education literature; see forexample [8]), relate to the support ofadult learners who are studying at a dis-tance from their instructor and class-mates. The following design guidelineswere extracted as particularly importantfor WWW-based course environments(Figure 3).

Design guidelines relating touse of the WWW technologyitself

Finally, WWW environments are a newform of technology for all concerned.From decades of experience with theimplementation of technological innova-tions in education, we know that instruct-ors and learners alike predictably en-counter confusion and frustration whenconfronted with a technology that is newto them (a review of this appears in [3],Chapter 7). Unless appropriate help isavailable, many never go past this frust-ration in order to continue with the newtechnology. Thus, the last of our designguidelines relates particularly to this im-portant aspect of WWW-based coursedesign, an aspect that may be outside ofthe course itself but critical to it (Figure 4).

We find it convenient to represent all thedesign guidelines in one graphic. Figure5 shows this representation.

Applying the guidelines

to the redesign of the

“Environmental Manage-

ment” course

An important way to validate andimprove design guidelines is to applythem in practice. From the designer’sperspective, the usefulness of the guide-lines during the design process can be

31 65

42 7

presentation

98 1110

hypertext

13

12 14

1615 1918

pedagogy

17

20

technologyGuideline 6: Becauseicons and virtual metaphorsare not unambiguouslyunderstood across all com-puter interfaces and inter-national borders, theyshould be investigatedbefore put into a page.

31 65

42 7

presentation

98 1110

hypertext

13

12 14

1615 1918

pedagogy

17

20

technologyGuideline 8: The usershould get a clear overviewof the structure of the Website. This can be accom-plished by adding naviga-tional buttons which link tothe previous and the nextpage in the same se-quence, as well as a linkto a local homepage.

31 65

42 7

presentation

98 1110

hypertext

13

12 14

1615 1918

pedagogy

17

20

technologyGuideline 16: Learnercommitment and motivationcan be increased by inte-grating questions into theinstructional materials.

31 65

42 7

presentation

98 1110

hypertext

13

12 14

1615 1918

pedagogy

17

20

technology

31 65

42 7

presentation

98 1110

hypertext

13

12 14

1615 1918

pedagogy

17

20

technology

Guideline 2: Use meta-phors that correspond tothe user’s daily life, so thathe has concrete expecta-tions about the result of anaction.

Guideline 18: The learnershould be able to get incontact with other learners,in the form of co-operativeworking or discussion aswell as informal contact.


evaluated. From the learners’ perspect-ive, the influence of the design guidelineswithin a course developed around thoseguidelines can be indirectly evaluatedbased on the learner’s reactions to theoverall course and its component aspects.Both of these perspectives were em-ployed during the application and valida-tion of the design guidelines with respectto a WWW-based course called “En-vironmental Management” (Version 2).In the following sections, we briefly de-scribe the course, illustrate a few of theways that the design guidelines wereapplied to its redesign, and give a sum-mary of learner reaction to the course.

The Environmental Manage-ment course

Version 1 of the Environmental Manage-ment course was developed in 1995 inthe framework of the TeleScopia Project,for trans-Europe delivery of coursesmaking use of a variety of technical plat-forms (see [1], [5]). One of the coursedeliverers in TeleScopia was the UETP-EEE (University-Enterprise TrainingPartnership in Environmental Manage-ment Engineering), a network of Euro-pean universities, enterprises and profes-sional organisations involved in environ-mental engineering. The EnvironmentalManagement course was especiallydesigned for the TeleScopia Project [13],and was the only WWW-based course inthat project.

Various evaluations were carried out ofthe first version of the course, and a sec-ond version, for a cohort of approxi-mately 120 learners participating duringthe period October 1996 and March1997, was designed1 [6]. Learners accessthe course via a WWW environment(http://www.dipoli.hut.fi/EEE; note that apassword is needed to enter, contact thefirst author for a guest password) via a

Figure 7 Index page of a module

Figure 8 Example of a question integrated in the learning materials

1 The second version of the course wasdeveloped and is used within the scopeof the ATLAS (Atlantic Mobility forAcademic Studies in Engineering andEnvironment) Project funded by theEuropean Union. The Lifelong Learn-ing Institute Dipoli at the HelsinkiUniversity of Technology, workingtogether with the University-EnterpriseTraining Partnership in EnvironmentManagement Engineering (UETP-EEE), is responsible for the re-designof the course.


modem or ISDN connection. Throughthe course environment, learners accessnot only the course materials and inform-ation about the organisation of thecourse, but also a variety of forms ofcommunication among themselves, andbetween themselves and local tutors andothers involved with the overall deliveryof the course. Local face-to-face tutoringsessions are also available. In addition,the course includes three internationalvideo-conferencing sessions which cur-rently are not integrated technically with-in the WWW environment. The languageused in the course is English.

Application of the design guide-lines in the design of Version 2

The design guidelines were applied toevery aspect of re-design of the Environ-mental Management course, Version 2.0(see [6] for a detailed summary). Becauseof space constraints here, we can onlygive a sample of how this applicationoccurred.

For example, the original course did notuse many graphics. In the materials sec-tion of the new EM course, use has beenmade of graphics, although they havebeen kept as simple as possible, andsmall in size (in kilobytes) because of thelong loading time that can occur whenusing sophisticated graphics. Figure 6shows an example of the presentationdesign of the course, in this case, themenu from the “Materials Index” compo-nent.

Guideline 6 was thus taken into accountin this design decision: icons should beinvestigated before put into a page.

Arrows are used in many user interfaces,perhaps the most universally understoodones relate to the interface of a radio-cas-sette player, with play, fast-forward andrewind buttons. Arrows suggest move-ment; an arrow that points to the left isinterpreted as going back, an arrowpointing to the right usually refers tomoving forward. An arrow pointing totop means usually “up”. Figure 7 showsthe index of Module 3 of the course. Thedouble arrows that point to the top referto the “material index”; the doublearrows pointing to the left take the learn-er to the index of the previous module;and the double arrows pointing to theright bring him to the index of the nextmodule. These decisions were made inreference to Guideline 8, the user should

Figure 9 Netscape newsgroup interface used for discussion in the EM course

Figure 10 Class of ’96-page


get a clear overview of the structure ofthe Web site.

Arrows are in this case navigational but-tons which link to the previous and thenext page in the same sequence. This atleast helps the user to have a sense of hisimmediate location in the site.

Questions integrated in the learningmaterials can increase motivation andinteractivity (Guideline 16). The EMCourse integrated several questions intothe learning materials. Figure 8 gives anexample.

In addition to the video-conferencingportion of the EM Course, Guideline 18is also applicable to a WWW-based two-way interactional element of the course,namely the Discussion component. Fromthe discussion page, the learner can jointwo newsgroups (see Figure 9). In thedesign of this part of the course, Guide-line 2, the use of metaphors was particu-larly important.

On the screen two buttons are visible thateach link to a different newsgroup. In thefirst newsgroup, learners can discuss thecontent of the course with other learners,with tutors, and with experts, as recom-mended in Guideline 18.

Since it is impossible to send the learnersin the EM course all over the world tomeet their fellow learners, an introduc-tion via video-conferencing has beenadded to the course. But the video-con-ferences are only short, and not all thelearners get the chance to talk to eachother. This is the reason that the Class of’96 pages have been integrated in thecourse (see Figure 10). These pages arebased on the idea of the “yearbook” or“almanac”, where pictures and smallbiographies about the learners are pre-sented. At the Class of ’96 site, thelearner is free to look around in the pagesof the tutors, to get a glimpse of thetutors, the experts, and of course theother learners, if they have added theirown “homepages”.

In the virtual workshop, associated withthe “Class of ‘96” component, the learnercan easily create his own Web page. Thisactivity is not compulsory, and does nothave any impact on the learner’s studyresults. In the virtual workshop, thelearner can fill in a standard CGI-basedform, displayed in Figure 11.

Figure 11 Form for adding a “personal homepage”

31 65

42 7

presentation

98 1110

hypertext

13

12 14

1615 1918

pedagogy

17

20

technology

used one time

used two times

used implicitly throughout the whole design process

Figure 12 Guidelines used in the design of the new EM Course


As can be seen in Figure 11, the learneris restricted in the amount of text andpictures he can add to his own page. Thisis done because the pages are mainlymeant for introduction, but also becausethere should be some consistency be-tween the learners’ homepages.

The above are just a sampling of exam-ples of how the design guidelines wereused to inform the many decisions thatwere taken in the re-design of theWWW-based EM course. Figure 12shows a general summary of how oftenthe design guidelines were applied toVersion 2 of the EM course.

Learner validation of theguidelines

The EM course is in session from Octo-ber 1996 through March 1997; thus, atthe time of writing of this report (No-vember 1996), only a first impression ofthe course could be obtained from thelearners. A survey returned by the learn-ers after their first experiences with theWWW-based course shows very positiveresponses: the learners feel they have aclear understanding of the structure ofthe course environment and particularlyvalue the easily-made personal homepages within the course. These and avariety of other comments from theWWW-based learner evaluation aregiven in detail in [6]; the important con-clusion to date is that the WWW-basedcourse environment appears to be wellreceived by its distributed learners. Thisin turn implicitly supports the designguidelines on which the course is based.

Other validation

approaches for the

design guidelines

Certainly more than one experience ofapplication of the design guidelines isneeded for their validation and refine-ment. Another type of validation is alsotaking place: the use of the design guide-lines within a course focused on the spe-cific topic of the design of WWW-basedlearning materials themselves. Such acourse (WWW-based itself) is in opera-tion at the University of Twente [2], andthe design guidelines have been integrat-ed into the Study Materials for thecourse. In addition, they are being usedas the criteria for the learners’ formativeevaluation of their self-designed WWWmaterials, and will be used as the criteriaused by the course instructors for sum-

mative evaluation of the learners’WWW-based products. The course site,integrating study materials discussing thedesign guidelines and various instru-ments for formative evaluation of WWWsites based on the guidelines, is open forinspection, so that this aspect of the vali-dation process for the design guidelinescan be seen in operation [2].

In conclusion, to date it appears that thedesign guidelines form a valuable contri-bution to support the design process forWWW-based course environments.However, it is also clear that the guide-lines should be continually evaluated andrevised, to refine their usefulness and testtheir applicability in a variety of contexts.

References

1 Collis, B (ed.). The TeleScopiacourses : experiences with the adapt-ation process for trans-Europeantele-learning. TeleScopia DeliverableUT/ DL1001/WG 3.3. Bonn, Deut-sche Telekom AG, Generaldirektion,1995.

2 Collis, B. The course ISM-1. 1996.[WWW-based course, Faculty ofEducational Science and Technology,University of Twente: seehttp://www.to.utwente.nl/ism/ism1-96/home.htm]

3 Collis, B. Tele-learning in a digitalworld : the future of distance learn-ing. London, International ThomsonComputer Press, 1996. [see WWWdocument: http://www.itcpmedia.com]

4 Collis, B, Andrenach, T, Diepen, Nvan. The Web as process tool andproduct environment for group-basedproject work in higher education. In:Proceedings of WebNet ’96, 1996.Maurer, H (ed.). Charlottesville, VA,AACE, p. 109–116. [see also WWWdocument: http://www.to.utwente.nl/ism/ism1-96/bookctr/san-fran/title_1.htm]

5 Collis, B, Parisi, D, Ligorio, B.Adaptation of courses for trans-Euro-pean tele-learning. Journal of Com-puter Assisted Learning, 12, 47–62,1996.

6 Eekma, A J. Defining guidelines foran individual learning environment.Graduate thesis, Faculty of Educa-

tional Science and Technology, Uni-versity of Twente, The Netherlands,1996.

7 Eekma, A J. Guidelines for educa-tionally oriented WWW sites forcross-cultural use. Internal report,Faculty of Educational Science andTechnology, University of Twente,The Netherlands, 1996.

8 Gastkemper, F H D. Pedagogy. In:Open and distance learning : criticalsuccess factors. Davies, G, Tinsley,D (eds.). Proceedings InternationalConference, Geneva, 10–12 October1994. Berne, FIM, Erlangen, 1994,19–21.

9 Lynch, P J. Yale C/AIM WWW stylemanual. Yale Center for AdvancedInstructional Media. [WWW docu-ment] URL: http:// info.med.yale.edu/caim/StyleManual_Top.HTML, 1995.

10 Minyoung, S. Hypermedia. Seoul,Korea, Seoul National University,Department of Computer Engineer-ing, 1995. [WWW document] URL:http://mmlab.snu.ac.kr/course/mm-seminar/hyperdoc/ node1.html.

11 Nielsen, J. Multimedia & hypertext :the Internet and beyond. London,Academic Press, 1995.

12 Preece, J et al. Human-computerinteraction. Wokingham, Addison-Wesley, 1994.

13 Taukojärvi, S, Järvilehto, P. In: TheTeleScopia courses : experienceswith the adaptation process fortrans-European tele-learning. Collis,B (ed.). TeleScopia DeliverableUT/DL1001/WG 3.3. Bonn, Deut-sche Telekom AG, Generaldirektion,1995, 43–51.

This paper focuses on aspects of col-laborative virtual environments (CVE)or Televirtuality. The first part will berelated to interaction issues and theresults of a field trial based on Tele-virtuality. The next part of this paperwill focus on technical aspects of ageneric framework for such systems,and how these utilise the network andthe technical platforms in an optimalway.

Part I

This part deals with the subject in theperspective of other conferencing sys-tems. It pays special attention to Vir-tual Reality (VR) as a medium andhow it can be used to support socialinteraction and collaboration betweenpersons located in remote areas. A net-worked application, based on the ideaof city planning (The Little Houses onthe Cyber Prairie), has been used in atrial consisting of several VR stationslinked in a local area network. In eachgroup three persons have entered thecity, confronted with some tasks whichthey had to solve in collaboration witheach other. This paper reveals some ofthe findings and experiences the testusers had. The experiences so farseems promising in using this newmedium in distance education andremote work.

Part II

This part gives an introduction toDOVRE1, which has been used tobuild the application for city planning.This system makes it possible for sev-eral users to enter a shared virtualspace and interact with enhanced func-tionality. Then it describes briefly theDovre framework, what design goals itis based on and some of the technicalsolutions we used when implementinga first version of the framework.

Part I

1 Introduction

The last 20 years the telecommunicationevolution has found its form in a hugerange of application areas and services.This evolution has coexisted with theinformation revolution and the mediamerger between telecommunication tech-nology, computer technology and contentproviders. We have seen the birth ofLAN, WAN, connected throughmodems, leased lines, ISDN, ATM. Thewide spread of IP-based services like theInternet, email and The WorldWideWeb.We can use services like videoconferenc-ing and multimedia conferences. Thepersonal computer in networks seem tobe the winning medium as its perform-ance and rendering capabilities seem todouble every two years. This is the back-ground for the next generation of com-munication medium; the networked vir-tual reality. The advantages seem quiteobvious, as it can overcome the obstaclesof traditional remote telephone confer-ence meetings, videoconferences, multi-media conferences and computer confer-ences. These obstacles are related to limi-tations in the various media and first ofall the ability to make the participantsfully engaged in the conferences. To feeland sense objects, be able to determineits own viewpoint, appearance and utilisea variety of functionality seems to bewhat we can expect of a future telecon-ferencing system. The alienating natureof many traditional media can be over-come in virtual reality conferencing. Theparticipants can experience a new kind ofpresence with each other and with thethree dimensional objects and documentsthey collaborate on, as they are present inthe same virtual room.

2 Field trial: Interaction in

The Cyber Prairie

The initiation of the networked trial wasthat we wanted to know more about thefollowing topics

• Interface, how do the participantsexperience the use of HMDs2, and joy-sticks in the virtual world

• Experiences and views on usingavatars3

• Co-operation, how do the participantsexperience being in the same virtualenvironment and work together tosolve tasks

• Views on networked VR as a mediumin education

• Views and expectations on networkedVR as medium.

About the applicationBased on the platform development (seesecond part of the paper) an applicationscenario was developed. The idea was togive the users the possibility to co-oper-ate on a city planning process in a dis-tributed environment. The virtual worldconsisted of a piece of virtual landscapewith several objects like houses, a hospi-tal, a ruin (historical), streets, a gas sta-tion, etc. The objects were scatteredaround in disorder, and the objective forthe participants was to put the objectsinto an order of a city. The system wasalso created in a way so that it wouldrequire co-operation. Before entering thevirtual environment, the participantswere given a list of tasks that they shouldtry to perform.

AvatarsThe three users were represented in the3D environment with so-called avatars orVirtual Identities (VID) (see Figure 1).The use of avatars is crucial in collabora-tive virtual environments, as they repre-sent the point of view of each participant.The avatars used in the Cyber Prairiewere designed as humanoids, but theycould have been without resemblance tohumans. To a certain extent we were alsointerested in examining the effect ofusing avatars, and how they were com-prehended.

InterfaceIt was possible for the users to movetheir avatars and their subjective point ofview with 6 degrees of freedom in space.The application did not use collisiondetection to objects or surface. The maininput device was a joystick for directionand movement along the X- and Z-axis,equipped with buttons for object (house)selection and rotation in the Y-axis.

51

Learning in collaborative virtual environments – Impressions from a trial using the Dovre framework

B Y O L A Ø D E G Å R D A N D K A R L A N D E R S Ø Y G A R D


1 Distributed Object oriented VirtualReality Environment. Incidentally,Dovre is also the name of a muchcherished mountain range in Norway.

3 Avatar, from Hindu; half god, halfhuman.

2 Head Mounted Display with head-tracking.

2.1 Technical configuration andmethod

Each of the three VR stations consistedof a Pentium 200 with a Head MountedDisplay4 with head tracking and a simplejoystick. The application Cyber Prairieperformed in a frame rate around 10frames per second in 640 by 400 pixelresolution. The three users were placed inseparate rooms. Each VR station ran aclient application, and they wereconnected through a Local Area Networkto a server which ran on a Silicon Graph-ics workstation5. This server also ran aclient application. This made it possiblefor the researchers to observe the interac-tion in the virtual world. The VR stationswere equipped with microphones whichwere connected to a mixer that sentsound back to the HMDs.

In this way, the researchers could followthe interaction both by choosing theirown point of view and listening to theconversation between the test persons.The idea was not to interrupt the testusers unnecessarily, and to let them learnby doing. On a couple of occasions oneof the client applications crashed, or thetest person got lost in the virtual world.Then it was easy for us to help whenneeded.

In addition to the trial a questionnairewas developed6. The test persons weregiven a hands-on experience using theVR equipment. Then they were intro-duced to some tasks which they should

attempt to solve. These tasks were relatedto environment issues, like where to putthe hospital and where to put the villas.

The background for the test was based onthe following assumption:

It is easier to collaborate and learn innetworked virtual environments thanin other teleconferencing environ-ments.

We wanted to acquire experiences withthe use of avatars and the possibility tochoose their own point of view. Wewanted to observe the shared awarenessof the participants. Compared to tradi-tional videoconferencing or multimediaconferencing some difficulties have beenrecorded in sharing objects and knowingwho is doing and seeing what. Wewanted to observe if these problems weremore easily overcome in virtual environ-ments.


4 I-O Glasses from Virtual I-0.5 SGI Indigo2 Maximum Impact.

6 The questionnaire and methodologicalapproach were developed in co-opera-tion with Marianne Skogerbø, who isusing this material for her masterthesis in Pedagogics at The Universityof Oslo.

Figure 1 Avatars on the Cyber Prairie

2.2 Interface, avatars and co-operation

The tests were carried out with threegroups of three persons each. Two of thegroups were high school students around16 to 17 years of age. The last group con-sisted of three researchers in the thirties.After the test a structured interview wasconducted with each group.

Interface:In general, the youngsters had more ex-perience in using the joystick. None ofthe test persons had any significant ex-perience in moving in virtual space orusing an HMD. Here are some quotesabout the interface:

“I had no control of the joystick, thehouses just flew back and forth, then Ihad to learn how to use the joystick.”

“In the beginning the head trackingdid not work, but later it did, and thenit became much more alive.”

“I thought we just should watch it inthe head mount. I did not know that wewere able to go around and dothings.”

“Some limitations on how and whereto move had made it easier.”

Each group spent around sixty minutesinside the virtual environment. Allgroups used around half the time to fig-ure out how to use the joystick and theHMD. There were two ways to choosedirection of the movements, either byusing the joystick, or by using the HMDby turning their head in the desired direc-tion.

Using an avatar:The avatars had various colours whichwere the only identificator, or differencebetween them. They were static in thesense of body limb movement. Theycould, however, see from the position ofthe avatar which direction they werelooking. Here are some quotes:

“It was no problem using an avatar;however, you did not see yourself, justthe other avatars.”

“Even though I was flying around, Idid not feel that I had any differentbody.”

One observation we did was that none ofthe groups had any difficulties in usingavatars. To most of them this was just anatural extension of themselves in thevirtual environment. Some of them


Figure 2 Three VR stations configuration

expressed a special kind of freedom invirtual space:

“I felt as if I were identical to myavatar, but I felt a bit handicapped notbeing able to use my hands.”

“This is a real experience, different toa movie, because you are in themovie.”

“It would be real nice to travel intospace with your friends.”

Co-operation:The next thing to do for the participantsafter getting used to the joystick andHMD, was to figure out how to selectobjects (houses) and to move them. Asthey had managed this, the next thingwould be to move them to desired posi-tion. The participants had to discussamong themselves to come up with anagreement on the desired position ordesired plan of the city. There was nocity plan that was “right” or “wrong”.The main idea was that they should co-

operate and organise themselves in theplanning and to put the plan into (virtual)life. Here are some quotes:

“In the beginning I had quite enoughmoving and orienting, but after a whilewe started co-operating.”

“We did not have any leader, but thatworked fine.”

“It is easier to work face to face, but ifI had this system at home for onemonth, it would have been easier.”

“I experienced it more intense to co-operate in VR than in real life.Because you were there all the time.”

“More fun and more interesting thanactual co-operation!”

“It was really important for co-opera-tion that we could both see the avatarsand speak to each other.”

Reflections on VR as a medium:“This is a different experience; I havebeen some place else, I have escapedfrom the normal world.”

“You have a TV screen in front of youreyes, and you move into these pictures,there is nothing dangerous there, youcan be driven over by a car andnothing happens, go through walls,go underground and into space.”

“We were in for 60 minutes, but thetime just disappeared.”

“Good for emergency meetings.”

Reflections on VR in schools:“Go to school in VR? Then therewould be a lot of dropouts!”

“In schools first of all, the pupilsshould have their own PC, then thebooks should be on CD-ROM and onthe Internet, then they would be lessexpensive and more up-to-date.”


Figure 3 The Cyber Prairie

“The teacher in future VR schools willbe the little avatar that flies around”

“I sometimes felt that it was the com-puter that had control over me, I hopethis is not how the school of the futurewill be.”

3 Conclusions on Part I

It became apparent for us that all testusers had no problem in enjoying thisexperiment. All three test groups man-aged during the period of sixty minutesto agree on some plan for the city, and tosome extent to realise this plan. Here weobserved that all groups were good inorganising themselves. In order to get thebest result, all groups had to place one ofthe avatars above the city (birds eye per-spective). This user had to guide theother users as they moved the houses.Most of the groups took turns in perform-ing the various task. One of them alwaystook on this leadership. We were also abit astonished by how naturally they re-lated to using avatars. All participants gotused to the joystick and HMD, but theyexpressed the need for a better 3D inputdevice, like a glove for grasping. Afterspending sixty minutes using an HMD,we expected that some of them mightfeel sick, but this proved not to be thecase. Some of them of course felt a bitdizzy. The VR application Cyber Prairiewas just a research prototype, and couldhave been improved in lots of ways tomake interaction easier, but it still provedpromising for this kind of collaboration.The three groups gave many similarviews on the experience. The greatestdifference was between the youngstersand the researchers. The researchersseemed much more analytic in theirapproach to the tasks they should per-form, but they were not so clever at per-forming them. The youngsters were morespontaneous and explorative in theirapproach to the virtual worlds. In this testwe observed two kinds of learning; firstthey learned how to move and orientthemselves in the environment, then theylearned how to collaborate on specifictasks in a virtual environment. Theresults of this trial have given us moreconfidence in Collaborative Virtual En-vironments as a useful way to learn andinteract.

Part II

4 Design issues

4.1 Rationale

When we set out to write our first appli-cations that would demonstrate some ofthe principles of Televirtuality, we soonfound that there existed no frameworksfor developing distributed Virtual Realityapplications on low end computer equip-ment. While solutions such as DIVEexisted for Silicon Graphics class ma-chines, we thought these unsuitable forconsumer Windows PCs. Also, manyVirtual Reality application frameworkswere strongly biased towards visualisa-tion of the virtual worlds, neglectingbasic features such as interaction with thevirtual world itself.

In the end, we decided that in order toeasily implement the application wewanted, we needed to develop a newframework. To this purport, we startedwork on Dovre.

4.2 Design goals

When we designed the Dovre framework,we had the following goals in mind:

Active objectsDuring the design phase of Dovre, wetook much inspiration from the realworld. Most notably, this has resulted inthe introduction of active objects. Ratherthan being passive objects that must bemanipulated by higher level algorithms,Dovre objects are active and have theirown behaviour and state. The activeobject paradigm is supported by astrongly object oriented line of thoughtand greatly simplifies tasks such asimplementing agents with artificial intel-ligence.

Biased towards ATM and TCP/IPOne of our main goals was that Dovreshould utilise the ATM network technol-ogy to its fullest. Since ATM is in-herently connection oriented, this effect-ively excludes connection-less protocols.Standards for multicasting over ATM arestill rudimentary, so the initial imple-mentation was based on a symmetricalclient/server model. Future versions ofDovre, however, may be extended to usemulticasting in a server-less environ-ment.

The TCP/IP protocol is a standard com-ponent of nearly all networking capablesystems and works with most networklayers, so we chose to base the initial


Figure 4 Features of the Dovre framework

implementation on TCP/IP. Note that theI/O system of Dovre has been completelyabstracted, so that implementing supportfor other network protocols is completelyinvisible to applications and most of theframework itself.

High degree of abstractionIn order to construct a system that isarchitecture and operating system inde-pendent, a high degree of abstraction isrequired. In the Dovre framework wehave used abstraction to hide operatingsystem, rendering engine and sound sys-tem details from the application.

The current implementation supports thefollowing operating systems and render-ing engines7:

• Windows NT 4.0 and 95/OpenGL• Windows NT 4.0 and 95/RenderWare8

• IRIX 6.x/OpenGL• Linux 2.x/Mesa9.

Sound is supported on all platforms.

The framework may be used withoutvisualisation and sound support, whichcan be highly useful for generating auto-mated agents. Such agents are entirelycomputer driven and act only on interac-tion or triggers from other participants,and have no use for visualisation ofsound locally, but may otherwise use allthe functionality that the frameworkoffers.

Support advanced sensors andinteractionIn order to support interaction with thevirtual world, Dovre supports sensorssuch as gloves, motion trackers, joy-sticks, etc. The sensors may be used fornavigating, triggering actions and gener-ally interfacing with the computer.

Sensors used in Virtual Reality applica-tions exhibit different behaviours andhave their own output formats. While amouse gives merely two degrees of free-dom, with no point of reference, a Polhe-

mus Fasttrak allows for six degrees offreedom in a fixed space. In order to sup-port the various sensors in a generic man-ner, a scheme for querying the sensorsfor their capabilities has been implemented.

Currently, Dovre supports the followingsensors:

• Mouse• Joystick• I/O glasses• 5th glove• Polhemus Fasttrak.

Dovre offers full object and ray intersec-tion detection in the object hierarchy.

EfficiencyTo maximise throughput and to constructa system that scales well, we have goneto great lengths to streamline and finetune Dovre. The framework is based onthe hierarchical world model, whichallows for many optimisations. We useour own highly optimised math libraryand advanced template mechanisms that,when inlined properly, generates codethat rivals that of hand coded assembly(see for example [8]).

The OpenGL implementation uses manyof the most common optimisation tech-niques in order to achieve high framerates, as well as special features of moreexpensive SGI hardware, such as anti-aliasing (multisampling) and real timeshadows.

With high end computers with multipleprocessors becoming cheaper, it is alsoimportant to consider how performancemay be increased by the parallelisation ofaccesses to the hierarchy. For example onhigh end computers, much of the render-ing engine has been implemented inhardware, which may execute in parallelwith the main CPU. On these systems,parallelisation is not only desirable butimperative in order to get reasonable per-formance.

5 Object organisation

The Dovre framework is strongly basedon the hierarchical model. This allowsfor many optimisations, and is usedextensively for scene culling, fast inter-section detection, efficient structuremanipulation, etc.

The world is partitioned in domains,which are separate hierarchies built up of

objects and containers. Domains may beconnected with portals, allowing for verylarge worlds.

All hosts (i.e. both clients and servers10)have at least one domain. This domain isentirely built up of real objects, objectswhich are local and belong to that host.All objects that are imported from otherhosts become virtual objects, objectswhich are mirrored from another host.

When two hosts connect, they willattempt to exchange object hierarchieswith each other. This is first done by thehosts exchanging any portals they mighthave with each other. If the hosts haveportals which may connect, they willthen proceed with distributing the hier-archies of those portals to each other.Eventually, all hosts in the network willhave a local copy of the part of the hier-archy that is relevant to that particularhost. All these imported objects are vir-tual objects. Any changes to the hier-archy or its objects will be distributed toall hosts that need to know about it.

All this is automatically done by Dovre,and is completely transparent to the pro-grammer.

5.1 Real objects

Real objects are, as the name implies, thereal objects in the hierarchy. They mayreceive messages and act upon them. Ifthe real object chooses to act on arequest, it reissues the request as a com-mand to itself, which is distributed to allits virtual counterparts on other hosts.This implies that both the real objectsand all its virtual objects have the samestate, and integrity is ensured.

5.2 Virtual objects

Virtual objects are mirrors of realobjects. They may receive requests, butmay not themselves take on any action,rather they must pass them on to its realobjects. Virtual objects receive com-mands from their real object which theymust then do their best to fulfil.


10The only thing that differs between aclient and a server is that a server canaccept connections from severalclients, while a client connects to oneserver. Note that a client can operatewithout a server.

7 Dovre may be ported relatively easy toany operating system which supportsnetworking and threads.

8 RenderWare is supported because ofits higher performance on inexpensivePCs.

9 Mesa is a freeware rendering enginewith an API which closely matches theOpenGL 1.1 specification.

6 The message system

The Dovre framework is an event-drivensystem, in that all interaction betweenobjects must be done with messages.Indeed, the very core of Dovre is themessage system. The message queue isbasically a distributed queue with mess-ages that can be updated if they losescope, not entirely unlike the updatablequeue described in [6]. The message sys-tem may change, however, with theadvent of MPEG4.

Messages may be sent directly to thereceiving object, for time critical pur-poses, effectively bypassing the messagesystem, or it may be queued for deliverywhen the system has time, optionallywith a time delay.

6.1 Messages

Any real object may issue a message toanother object. The message will then bedelivered to the real object. The receivingobject may choose to act on the messageor ignore it. Note that the receivingobject may not update internal stateaccording to the message, as this wouldintroduce inconsistencies in the hierar-chy. This has to be done with commands.

6.2 Commands

Commands are used for updating internalstate in an object. When a real objectneeds to update internal state, it issues acommand, which is distributed to all itsvirtual objects, as well as itself. In thisway, the hierarchies of all hosts are up-dated simultaneously, and consistency isensured. Note that only real objects areallowed to issue commands.

7 Object types

The following is a short description ofthe basic object types in Dovre. Theseobject types are the building blocks forvirtual worlds, and may be subclassedand overloaded to get new behaviour.

• NodeThe Node class is an abstract baseclass, from which all objects that canbe put in the hierarchy have been sub-classed. It simply implements thecapability of being a child (i.e. havinga parent).

• ContainerThe Container node can contain anynumber of nodes, and may be used forgrouping objects and optimising thehierarchy.

• Coordinate_SystemIn order to specify spatial position, co-ordinate system containers may beused. A Coordinate_System has atransformation matrix which trans-forms the co-ordinate system of all itschildren.

• ElementElements are the visual objects in theobject hierarchy. These have a geo-metrical model, which can be generat-ed dynamically or loaded from file.File formats currently supportedinclude DXF, NFF, 3DMF and somemore obscure formats. VRML supportis planned.

• LightsourceLightsources provide light in thescene. Lightsources may be direc-tional, positional or spotlights.

• BillboardBillboards are containers that alwaysturn to face the viewpoint. This may beuseful for implementing menus andsimilar. Note that billboards can bepretty non-deterministic to work with;they are relative to the viewpoint, which

may give some surprises when usingintersection detection and similar.

• ViewpointViewpoints implement a means oflooking at object hierarchies. By bind-ing a viewpoint to a window, the scenecan be rendered real time on thescreen.

• Sound_SourceA sound source emits sound whichmay be heard by listeners within thedomain that the sound source is locat-ed in. Sound sources may read sampledata from a file, or may receive itdirectly from a microphone. They mayalso have different characteristics asregards sample rate, number of bits,etc. In the future we plan to implementadvanced features such as customis-able filtering of sound sources (forexample flanging, chorus or distor-tion).

• ListenerListeners hear sound sources which arelocated in the same domain as thelistener itself. Sound is spatialisedaccording to the relative positions ofthe sound sources.

• MirrorMirrors make simple reflections ofobjects located perpendicular to themirror. It actually only redraws thescene with a new transformationmatrix that has the Z axis inverted.Mirrors should be used with care, asthey effectively double the number ofelements that need to be drawn in thescene.

• PortalPortals are used to connect differenthierarchies with each other. This isuseful for a number of reasons. Twoportals that can connect may be locat-ed on different hosts and will notrendezvous before a connection be-tween the hosts has been made. Thisallows for distributed worlds.


Node

Listener Sound_Source Lightsource Element Portal Mirror Container

Billboard Coordinate_system

Viewpoint

Figure 5 The class hierarchy of Dovre

Portals also allow splitting a very largeworld into smaller domains, which willimprove performance considerably.During rendering of a scene, the portalitself is tested for intersection with theviewing frustum, similar to the Poten-tially Visible Set technique describedin [7].

8 Future development

Dovre is a platform for constant researchand development. Some of the mostimportant features that we are currentlyworking with include:

8.1 Inverse kinematics

In order to simulate the real world asclosely as possible, natural and realisticmovement of avatars is necessary.Inverse kinematics provides a solution tothis problem, but is computationallyintensive and currently difficult to use.We are undertaking research on how tooptimise such algorithms, so they may beapplied in real time simulation.

8.2 Dynamic models andMPEG4

Currently, Dovre does not have the capa-bility of coding models and distributingthem to other hosts. Work is being donethat will change this. We will proceedwith implementing support for audio andvideo streams, most likely through supp-ort for MPEG4.

8.3 Graceful degradation ofnetwork services

With ATM comes a wealth of new possi-bilities as regards controlling the qualityof the network services provided. Inorder to handle the highly variant natureof the data streams associated withDovre, we need to support QoS manage-ment and graceful degradation of net-work services. The current implementa-tion does not do this, but we are workingactively with supporting it. The MPEG4SNHC11 group is performing research inthis field, which we hope to put to use inDovre.

8.4 Java

Java is making a great impact on thecomputer world, and is particularly inter-esting in the context of active objects.We are looking into how Java amongstother things could be used for agentswith programmable intelligence.

9 Conclusion on Part II

So far, we have had very good experi-ences with Dovre. The framework pro-vides us with a platform for further Tele-virtuality research and a quick and sim-ple way of implementing new conceptsand ideas for distributed virtual realityapplications.

Performance is excellent, and Dovre hasno problems handling relatively largeworlds. We have more ideas for optimis-ations that should increase performanceeven further, and we will continue addingsupport for new technology.

The object oriented, active object pointof view provides a good basis for deve-loping distributed virtual reality applica-tions. Since objects have their own be-haviour and state, they can virtually acton their own. In the future we will con-tinue to explore how the active objectparadigm can be used to bring virtualenvironments to life and to implementintelligent user interfaces.

References

1 Bowers, J, Pycook, J, O’Brien, J.Talk and embodyment in collabora-tive virtual environments. In:Proceedings of CHI’96, New York,ACM Press, 1996.

2 Bowers, J, Pycook, J, O’Brien, J.Practically accomplishing immer-sion : cooperation in and for virtualenvironments. In: Proceedings ofCSCW’96, New York, ACM Press,1996.

3 Pycook, J, Bowers, J. Getting othersto get it right : an ethnography ofdesign work in the fashion industry.In: Proceedings of CSCW’96, NewYork, ACM Press, 1996.

4 Søby, M. Cyborg and the prosthesis :about socialisation in Cyberspace.Paper to the 5th National Conferenceon Pedagogic, Stavanger Nov. 1996.

5 Ødegård, O. Social interaction intelevirtuality. Paper for Virtual Real-ity World ’96, Stuttgart, Germany,13–15 February 1996.http://televr.fou.telenor.no/papers.html

6 Kessler, D G, Hodges, L F. A net-work communication protocol fordistributed virtual environment sys-tems. 1996. Available atftp://ftp.gvu.gatech.edu/pub/gvu/tr/96-03.ps.Z.

7 Luebke, D P, Georges, C. Portals andmirrors : simple, fast evaluation ofpotentially visible sets. Proceedingsof the 1995 Symposium on Interact-ive 3-D Graphics, 1996. Available athttp://www.cs.unc.edu/˜luebke/publications/portals.html.

8 Veldhuizen, T. Expression templates.C++ Report, 7, (5),26–31, 1995.Available at http://monet.uwaterloo.ca/˜tveldhui/papers/Expression-Templates/exprtmpl.html.


11Moving Pictures Experts Group, Syn-thetic and Natural Hybrid Coding.

1 Introduction

Over the last decades we have witnesseda very rapid diffusion of telecommunica-tion and computer technology in privatehomes as well as in the education andbusiness sectors. These new communica-tion tools have radically changed thescope and potential for distance educa-tion, and they are gradually giving newmeaning to the concept itself. Still, thereis reason to believe that we are onlybeginning to see a new generation of dis-tance education, based on advancedbroadband communication technologyand digital networks.1

Telenor Research and Development hasclosely supervised this development, andthe first comprehensive survey on the useof telebased distance education in Nor-way was launched in the late eighties[24]. A field study of higher educationand universities was followed by a simi-lar study of manufacturing industry [10].Five years later the two surveys wererepeated, using the same questionnaires(but not the same sample) [12], [13].

These broad field studies cover the rapiddiffusion of new telebased distance edu-cation within the crucial part of societyconcerned with education, teaching andtraining. In retrospect, this gives us aninteresting view of the rapid changeswithin this field. The object of thesestudies, however, was not restricted torecording only the number of students orteachers using new technology for educa-tional purposes. In addition, the researchintended to explore the future needs fordistance education, both in a latent andmore overt sense. For this reason the sur-veys deploy a broad view of the conceptof distance education, including all activ-ities where teacher and student are sepa-

rated in time and/or space.2 We alsogathered information in fields whichcould tell us about the possible long-termneeds for telebased education. For exam-ple, the surveys of manufacturing indus-try cover all kinds of training and educa-tion within the enterprises, and the sur-veys of the education sector cover the useof external teachers and lecturers in tradi-tional education.

This article presents some main resultsfrom the quantitative user-need researchat Telenor Research and Development,focusing on the changing patterns of dis-tance education over the five year periodfrom 1989 to 1994. Manufacturing indus-try and the education sector, includingupper secondary schools, regional col-leges and the universities, will here betreated in separate chapters. One of thereasons for this is, as will appear fromthe discussion, that the areas are quitedifferent regarding the way distance edu-cation is initiated, managed and imple-mented. Each of the chapters will, how-ever, report both on the current situationand the future needs. Finally, I will pre-sent some general conclusions and rec-ommendations on further research anddevelopment within the field.

2 Distance education in

manufacturing industry

2.1 Introduction

A central idea in Norwegian educationalpolicy has been the principle of “lifelonglearning”, where knowledge and compe-tence is maintained, renewed and extend-ed throughout a lifetime [22]. The abilityto build personal knowledge is viewed asa common good which as many as possi-ble should have access to. Education andknowledge open the doors to individualdevelopment, and are an importantpremise behind a living democracy. Inthis context, education and skills is afounding element in the Norwegian ideal“welfare society”. At the same time,competence and knowledge represent afactor of major importance for the econ-omic development of the country. Duringthe last 10–15 years, the value of educa-tion and training has been emphasised in

official documents, focusing on thenational need for a more competitiveindustry.3 In the White Paper “Knowl-edge for all” (“Mer kunnskap til flere”)this was made very clear:

“To develop knowledge is a precon-dition for strengthening Norwegianeconomy, ensuring full employment,achieving effective changes and pre-paring the way for new activities in theeconomic life. (...) in the economicsense education is of strategic import-ance for our value creation.” [14]

The quotation reflects a central theme inthe discussion of education for new skillswithin the industry: To elevate the levelof competence of employees represents amajor strategic weapon to meet new,international competition. At stake hereis not only the question of strengtheningthe local labour markets. Inability to copewith the challenge of upgrading skills inthe industry will threaten the competi-tiveness in the future information society.‘Knowledge’ is the keyword for every-one who wants to take part in the newinformation-intensive global economy.

The dynamics behind the growingimportance of skills and competencewithin the labour market are of coursemany-faceted and complex. Neverthe-less, it is possible to emphasise at leastsome central factors: First, the fast intro-duction of new technology into thelabour market has raised new demandsfor knowledge and skills. The extendeduse of information technology in almostevery sector of the industry, has on theone hand radically changed the workoperations involved, demanding newskills to make an optimal use of the tech-nology. On the other hand, the deploy-ment of new technology has in some

59

Distance education in Norwegian manufacturing industryand the education sector – Current situation and future needs

B Y T O M E R I K J U L S R U D


2 The definition of distance educationused in this article is as cited in thearticle “Telecommunications in dis-tance education” in this issue of Telek-tronikk.

3 This is true on a European level aswell. A report from the EuropeanCommittee for Research and Develop-ment (IRDAC) states that; “the outputof education and training systems (...)in terms of both quantity and quality ofskills at all levels, is the prime determ-inant of a country’s level of industrialproductivity and hence competetive-ness. (...) If sufficient attention is notgiven to the skill shortage problem, inparticular technological advance,Europe’s competitive position will bethreatened.” [9]

See also [4], [21] and [28] for similarargumentation.

1 It has been discussed if this technolog-ical change is best characterized as a“paradigm shift” or a “shift in gener-ations”. Some scholars have presentedthis development in different phases(generations), from traditional letter-based education to education throughcomputer networks. Rekkedal and Søby[25] have remarked that it may bemore correct to view these chategoriesas a pool of technologies and methodsto choose from when organizing dis-tance education.

cases substituted for jobs and tasks,resulting in a need for re-educatingemployees.

Second, there has been a growing glob-alisation of trade and commerce activitiesin the last 10–15 years. One effect hasbeen that major parts of traditional indus-try have been re-located to countries withlower wages and expenses. Technicaland industrial co-operation has expandedbetween nations, together with the rise ofmore global markets. For Norway andother European countries, this has fuelledan engagement in more knowledge-basedindustry, and more specialised produc-tion. One of the consequences has been ageneral expansion of jobs generated frommarket and non-market services such asfinance, banking, tourism, consultancy,etc., at the expense of manufacturing andproducts.

Third, it should also be mentioned thatthe amount of knowledge is growingvery fast, and still more efforts are need-ed to keep informed. Today, a consider-able number of those employed are en-gaged in the production of new informa-tion, and even more people are busy col-lecting and structuring this. It is estimat-ed that the total amount of information inthe country is doubled every 10 years[23]. One effect is that constantly moreefforts and skills are required to stay wellinformed.

Thus, the process of strengthening thecompetence of employees has beenstrongly advocated by representatives ofthe state and the government. Thereseems to be an urgent need for industryto redevelop and upgrade the internallevel of competence. In practice, how-ever, it is local industry who will have toinvest money and time in new compe-tence, and there are several barriers toovercome to manage this.

2.2 The implementation ofdistance education

The training and the development com-petence at the workplace include a lot ofdifferent activities. In a very generalway, the concept of competence refers to“knowledge and skills which are used ormay be used in professional activities”4

[9]. This is the way we will make use ofthe concept here. In a more extendedsense, the concept may refer to a collec-tive body of knowledge, reflected in anorganisation or a social system.

The development of competence in anorganisation may be described as a pro-cess of both planning, implementationand exploitation [20]. Various internal orexternal changes may trigger a need forhigher education or development ofskills. This may come as a decision fromthe management, but also as a demandfrom the employees. The experiencedneed for new competence in the organis-ation may be met with new employmentor development of the skills internal tothe organisation, which in turn may beimplemented and exploited. This is whatmay be called “the chain of competence”.(Ibid).

The development of competence in theorganisation includes several possiblechoices, such as investing in more re-search, or trying to strengthen the workethic or organisational culture. Of specialinterest to the use of distance education,however, is the more traditional formaltraining of the personnel. This training iseither organised by the organisation itself,or by external institutions. The former ismore common in larger enterprises orofficial organisations, while small andmedium sized enterprises mostly useexternally organised courses or seminarsin co-operation with universities or othereducational institutions.5 [19]

In this context, tele-based distance edu-cation represents expanded opportunitiesfor enterprises to develop their compe-tence, and more specific, the skills andknowledge of the individual employee. Itcomes as a result of the same needs asordinary education and training (see Fig-ure 1). The difference is that the teachersare seated at a distance from the students,and different communication media areused to support the learning. Althoughthe most common organisational form isto use technology to bring the distantcompetence into a group of employees inthe organisation, it will also include morecomplex forms: Students may be locatedat different places, as well as the teachersand different sources of information usedfor instruction.

The advantage of distance educationcompared to regular face-to-face educa-tion is first of all that it may be more effi-cient and economic than traditional edu-cation. The amount of money spent ontransporting employees to and from theexternal locations is significant, and will,it is argued, soon compensate for the nec-essary investment in communicationtechnology. When this investment isdone, however, it will open up a widerrange of education possibilities thanbefore. For instance, implementing avideo conference studio will in principlegive access to every learning institutionand university with similar equipment.Moreover, the equipment can easily beused for other purposes, such as businessmeetings and video conferences.

The factors of special importance fordevelopment of distance education inindustry are therefore the access to thedesired knowledge (both in time andspace), and the economic situation of theenterprise.

2.3 Distance education in man-ufacturing industry

Let us then turn to the result of the re-search on the use of distance educationamong Norwegian firms in the manufac-turing industry. The survey was based ontelephone interviews with representativesfrom 795 enterprises, which provided arepresentative sample of about 7 % of thepopulation. The respondant was in mostcases the manager of the enterprise or theperson responsible for internal training.As mentioned in the introduction, the1994 survey succeeded a former study inthe field.


4 Original formulation: “... kunnskaperog ferdigheter som er i bruk eller somkan brukes i rollen som ansatt.”

5 There are certain differences betweeninternally and externally organizedtraining, with respect to motivation bythe organisation. While internal train-ing in most cases is triggered by cer-tain needs within the organization,external education appears on an openmarket. For this reason the latter maybe less dependent upon the needs andstrategies of the single enterprise ororganisation [8].

Distance education

External trainingInternal training

Formal training

A perceived need for new skills and knowledge

Figure 1 The relation between distance education and formal training

Firstly, the survey showed that there hasbeen a growing demand for distance edu-cation in the training programme of theenterprises. While in 1989, 7 % had useddistance education in their training pro-gramme, the survey showed that this hadincreased to 11 % five years later. Acc-ording to the general penetration of com-munication technologies in the industry,this does not indicate a very rapid diffu-sion of distance education (see [27]). Themajor explanation for this seems to be inthe general economic conditions duringthis period. This becomes more evidentwhen we compare the results to the gen-eral investment in training programmes,which in the same period was cut byalmost 20 % (see Figure 2). In relation tothis general reduction, the share of dis-tance education in training programmeshas actually increased by 9 %. An inter-esting supplementary finding was that thenumber who reported that they “did notknow” if their enterprise used distanceeducation, had dropped from 16 % to 1,suggesting that general knowledge of thetopic had been strengthened.

Most of the distance education consistedof small-size projects, involving no morethan 1 – 5 employees. There were, how-ever, some large projects pushing theaverage up to 11 employees per pro-gramme. Altogether, there were notmany long-lasting projects, and everysecond one had been started during thelast three years.

Secondly, the survey showed that therewere certain differences among the firms,both in respect to number of employeesand line of business.6 On average, thelarge enterprises more frequently madeuse of distance education in recent years,than the smaller. In enterprises with morethan 150 employees, 25 % of the

employees had used distance educationin the last years, but for the enterpriseswith less than 30 employees, the corre-sponding number was only 5 %. Thesefindings are in accordance with theresults from the 1989 survey, and theyreflect a pattern known from similar re-search, reporting that the smaller enter-prises have problems giving their staff asmuch training as the larger ones (see [2]).

Furthermore, the results showed that dis-tance education was more common in thesectors of “chemical production”, follow-ed by “basic metals” and “food, bever-ages and tobacco”. The high number ofdistance education in the latter two ispartially explained by the many bigenterprises in these sectors. “Chemicalproduction” and “food, beverages andtobacco” were also characterised by highrates of regular training, as were “Otherindustries” (see Figure 3).

What kind of communication technolo-gies did the enterprises use in their dis-tance education programmes? Letter/postservices was the most frequently usedtechnology, mentioned by nearly 70 % ofthe respondents, followed by ordinarytelephone, used by 20 %, and electronicmail / computer conferences and videocassettes used by 15 %. Hence, the tradi-tional ways of carrying out distance edu-cation were still dominant, although

computer based communication (CMC)played an increasingly important role.The result also confirmed earlier find-ings, showing that ordinary telephony isan important means for consultations be-tween teacher and students (see [3]). Thetelephone service was used to an equalextent by small and medium sized enter-prises, as by larger. The number of avail-able media technologies in general in-creased with the number of employees.

2.4 Future needs for distanceeducation

There are several ways to estimate thefuture demands for distance education.The easiest way is simply to ask theinformants what they intend to do in thisfield in the coming years. On the ques-tion of their future preferences, 6 % of allthe enterprises answered that they wereplanning to launch a project next year.These were mostly large corporationswhich expressed interest, with a predom-inance of companies from “textiles” and“food, beverages & tobacco”. There was,however, a certain interest for startingdistance education in every industrialbranch, ranging from 3 to 10 %. Thiskind of self-reported judgement, how-ever, is not necessarily a very reliableindex: In the 1989 study we asked therespondants about their preferences forthe next years. Compared to the actual


6 The enterprises were analysed inaccordance with the Standard Indus-trial Classification used by the officialStatistics in Norway [31]. The classifi-cation of manufacturing industry con-sists of nine main activities (lines ofbusiness): Manufacture of food, bever-ages and tobacco; Manufacture of tex-tiles; Manufacture of wood and woodproducts; Manufacture of paper andpaper products; Manufacture of chem-icals; Manufacture of mineral prod-ucts; Manufacture of basic metals;Manufacture of fabricated metal prod-ucts, machinery & equipment; andOther manufacturing industries.

80

70

60

50

40

30

20

10

01989 1994

TrainingDistance educationPlans for next year

Figure 2 The extent of training programmes, distanceeducation and estimated future use of distance education in

the manufacturing industry, 1989 and 1994, in per cent

registered number of distance educationprogrammes in the later study, thisproved to be a rather poor indication.

A more indirect way of valuing the futureneed for these services, is to take a closerlook at the way the enterprises are struc-turing their regular training activities. Inthe introduction above, we made theassumption that both internal and extern-al circumstances were facilitating theneed for formal training programmes.Thus, a new market situation may triggerdecisions to change the products, toengage in new markets, to adopt newtechnologies in production, etc. Internaltraining is here one of many alternativeresponses. The strategic reason forimplementing distance education is that itoffers easy and fast access to the educa-tion providers with lower expenses, whena need for training already exists. Dis-tance education may therefore be viewedas a product of 1) high demand for train-ing, combined with 2) distance to theknowledge/competence, and 3) a needfor reduction in training budgets.

How is the outlook, with respect to thesethree conditions? Firstly, it appearedfrom our analysis that the general use ofinternal training in the industry is ratherunstable. As mentioned above, there wasa significant reduction in the trainingprogrammes in the period 1989 – 1994.This was in a period when the economic

situation in the country was worsening,resulting for instance in a reduction ofthe number of people in employment.7

This may be part of the explanationbehind the reduced efforts in training.

The reduction was most evident in thesmaller firms; many of the larger enter-prises had actually increased their formaltraining. For instance, 50 % of the small-est companies (less than 30 employees)did not have any training at all the pre-vious year, but all of the larger firms did.The number of enterprises where morethan 50 % of the staff had enjoyed atraining programme, had also increasedamong the largest firms, but decreasedamong the smaller ones. Hence, the train-ing programmes in the larger firms seemto be less vulnerable to economic cyclesthan in the smaller. Interestingly, dis-tance education seemed to hold itsground, despite the general cuts in inter-nal training, probably due to the fact thatmost of this activity was taking part in

larger enterprises. Still, these fluctuationsin the demand for formal training make itdifficult to predict the future use of dis-tance education.8

Secondly, it appeared that the use ofexternal teachers or lecturers was verycommon, and had increased over the fiveyear period. In 1989 every third enter-prise relied totally on their internal teach-ers, but five years later, this constitutedonly 8 %. In 1994 every second trainingprogramme was held by an external spe-cialist, and over 40 % used both internaland external teachers.9 In reality then,over 90 % of the enterprises had usedteachers from other institutions, or organ-isations in their courses. This finding wassupported when we considered the venuefor the courses. In 1994 the number ofenterprises who held their training out-side the company building had increasedfrom 31 % to almost 80 %. The majoritypreferred a combination of internal andexternal courses. Thus, it seems to be thecase that the use of external teachers andexternal courses is supplementing, andnot substituting, the regular (internal)training. The use of internal teachers orlecturers was more common among theenterprises with many employees, whilesmaller enterprises as a rule hired thenecessary competence (see Figure 4).This was evident in both the surveys. Theincrease of external organisers and tutorsseems to be the result of higher demandsfor training in the larger companies,


7 From 1989 to 1991 the number ofemployees in manufacturing industrydropped by 13,000 people [31]. Dur-ing the period 1986 to 1990 there wasa reduction of the total investments indevelopment of skills and knowledge inthe industry, even if the non-materialinvestments were kept relatively con-stant [6].

8 There are signs indicating that theinvestment in education has beenincreasing in the last year, at least insome lines of business. A recent surveyof European business and enterprisesshowed that the Norwegian privatesector was spending about 3.5 % oftheir salary expenses on competencedevelopment, only surpassed bySwedish, French and Swiss enter-prises. The results showed, however,that the local government in Norwayspent minimal amounts on educatingtheir employees, pulling Norway downin the international statistics (Aften-posten, 17.10.96).

9 There are some differences in the waythe registration was done in 1989 and1994, on this question. In the first one,the answers were registered as either“internal” or “external”. In the latterone was added a registration possibil-ity for “both”. This might have influ-enced the reliability of the data.

70

60

50

40

30

20

10

0Food, b&t Paper and

productsWood andproducts

Chemicalproducts

Basic metallproducts

Textiles Other industry

Training Distance education

Figure 3 The use of training programmes and distance education in manufacturingindustry and activity classification, 1994, in per cent

resulting in extended use of supplement-ary providers.

Summing up, this might point in thedirection of higher demands for distanceeducation, which in essence is a veryefficient way of getting access to know-how located outside the enterprise. Onthe other hand, there may be several rea-sons for bringing an external expert intothe organisation, and it is not easy to sayto what extent distance education cansubstitute for this. In a very similar way,there may be different reasons for lettingthe staff go to externally arrangedcourses, reasons that are not easily sub-stituted by distance education. Examplescould be: to get away from the everydayenvironment, have a new experience,enjoy a trip together with colleagues, orto get an award. To gain knowledgeabout these circumstances would requirefurther research of a more qualitativecharacter.

Thirdly, the enterprises reported havinglarge expenses on training activities.Most of the enterprises have separatebudgets dedicated for this, and the aver-age sum spent on training was almost5,000 NOK. The respondants who report-ed having no training budget were in90 % of the cases representing firms withless than 30 employees. Compared to thetotal budget, however, it was themedium-sized enterprises which spentthe largest sum on education. “Otherindustry” had the highest proportion oftheir total budget dedicated to formaltraining; about 4 % on average. When itcomes to expenditure on travelling, thegeographic location of the enterprisesproved to be a decisive variable. Indensely populated areas (Akershus, Oslo,Nord-Trøndelag) almost half of therespondents reported that they did notuse any of the training budget on travel-ling. In contrast, every second enterpriselocated in certain districts on the westcoast (Aust-Agder) and in the north(Finnmark) reported using more than halfof the total training budget on travellingand accommodation.

Hence, the budget on training and travel-ling interacts closely with the rate oftraining in general, as well as the numberof employees. The prediction value ofthis variable is therefore not very strong,and the analysis did not investigate theneed for cutting costs related to the train-ing. The analysis showed, however, thatthe use of distance education was morecommon among enterprises outside the

city areas, suggesting that geographicaldistance may be of importance. The lowbudget for training in smaller firms indi-cates that the economic situation is morecritical here than in larger firms, even ifthis has not resulted in more distanceeducation.

2.5 Summing up

The survey on the use of distance educa-tion in manufacturing industry showedthat there is a growing number of enter-prises which use distance education,despite a general decrease in formaltraining, during the period 1989 – 1994.Distance education is primarily used bymedium-sized and larger enterprises, par-ticularly within “chemical production”,“basic metals” and “food, beverages andtobacco”. There was also more distanceeducation among enterprises outside thecity areas.

Moreover, there appeared to be a certaininterest for more distance educationamong the respondants. The increaseduse of external education and externallyorganised courses, and the rapid diffu-sion of computer based communicationalso suggests a higher adoption of this inthe future. Still, it is difficult to predictthe development. Firstly, because thereare different needs accounting for the useof external competence, not all of themcan be supplemented by distance educa-

tion. Secondly, the training activities inthe enterprises seem to be very vulner-able to changes in external factors, par-ticularly among smaller and medium-sized enterprises (SME).

Thus, the political arguments for moreeducation and training in the industry arenot necessarily followed up by the indi-vidual corporation or enterprise. Thereduction of education programmes indi-cate, on the contrary, that more insecureeconomic situations make enterprisesreduce their training activities. The gen-eral investment in necessary technologyand equipment seems to be difficult tomanage, especially for smaller enter-prises, even if distance education in thelong run could help reduce the expensesof training. This may end up in a “viciouscircle” where the SMEs are closed offfrom sufficient development of trainingand education, while the larger corpora-tions enhance their lead.

3 The education sector

3.1 Introduction

The debate on the use of informationtechnology in the education sector hasbeen very intense in recent years. Thebackground for this is a general aware-ness and understanding of the prospec-tive importance of this technology inalmost every higher level study, occupa-


100

90

80

70

60

50

40

30

20

10

0<30 30-44 45-75 76-149 150<

Both External Internal

Figure 4 The use of internal and external lecturers in manufacturing industry, andnumber of employed, 1994, in per cent

tion or workplace. It appears to be of cru-cial importance to provide students onevery level with a good understanding oftechnology, and how it best can beexploited. Moreover, it has become evi-dent that information technology has apotential for making the education itselfbetter and more efficient. The very rapiddiffusion of the Internet in recent yearshas demonstrated the great advantagesthis may imply for students and teachers;the innumerable databases connected onthe network offer a unique opportunity toenrich and supply the education sector.

The discussion in Norway has also in-cluded a regional dimension. Since theeducation sector has a very decentralisedstructure, it has been an intention to usetelecommunications and IT to link theinstitutions together in a “knowledge-net-work” (“Norgesnettet”). In that respect,an infrastructure has been establishedbetween the central universities andregional colleges, with the intention ofco-ordinating the development ofspecialised competence and training. Inaddition, a broadband network has beenimplemented between the universities(34 Mb/sec), and a wide access to theInternet has been provided for all theupper secondary education and regionalcolleges.

The acknowledgement of the importanceof information technology and telecom-munication in education has led to aninterest for telebased distance educationin the education sector. Actions have

been put forth to speed up the develop-ment of such activities, and considerableexpenses have been invested in equip-ment [14]. A recent survey shows thatthere were 5.9 students per PC in uppersecondary education institutions, and 6.1in the colleges, which indicates a fastimplementation [15], [16].

Telenor Research & Development con-ducted its first survey on the educationsystem in 1989, and a subsequent studywas launched in 1994. The latter, whichwill be emphasised here, included a rep-resentative sample of 733 institutions,including 560 upper secondary educationinstitutions, 192 colleges and 38 univer-sity faculties. Both studies were directedtowards all schools, colleges and univer-sities in Norway.10 In this chapter, I willpresent some of the main results fromthis work. For the purpose of the present-ation, I will discuss the results from allthe levels in parallel. It should be men-tioned, however, that the entities here arenot entirely comparable, since universi-ties are organised differently from themore traditional upper secondary educa-tion institutions and colleges. There arealso important differences regarding theway they are managed and financed:While universities and colleges are stateowned, upper secondary education isfinanced by the local administration.

3.2 Distance education in theeducation sector

The number of Norwegian schools usingdistance education appeared to haveincreased markedly in the period from1989 to 1994. In upper secondary educa-tion institutions 13 % were offering dis-tance education in 1994, which repre-sents a 10 % increase, compared to theformer survey (see Figure 5). Among theregional colleges, about 35 % of theschools were offering distance education,indicating an increase by more than20 %. The fastest implementation of dis-tance education, however, was registeredat the universities, where 14 out of 37faculties were offering distance educa-tion in 1994. Compared to the 1989 fig-ures, when only 3 out of 45 faculties useddistance education, this implied an in-crease by more than 30 %. It was theregional colleges that included the largestgroup of students, and in the school year1993–94, a total of 7,825 college stu-dents were receiving distance education.

Most distance education in upper second-ary education involved subjects such as“general and business studies”, “healthand social studies” and “music, danceand drama”. At the regional colleges, thetypical distance education was takingplace within “pedagogics”, “general andbusiness studies” and “technicalscience”. At the universities this involveda wide range of subjects, including socialscience, humanities, medicine, law andtechnical studies.

Distance education played a role inschools all over the country, but in uppersecondary education there was more dis-tance education in schools located inrural parts of the country. In northern andwestern parts (Finnmark, Nordland,Møre- and Romsdal) the rate of schoolswho used distance education was twicethe total average. This difference was rel-evant to the distance between regions andcity areas. The predominance of distanceeducation outside the central areasbecame even more evident when we con-sidered the schools which were receivingthis from other institutions. While lessthan 18 % of the upper secondary educa-tion institutions in the cities were receiv-ing distance education, 35 % of theschools in rural areas did. This was thecase even for regional colleges: At thecity colleges, approximately 30 % offer-ed distance education and 17 % receivedit. Among the rural schools, in contrast,


10This included students from privatecolleges, but no private upper second-ary education.

40

35

30

25

20

15

10

5

0

Upper secondary schools Regional colleges University faculities

1989

1994

Figure 5 Number of upper secondary schools, regional colleges and university fac-ulties offering distance education, 1989 and 1994, in per cent

46 % were offering and receiving dis-tance education.

Furthermore, the occurrence of distanceeducation was related to the number ofstudents at the regional colleges and fac-ulties. Among the largest colleges (withmore than 700 students) more than 60 %had given distance education, but only20 % of the smallest schools (fewer than300 students). Among the biggest facul-ties (more than 1,000 students), 69 % hadused distance education, but only 8 % ofthe smallest (fewer than 500 students).

Which medium was used in the educa-tion sector? In upper secondary educationinstitutions and the universities ordinarytelephone was the most used technology,used by 50 % and 77 %, respectively (seeFigure 6). In the colleges letter/post wasthe most frequently used medium, men-tioned by over 60 %. In the uppersecondary education institutions 48 %had used letters in distance education,and the equivalent figure for the facultieswas 62 %. It is interesting to see that atthe regional colleges, e-mail is beingused more frequently than for exampletelephone conferences and television. Byand large, however, the colleges and uni-versities used more “advanced” technolo-gies than upper secondary educationinstitutions. Telefax, e-mail, telephoneconferences and databases were moreoften applied on the higher levels.

3.3 Future needs for distanceeducation

The results presented above document avery rapid implementation of distanceeducation in the education sector. If welook at the respondents’ plans for thefuture, there will be no slowing down ofthis development. First, at all levels themajority planned to escalate their dis-tance education activities. At the uppersecondary education and college level,approximately 86 % of the schools whichtoday have distance education pro-grammes, reported that they where goingto extend the activities. At the universi-ties, 11 out of 14 faculties which cur-rently offer distance education, reportedthat they would extend their activities.Second, a lot of schools which had notyet used distance education, planned tostart new projects within the next years.20 % of the schools on the upper second-ary education level and 30 % on the col-lege level were planning to launch a pro-ject during the following three years. Atthe universities, 9 out of 23 faculties

without distance education planned tolaunch projects.

To gain a better understanding of thefuture possibilities, however, we shouldtake a closer look at the underlying needsthat may trigger further distance educa-tion in the educational sector. The con-text of distance education is quite differ-ent to what we found in manufacturingindustry, primarily because the educationsystem in essence is directed towardstraining and teaching, and not imple-menting products, as in industry. Theschools consist of larger entities, morecontrolled by political decisions and res-olutions than fluctuations in commerceand business. On the other hand, theadvantages of using distance educationare more or less equal; easy and fastaccess to knowledge, reduction of ex-penses in relation to travelling andaccommodation.

Thus, one major reason for using dis-tance education is to reduce travellingfor employees at schools and universi-ties. In Norway, there is quite a traditionfor “decentralised education”, whereteachers travel to small groups of pupils.Our survey found that 164 upper second-ary education schools practised this typeof education, that is almost one third (seeFigure 7). Further, this was a 20 % in-crease since 1989. This pattern was evi-

dent among the regional colleges as wellas universities. Every second regionalcollege had provided decentralised edu-cation to students in 1994, a 10 %increase since 1989; and every third uni-versity faculty, representing a 15 %increase. This indicates that the potentialfor reducing travelling among teachersmay be one important reason for thegreat interest in distance education. Italso suggests that there may still be aconsiderable potential for substitutingtravelling activities by distance educa-tion.

Another aim for using distance educationis to gain fast and easy access to externalteachers. It appeared to be quite usual todraw upon externally employed person-nel in education. At upper secondaryeducation level, two thirds of the schoolsreported to have obtained competencefrom other schools or institutions. Thiswas done by 89 % of the faculties, andalmost every regional college. Despitethis, however, the schools reported re-ceiving a lot of inquiries from studentson subjects which they were unable toaccommodate. This was most prominentamong the regional colleges where 60 %admitted that they had received this kindof enquiry, followed by 50 % at universi-ties and 40 % in upper secondary educa-tion. At the same time, there was a lot ofcompetence among teachers not currently


80

70

60

50

40

30

20

10

0

Vid

eo c

onfe

renc

es

Vid

eore

cord

er

Inte

ract

ive

vide

o

CD

-rom

Vid

eoph

one

Sea

rch

on d

atab

aes

Ele

ctro

nic

mai

l

Pos

t

Slid

es

Doc

umen

t cam

era

Tele

phon

e

Tele

phon

e m

eetin

g

Tele

fax

Tele

visi

on

Rad

io

Upper secondary school

Regional colleges

Univerity faculities

Figure 6 Media used in distance education in upper secondary schools, regionalcolleges and university faculties, in per cent

used in education. Every other collegeand upper secondary education institu-tion reported that the school was in pos-session of competence that was notapplied, and 40 % of the university facul-ties. At all levels, there was more “sur-plus” than “deficiency” of competence inmost subjects. However, certain subjects,such as “general and business studies”,appeared to be a field where there wereparallel needs for more competence and alot of unused knowledge.

Interestingly, this reported “deficiency”and “surplus” of competence seemed tohave increased over the period. Forinstance, the number of regional collegeswith unapplied competence had increas-ed by 30 %, and the number without nec-essary competence by 10 %. This tend-ency was apparent for upper secondaryeducation schools as well as for faculties.This would suggest that the educationsector has become more “competence-intensive” over the five year period, withhigher demands for knowledge andteaching capacities. This seems to be ofparticular relevance for subjects whichfall within the category “general andbusiness studies”.

3.4 Summing up

The analysis of the education sector doc-umented that distance education haschanged, both qualitatively and quantita-tively during the years 1989 – 1994. Thenumber of distance education projects,and the number of students involved,

have increased at all levels. Upper sec-ondary education has the highest rate ofdistance education projects, while theeducation at regional colleges involvesmost students. The fastest implement-ation rate, however, has taken place atthe university faculties. To a certainextent, this must be considered a deliber-ate consequence of the official politics inthis field, which since the late eightieshave been emphasising the use of dis-tance education in various forms (see[14]).

New computer and telecommunicationmedia play an important part in thisdevelopment, and e-mail has been greatlyadopted by distance education in recentyears, especially at the regional collegesand the universities. Here, e-mail is morefrequently used in distance educationthan television and telephone confer-ences. The broad range of media indi-cates that education has become moremedia-intensive, even if it is difficult totell how the technology is used (as sup-plement, consulting means, etc.). How-ever, subjects such as “music, dance anddrama” reported on the list of distanceeducation, suggest that the technologyoften plays a secondary or supplementaryrole.

Moreover, the analysis indicates that theeducation sector has become moredependent on technology, as well. Thereis a general demand that schools offermore diverse and specialised educationand training, putting pressure on theschools to develop and expand their own

competence. Since a lot of schools havelimited budgets, distance education isoften an acceptable substitute for hiringcompetence or launching regular courses.The increased demands for specialisedknowledge may be the single mostimportant factor behind the push towardsmore distance education.

4 Conclusions

4.1 Differences and similarities

The building of a high capacity inform-ation and telecommunication network issaid to represent one of the greatest long-term investments in the 20th century[26]. Distance education is in this con-nection repeatedly emphasised as one ofthe applications which justify this invest-ment [28], [4]. There seems to be a gen-eral agreement that education will be oneof the most important elements in thefuture information society, both inrespect to personal development, busi-ness and industry, and culture.

In many ways, this trend has been visiblefor several years already; the number ofhigher-level students is still increasing inall western countries, and it is estimatedthat the working part of the populationmust be prepared to re-educate them-selves at least four times during theircareer [26].

This development costs money and time,and industry and the education sectorshould have a shared interest in applyinginformation technology and telecommu-nications to make education and trainingmore efficient, and of a higher quality.Computer mediated communication has apotential to function as a commonknowledge-network, enhancing the con-tact between industry and teaching. Oursurvey research on the manufacturingindustry and the education system hasproved that the education sector is in thevanguard, implementing distance educa-tion on a much faster scale than manufac-turing industry. While the rise of distanceeducation in both upper secondary educa-tion, regional colleges and university fac-ulties must be characterised as signifi-cant, manufacturing industry is develop-ing at a slower pace.

The differences between the two sectorsmust be viewed against the backgroundof very different contexts surrounding theindustrial enterprise and the school orfaculty. For the enterprise, training activ-


Upper secondary schools Regional colleges University faculities

50

40

30

20

10

0

1989

1994

Figure 7 Number of upper secondary schools, regional colleges and university fac-ulties offering decentralized education, 1989 and 1994, in per cent

ity is only one among several strategicinitiatives to develop the organisation,and in turn be more competitive. Aninvestment in internal competence is bal-anced against the long and short termassets and expenses. This makes theefforts behind the use of distance educa-tion far more vulnerable to changes fromexternal sources or internal strategies.The school, on the other hand, is part of agreater education system, more politic-ally controlled and co-ordinated. It has aresponsibility to provide education to stu-dents connected to the school (even ifthere are differences regarding the threelevels) and is not in a position to changethis from day to day.

These different contexts explain the dis-tinct patterns for adopting distance edu-cation. Regarding the use of telebaseddistance education, however, there arealso certain interesting similarities. First,in both sectors, distance education isdependent on large institutions. Thesmall schools, as well as the small enter-prises, did not apply distance educationon the same scale as the larger ones. Thesize of the institution seems to be closelyrelated to the ability to invest in neces-sary technology, but also the willingness(and financial capacity) to experimentwith new organisational forms. Second,the geographic location had to a certainextent impact on the use of distance edu-cation, among both schools and enter-prises. This was most evident in theupper secondary education schools out-side regional areas, but also regionalenterprises were more prepared to usedistance education in their training pro-grammes than the central industry. Third,there were parallel patterns in respect tothe use of media technologies: Post/letterand telephone was still the most usedtechnology in distance education, al-though at the regional colleges and uni-versities, the telephone was more usedthan letter/post. E-mail appeared as themost rapidly growing medium for dis-tance education purposes, both in theeducation sector and among industrialenterprises. Altogether, however, theeducation sector used a wider range ofelectronic media in their education, par-ticularly at colleges and universities.

4.2 Future challenges

What, then, are the future developmentsappearing from this analysis? I will con-tent myself with pointing out some majorchallenges, which follow from the dis-


cussion above. First, the investigationshows that there is no obvious link be-tween the political requirements of anescalation of knowledge and education,and the actual development in the small-est parts of the system. Small enterprisesand schools have generally less access todistance education with telecommunica-tion and computers. In Norway, wherethe majority of enterprises are small andmedium-sized, there are only vague signsof a regional knowledge network. Thechallenge will be to help these organ-isations gain easy access to both theglobal and the regional networks. Lowerprices on simple technological solutionsrelevant for distance education, may beone important effort. Still, it is importantfor the managers to orientate themselvesin the new possibilities for getting accessto training via distance education, that isto say, to “be competent in competence”.

Second, it is to a certain degree possibleto trace a development of polarisationbetween different regions. In the educa-tion sector, the universities seem toappear as the “knowledge providers” forupper secondary schools and the col-leges. None of the university facultieswere receiving distance education fromexternal institutions, while this was animportant trend in the regional collegesand the upper secondary educationschools. This approach to the problem isrelevant for the development of knowl-edge-based products in different parts ofmanufacturing industry as well. There isa risk for transforming the city areas(where universities and most of theindustry are located) to “knowledgeproviders”, and the regional areas to“knowledge users”.

Third, the analysis indicates that a lot ofthe distance education projects are rathershort lived. This is especially apparent inmanufacturing industry, where every sec-ond project has been launched during thelast three years. It has not been within therange of this work to analyse the distanceeducation projects in detail, but this pro-vides some hints of the problems con-nected to developing good technical andorganisational solutions. To get a clearerview of the situation, more researchshould be conducted in this field, prefer-ably trying to analyse the individual dis-tance education projects in a concreteorganisational setting.

References

1 Bates, A W. Educational aspects ofthe telecommunications revolution.In: Teleteaching. Proceedings of theIFIP TC3 Teleteaching Conference,Trondheim, Norway, 20–25 August,1993. Davies, G et al. Amsterdam,North Holland, 1993. (IFIP transac-tions A-29.)

2 Berg, L. Verkstedindustriens langsik-tige kompetansebehov : personalop-plæring i verkstedindustrien. Oslo,NAVFs utredningsinstitutt, 1987.(Notat nr. 7.)

3 Carmichael, J. Voice mail and thetelephone : a new student supportstrategy. In: Distance Education, 16,(1), 1995

4 Club de Bruxelles. The future of theinformation society. ANNEXE II.Brussels, 1994. (Bangemann groupreport.)

5 Fredrikson, C. De nya nordiska före-tagen. København, Nordisk Minister-råd, 1993. (Nord-rapport nr. 26.)

6 Frengen, G. Immaterielleinvesteringer i oljeutvinning,bergverksdrift og industri for1986–90. Oslo, Statistisk Sentral-byrå, 1993. (Notat nr. 14.)

7 Garrison, D R. Understanding dis-tance education : a framework forthe future. Routledge, London, 1989.

8 Gooderham, P. Bedriftsinternopplæring. Trondheim, Norsk Vok-senpedagogisk Institutt, 1985.

9 Gullichsen, A H. Strategisk kompe-tanseutvikling eller profesjonsstyrtetterutdanning : en analyse avopplæringsatferd i et utvalg kom-muner. Trondheim, Norsk Vok-senpedagogisk Institutt, 1992.

10 Helljesen, L, Rasmussen, T.Telekommunikasjon, fjernundervis-ning 2 : en undersøkelse av indus-trien. Kjeller, Televerkets Forskn-ingsinstitutt, 1992. (TF report R2/91.)

11 Industrial Research and DevelopmentAdvisory Committee of the Commis-sion of the European Communities(IRDAC). Skills Shortages in

Europe. Brussels, IRDAC Opinion,1991.

12 Julsrud, T E. Fjernundervisning inorsk industri 1989–1993 : resultaterfra en survey-undersøkelse. Kjeller,Telenor FoU, 1995. (FoU reportR 6/95.)

13 Julsrud, T E. Fjernundervisning iskolesektoren, 1989–1994 : resul-tater fra en survey-undersøkelse.Kjeller, Telenor FoU, 1996. (FoUreport R 19/96.)

14 Kirke- og Undervisningsdeparte-mentet. Mer kunnskap til flere. Oslo,1989. (Stortingsmelding nr. 43/89.)

15 Kirke- og Undervisningsdeparte-mentet. IT i opplæringa 91/92. Oslo,1992. (Rapport fra spørreunder-søkelsen.)

16 Kirke- og Undervisningsdeparte-mentet. IT i norsk utdanning : planfor 1996–98. Oslo, 1995.

17 Kristiansen, T. Fem års forskninginnen fjernundervisning : erfaringerfra forsøk med telekommunikasjon.Kjeller, Televerkets Forskningsinsti-tutt, 1992. (TF report R 17/92.)

18 Ljoså, E. Brev eller ny teknologi.Administrative og økonomiske siderved bruk av ny teknologi for fjernun-dervisning. In: Fjernundervisning :utvikling og mangfold. Oslo, Norwe-gian Centre for Distance Education,1993.

19 Nordhaug, O. 1985. Effekter av per-sonalopplæring : individ og organ-isasjon. Trondheim, Norsk Voksen-pedagogisk Institutt, 1985.

20 Nordhaug, O. Kompetansekjeden iorganisasjoner. In: Kompetanses-tyring, chapter 3. Nordhaug, O et al.Oslo, TANO, 1991.

21 Nordisk Ministerråd. Utveckling avkompetens för den lokala arbets-marknaden : flaskhalsprosjektetsslutrapport. Oslo, 1993. (NordiskeSeminar- og Arbejdsrapport nr. 580.)

22 Norges Offentlige Utredninger. Livs-lang læring. Oslo, 1986. (NOU1986:23.)

23 Norges Offentlige Utredninger. Medviten og vilje. Oslo, 1988. (NOU1988:28.)

24 Rasmusen, T. 1990. Telekommu-nikasjoner i fjernundervisning : enundersøkelse av skolesektoren.Kjeller, Televerkets forskningsinsti-tutt, 1990. (TF report R 4/90.)

25 Rekkedal, T, Søby, M. Fjernun-dervisningsmodeller, datakonferanserog nordiske utviklingstrekk. In: Fjer-nundervisning : utvikling og mang-fold. Oslo, Norwegian Centre forDistance Education, 1993.

26 Rodriguez-Rosello, L. Existing andfuture technologies for teleteaching.In: Teleteaching. Proceedings of theIFIP TC3 Teleteaching Conference,Trondheim, Norway, 20–25 August,1993. Davies, G et al. Amsterdam,North Holland, 1993. (IFIP transac-tions A-29.)

27 Rusten, G. 1995. The diffusion oftelecommunications among firms : anempirical regional approach. Bergen,Stiftelsen for samfunns- ognæringslivsforskning, 1995. (SNF-rapport 93/95.)

28 Samferdselsdepartementet. Dennorske IT-veien : bit for bit. Oslo,1996. (Rapport fra statssekretærut-valget for IT.)

29 Sentralorganet for fjernundervisningpå universitets og høyskolenivå(SOFF). Det nasjonale kunnskapsnet-tet : fleksibel læring i Norgesnettet.Tromsø, 1993. (Rapportserie nr. 5.)

30 Statistisk Sentralbyrå. Standard forkommuneklassifisering 1994. Oslo,SSB, 1994.

31 Statistisk Sentralbyrå. Industristatis-tikk 1991 : næringstall. Oslo, SSB,1991.


Do we need it?

The reaction of many people to thenotion of global education is resignedexasperation, querying whether we reallyneed yet another revolution in the teach-ing/learning paradigm. The spectre ofpeople dotted in continents here andthere around the world studying the samePhysics 101 course does not conjure upimages of a high quality learning envi-ronment comparable to the cosy pictureof student-led seminars, heads bent overbooks in library carrels, passionate argu-ments in late night coffee houses, andbrilliant lecturers spellbinding youngundergraduates. Why is global educationthen such a burgeoning phenomenon?Are we being driven to this by non-edu-cational pressures, or are we expandingin this direction voluntarily? And who isit that is leading the pack?

I begin by considering the extent of thephenomenon – how much education onan international scale is actually takingplace? It is hard to avoid encounteringthe avowedly evangelical predictionsabout the educational implications of thedigital this and the electronic that, andthe triumphalist announcements of theoptimists introducing the educationsuperhighway here and the virtual uni-versity there. It would be easy to thinkthat global education is already establish-ed practice, with courses, programmesand whole institutions jostling for posi-tion in the global marketplace. Whilstcompetition amongst tertiary educationproviders is certainly brisk in most devel-oped countries, the market in question,even for those offering distance educa-tion courses, is primarily regional, insome instances national, but rarely inter-national. In fact, the number of trulyglobal courses, let alone programmes orinstitutions, is still very limited. Never-theless, I can only conclude from mystudies that global education is a phe-nomenon to be reckoned with for allthose in higher education and training.

The arguments for global

education

There are a good many economic, socio-political and technological reasonsunderpinning current developments bothof the plans for and the practice of globaleducation. Perhaps they will prove to bethe strongest forces driving change.However, committed teachers and educa-tionalists would prefer to feel that a peda-

gogical rationale for a global studentbody or a global curriculum is the pri-mary consideration. Undoubtedly there isa case to be made for global courses onpurely educational grounds, and I havedivided these according to the reach ofthe course, the access to the course andthe development of the course.

The strongest argument relates to thebenefits of a global student body:

The diversity of participants made fora far richer course than I could everteach myself. Take, e.g., the time ourcorrespondent in Istanbul reported on alecture given there on medieval Chris-tian philosophy by a Franciscan priestto the faculty of the (Islamic) Univer-sity of the Bosporus. Well and good,the faculty opened when it was over,but it’s too bad Christianity is not atruly rational religion, like Islam.Leaving aside the question of compar-ative rationality of religions, I think itis undoubtedly good for my students atPenn [State University], taking acourse on a very traditional, ‘western’figure, to be reminded that the wholepicture looks quite differently if youhappen to be in a different seat. ([20]p. 113)

Here is the case of a course on the workand thought of Augustine, taught face-to-face to undergraduates and beginninggraduates on a traditional campus, beingoffered over the Internet in order to sup-plement the intake on such a specialisedtopic. Over 550 subscribers from aroundthe world – Bangkok to Istanbul – joinedup and undergraduates, scholars, andInternet-surfers all contributed to the dis-cussions. The taste of global interactiongiven to this teacher by his first ventureonto the information superhighway, lefthim concluding:

I cannot imagine ever passing asemester in the classroom again with-out the umbilical cord to the networkto energize, diversify, and deepen whatwe do. (op cit, page 114)

O’Donnell’s experience is echoed bymany others who have discovered theeducational benefits of a global studentbody. What is particularly welcome inthis example is that the globalisation ofthe course countered the usually domi-nant western world view.

The second most compelling educationaljustification for globalised education isthat of access. Whether potential students

be geographically remote, time constrain-ed, financially constrained, house-bound,disabled, or simply unable to find acourse on the subject they want locally,there exist large unmet educational needswhich every research report, policy studyand educational analysis shows is in-creasing.

For an institution which has an accessremit in its mission statement, globaleducation is now part of the moral highground and new technologies are provid-ing them with the means for reaching outto people anywhere, anytime, who wantto learn. The UK Open University(UKOU) is one of the most prominentexamples of an institution enshriningaccess to educational opportunity andopenness to students of all backgroundsnot only into its name and mission state-ments, but into the very heart of theorganisation: its preparation of coursematerials, its enrolment procedures, itstutorial support provisions and its com-mitment to developing similar organisa-tions around the world. This is how theUKOU rationalises the extension of thisapproach into a global strategy:

The University is committed to add-ressing educational disadvantage andwidening educational opportunities foran increasingly large and diverse num-ber of learners. Its mission statement,with its emphasis on openness to peo-ple, places, and methods, underlinesthe University’s commitment to re-spond to need and demand whereverthe means of delivery exist. In the past,limitations of educational technologyand funding have confined that abilityto deliver to fairly strict geographiclimits. Those constraints are rapidlydiminishing. The OU now has thepotential to extend educational oppor-tunities to a much wider body of learn-ers not only in the UK but throughoutEurope and more widely in the world.In doing so, it has the ability morefully to satisfy its mission. It has thepower to transform peoples’ lives, with-out regard to geographic frontiers. [21]

Related to the notion of access is thethird purely educational rationale forglobal courses: that the expertise of thefew can be made available to the many,such that those in remote areas can havethe same access to educational resources,specialist courses and renowned expertsas those located in large cities and devel-oped parts of the world. A student whosigned onto a Web-based course on the

69

The globalisation of educationB Y R O B I N M A S O N


Principles of Protein Structure offered byBirkbeck College, London, listed one ofthe main advantages of this virtual courseas the easy and informal contact it pro-vided with a large number of fellow sci-entists around the world.

Everyone is on first name terms andthe mailing lists, along with the groupstructure, provide a convenient way todiscuss issues raised by the coursematerial, and to deal with any prob-lems people may be having. [22]

Athena University, one of the newlyestablished electronic universities, seesits mission as providing high-quality edu-cational opportunities on the Internet asinexpensively as possible. It is able tocarry out this mission by utilizing theInternet resources of the entire globe(http://www.athena.edu).

The expertise located at one university inthe United States is shared with ninetystudents in three other countries: Mexico,Finland, and Estonia, providing access toa Certificate Programme in DistanceEducation, through a combination ofaudioconferencing, computer conferenc-ing and recorded media. [5]

Finally, and some might say most impor-tantly, a good many areas of the curricu-lum are inherently global in nature andsome particularly lend themselves tocourse development on an internationalscale, providing students with a muchbroader perspective than a course pre-sented by a single lecturer or developedby a single institution. A good exampleof trans-national course development isthat of the European Association of Dis-tance Teaching Universities which hasencouraged the joint authoring amongstits members of two programmes: theEuropean MBA and a comprehensivecourse in the humanities, entitled ’Whatis Europe’. The advantage of this kind ofinternational collaboration is elaboratedby Trindade, President of the PortugueseOpen University:

In Europe, the very old universitiestend to perpetuate rigid (not to sayextremely conservative) educationalsystems ... This weight of tradition isreflected in university programs thathave scarcely varied in format or des-ignation for at least half a century,despite intense basic and applied re-search activity in advanced or innova-tive fields ... Distance Teaching Uni-versities seem to be more accepting ofnew ideas or new models, more prag-

matic in their approach to cooperatingwith each other, and more daring intheir attitude towards trans-nationalcollaboration. ([28] p. 40–41)

These, then are the major advantages toglobal education as cited by some of thepractitioners teaching and learning at thecoal face. At its most visionary, the idealof global education is one of a movementaway from the bounded classroom, seenas a haven from the world, self-containedand static, to a dynamic synergy of teach-ers, computer-mediated instructionaldevices, and students collaborating tocreate a window on the world. Interactionwith learners on a global scale leads to anincreased awareness of the extraordinarycomplexity of interrelations and a rela-tivistic comprehension and tolerance ofdiverse approaches to understanding.

Other advocates of the movement in thisdirection, however, see a range of lessidealistic factors driving education ontothe Superhighway.

Pressures encouraging

global education

The reasons behind the, in some cases,drastic reduction in funding of publicuniversities, as well as the falling popula-tion of traditional 18 – 20 year old stu-dents, at least in western countries, areoutside the scope of this article. Suffice itis to say that financial pressure on institu-tions of higher education world-wide isprobably the most critical factor forcingadministrators and policy makers to lookto global markets as a way of making upfor falling government revenues and fall-ing numbers of traditional learners.

While the search for new sources of rev-enue has provided the impetus for uni-versities to look for new markets, largersocial changes have steered that searchtowards globalisation. In fact, global edu-cation is a reflection or extension of soci-ety’s increasing understanding of theinterrelatedness and interdependence inthe physical world. The development ofthis global consciousness has beenheightened by the spread of global com-munication systems and particularly theentertainment media.

Distance open learning appears to beuniquely suited to the emerging worldorder. As borders open up across theglobe to traffic of almost every kind,so distance open learning flows in-

creasingly across national frontiers.[11]

Field contends that trans-national educa-tion represents both an outcome of and aprimary factor in the intensification ofglobal interconnectedness. Education, hepoints out, used to be regarded as anessential element in nation-building. Aswith other social and political trends,education is increasingly being thoughtof as a commodity to be shaped accord-ing to consumer demand.

Some commentators on globalisation seethe trend towards student-as-consumer ashaving positive educational outcomes.Teachers and course developers are beingforced to consider the requirements oflearners, and the global telecommunica-tion systems allow students’ opinions tobe embedded into the learning environ-ment.

Technologically-mediated knowledgeprovides the basis for individualisinglearning in a more complete and activeway ... Here distance is subservient tothe discourse of open learning and‘educative’ processes are displacedand reconstituted as relationships be-tween producers and consumers inwhich knowledge is exchanged on thebasis of the usefulness it has to theconsumer. It is therefore a discourse ofopen learning which might be said tomore fully govern the practices ofthose operating at a distance in thepost-modern moment, as increasedmarketisation is introduced into theprovision of learning opportunities andmass markets fragment and becomemore volatile across the globe. [8]

For Edwards, the globalisation of educa-tion through the use of telecommunica-tions technologies will empower thelearner and force the providers of educa-tion to concern themselves with students’needs, rather than with the transmissionof a pre-established canon of knowledge.Educators, just like businesses, will haveto become more flexible – in their staff-ing ratios, in their approach to students,and in their considerations of the curricu-lum. Various structural rigidities of tradi-tional universities will have to be over-come: constraints on what constitutes theacademic year, on where credits can beaccumulated, and on how courses can bemodularised. The kinds of courses whichthe global consumer is demanding areflexible, adaptable, portable and inter-active.


Dangers of global

education

Although the enthusiasts for the globali-sation of education are more prominentin the media, they are probably equalledin number by the voices crying for a haltto the headlong expansion of global edu-cation on the Superhighway. I havefound it useful to classify their argumentsas primarily cognitive, educational,social and cultural.

The cognitive argument is based on thefact that the new delivery mechanismsfor most global education are electronicand rely largely on the digitisation andcomputerisation of knowledge. Manypeople decry the cognitive effects oflearning from screen-based informationrather than traditional text-based mate-rial, pointing to the breakdown of linear,narrative structures associated with thebook, and the resulting fragmentationand superficiality induced by the hyper-linked structures of the Web and multi-media CD-ROMS. One of the more elo-quent apologists for the culture of booksis Sven Birkerts, who tots up the cogni-tive losses we are encurring with the riseof an electronic culture:

In the loss column are (a) a fragmentedsense of time and a loss of the so-call-ed duration experience, that depth phe-nomenon we associate with reverie;(b) a reduced attention span and a gen-eral impatience with sustained inquiry;(c) a shattered faith in institutions andin the explanatory narratives that for-merly gave shape to subjective experi-ence; (d) a divorce from the past, froma vital sense of history as a cumulativeor organic process; (e) an estrangementfrom geographic place and commu-nity; and (f) an absence of any strongvision of a personal or collectivefuture. ([3] p. 27)

Birkerts acknowledges that these areenormous generalisations, but he feelsthat they accurately reflect the commentsof his students about their own experi-ence.

The educational argument against globaleducation centres on the undesirableaspects of consumerism, wherein learn-ing ceases to be about analysis, discus-sion and examination, and becomes aproduct to be bought and sold, to bepackaged, advertised and marketed. Inthe previous section I presented the casein favour of competition in course provi-

sion. However, just as there are manywho do not accept its value in a wholerange of social services, so there arethose who view the growing competitive,consumer spirit amongst educators asdetrimental to the learning outcomes(see, for example [19]). This marketplacephilosophy of higher education is partic-ularly associated with distance education,which in turn is the foundation of themovement towards globalisation.

One example of the shoddiness of globalcompetition in education is the numberof MBAs offered by a range of Westerninstitutions in the advanced countries ofthe Pacific Rim. Some are outright cons;others are just poor quality; many are‘sold’ for high prices which are then usedto defray the deficit incurred by studentstaught face-to-face in the home institu-tion. Together with legitimate, high qual-ity, well supported programmes, they alljostle in the marketplace to the bewilder-ment of prospective students, who havefewer standards for and less experiencewith assessing the quality of globalcourses than they have in judging localprogrammes.

Social arguments against globalisationare related to the breakdown of commu-nity. This phenomenon is part of a muchlarger, more complex web of changesassociated with Post-modern society;nevertheless, education which has alwaysbeen a net contributor to the positive ben-efits of physical communities, is nowseen as undermining still further thephysical experience of community andoffering instead a much less substantialsubstitute in the form of virtual commu-nities.

The global classroom does have a com-munal atmosphere: it has interaction be-tween students and teachers, and net-working and serendipitous encounterswith other learners both those followingthe same course and those simply outbrowsing, searching or talking on theInternet. There is evidence of many ofthe elements of traditional communities:people help each other (by forwardinginformation to each other which theythink will be of interest; they answerqueries from strangers, providing themwith the benefits of their expertise, for nopersonal gain; they trade confidences andintimacies; they play games together;they meet regularly. Nevertheless, someobservers of this new phenomenon seegreat danger and significant social loss inthe demise of physical community and its

replacement by virtual community nomatter how educational:

There seems to be a great difficulty inholding onto the truth – as obvious asit is – that ease and flexibility ofswitching do not constitute ease anddepth in making human contact. Cer-tainly the connectivity makes commu-nication ‘easier’, but it achieves thisprecisely by rendering contact moreincidental, more shallow.

I suspect that every major technicaladvance in communication since theinvention of writing has marked thepath away from community rather thantoward it. Community is, in the firstinstance, something to be salvagedfrom information technology, notfurthered by it. ([26] p. 74)

Others warn about the loss of our funda-mental assumptions regarding identityand subjective meaning by moving fromthe physical and substantial to the virtualand electronic. The notion of communityhas always been rooted in a sense of theparticular and this has characterised ourexperience over millennia. The essenceof our education system has been thecommunity of the classroom and thephysical reality of the textbook. It haschanged relatively little over the past fewhundred years.

What we will have in the next fewyears is an education system that ispart of computer culture. It is not justthe physical environment that will betransformed. Whereas books haveencouraged us to think in terms of astable body of knowledge, a form ofcontent that we can read, digest, learn,and know, computers dispose us tothink differently – to be engaged in aconstantly changing process whereinformation is not stable or fixed. ([25]p. xxiii)

The effects of this destabilisation onsocial and personal assumptions has beenanalysed at length by Poster [23], whospeaks of the dispersal of identitythrough computer networking.

Finally, the cultural arguments againstglobal education systems are equallycompelling, and harken back, of course,to old concerns about imperialist atti-tudes, the loss of indigenous cultures andthe relentless imposition of Western val-ues. Global educators are seen as the newcolonisers, insensitively spreading theirown views of the world onto developing


nations in the mistaken belief that theyare actually helping people:

The access to this sort of provision canbe seen to have considerable educa-tional (and possibly, social and eco-nomic) benefits to a developing nation.However, as an invasion, it can be seento weaken national initiatives todevelop local educational provisionwhich might be better suited to localneeds. It also creates the potential for apost-colonial dependency on another‘developed’ nation. [9]

Evans argues that despite the value ofglobal education in offering diversity ofchoice, this comes at the expense ofencouraging local initiatives which valuelocal culture and promote national be-liefs, skills and knowledge. The potentialpower of globalised teaching to spreaddominant ideologies and to crush emerg-ing structures, whether wittingly orunwittingly, is the main cause for con-cern. Moore, who also expresses concernabout the cultural implications of globaleducation, says that truly internationaldistance education courses would involveall participants (including the teachers,experts and Western students) in a re-examination of their educational philoso-phies, their views of the subject beingtaught and their cultural perspective ofthe content of the course. In practice, hefeels, Westerners tend to be arrogantlyuncritical of the assumptions underlyingtheir teaching and unreflective of theirfitness for teaching across cultures ([19]p. 189).

Current practice

I have presented some of the most signif-icant views of the enthusiasts and practi-tioners of global education and have out-lined the wider pressures leading to theglobal expansion of higher education. Ihave also represented the arguments ofthose concerned with the negative impactof global education. So having brieflyconsidered why it is happening, I turnnow to an analysis of what is happeningunder the guise of global education.

I suggested earlier that there is a gooddeal less global education being practisedthan would be imagined from readingnewspaper articles, browsing the Web orlistening to media hype. Let me beginwith a listing of the various elementswhich characterise global education.Many self-designated global courses arein progress which do not meet all of

these criteria; indeed, I have not found asingle course which has all of these char-acteristics. My list is roughly in order ofsignificance:

1. students spread roughly evenly in threeor more countries of the world

2. an express aim on the part of theteacher or institution to attract interna-tional participation

3. course content which is devised withtrans-national participation in mind

4. support structures – both institutionaland technological – to tutor andadminister to a global student body.

For the purposes of this article I am notgoing to consider the various global ini-tiatives operating at the level of primaryand secondary education. While some ofthe issues I discuss will also apply tosome extent to these schemes – culturalbenefits and threats, cognitive effects ofcomputerisation and screen-based learn-ing, educational value of a global cur-riculum – I want to focus primarily onthe higher education sector (includingtraining and professional updating), andon the notion of a university withoutplace, rather than on a classroom withoutwalls.

Secondly, I know that there have been anumber of global programmes runningfor some time, even quite large-scaleoperations, which are based on print andpostal systems only. I do not intend toinclude these non-telecommunicationsbased programmes within the compass ofmy discussions, because I consider thatthey have been effectively superseded bythe advent of the Internet and globaltelecommunications systems. In fact,many of these programmes are alreadybeginning to introduce elements oftelecommunications into their support oradministrative procedures.

As is apparent from some of my argu-ments so far, I will include current de-bates surrounding Virtual Universitiesand on-line courses, whether or not theyare actually operating globally, becausethe issues and outcomes of these prac-tices are fundamental to global educa-tion; indeed they are the forerunners of it.

Using the four attributes of a globalcourse listed above, I can discuss currentpractice on a continuum from courseswith global aspirations to those alreadyin operation.

1 A global student body

By far the largest number of ‘pseudo-global’ courses – and most originatefrom North America – have no face-to-face requirements, and all course mate-rial, administration and support are pro-vided either electronically or by post.The majority of students are NorthAmericans, but occasionally a studentabroad (British or Australian perhaps)will hear of the course and ask to enrol,particularly in specialist areas where asimilar offering is not available locally.More commonly, a number of NorthAmericans move abroad or are sent bytheir company during the period of thecourse or programme, and they continuetheir studies from the new country. Thesecourses are global in fact but not in spirit– the content has not been altered; theinteractions are amongst people of thesame culture; the institution has not beenre-engineered for a global mission.

Nevertheless, there are some courses andeven programmes currently taking placewhich have students from a number ofdifferent countries and cultural back-grounds. For example, many trainingprogrammes offered by global companieshave trainees in North America, Europe,Asia and Africa following the samecourse. The delivery medium is usuallysatellite television with the employeesaccessing from the workplace.

2 An international aim

A most interesting class of courses arethose where the instructor, often a singleenthusiast within the institution, designsand writes a course (usually Web-based)and makes it available globally to anyonewith access. In many respects this ‘earlyadopter’ class of courses provides themodel of best practice for true globalisa-tion. The big difference obviously, is thatthese courses are free, are not accreditedand usually ‘one-off’. Because of theiroutstanding success, the instigators moveon to institutionalising (accrediting,charging, marketing) them. One examplehas already been mentioned: the Princi-ples of Protein Structures at Birkbeck(http://www.cryst.bbk.ac.uk/PPS/index.html). Others in this category continue tobe offered in their original form: forexample, Roadmap is a free, twenty-seven lesson Internet training coursewhich 80,000 people in 77 countries hadtaken by email in 1994, and by 1995, thenumber had increased to 500,000. Spec-trum Virtual University offers freecourses about the Web and the Internet,


as well as ‘focus groups’ for ‘hands-on’learning in small exploratory teams(http://horizons.org).

A few courses specifically aimed at aglobal audience, which nevertheless arefee-paying, accredited and on-going, doexist, and many more are in the planningstages. See, for example, the GraduateCertificate in Open and Distance Learn-ing by the University of SouthernQueensland (http://www.usq.edu.au), andthe UKOU’s Masters Degree in the samesubject (http://www-iet.open.ac.uk/iet/iet.html).

3 A multi-cultural coursecontent

There have for many years been twinningprojects between universities in whichone (usually a Western) institution devel-ops a course or programme in conjunc-tion with one or several (usually non-Western) institutions. Recently, thecourses resulting from these collabora-tions have been delivered electronically.The Télé-université in Quebec has car-ried out this kind of course developmentwith other French-speaking nations [29]for example, and many universities havelong standing arrangements with institu-tions in South America, Asia and Russia.

A number of franchise arrangements alsofall into this category – the core contentremains the same, but the local institutionadapts the material (by translation, byincluding local case study material, or bycustomising the length or degree of diffi-culty of the material). The resultingcourse is often of very high quality, con-taining the best of both worlds – an inter-national perspective with a grounding inlocal or national concerns. The UKOUhas large programmes in Eastern Europe,Singapore and Hong Kong which demon-strate this quality of globalisation [27].

Finally, there are a range of collabora-tively developed courses across all areasof the curriculum which encompass theperspectives of different cultures andapproaches to the discipline. Gayol de-scribes the activities of NADERN, theNorth American Distance Education andResearch Network [12] and the ‘What isEurope’ course authored by several Euro-pean institutions is another example.

The shortcoming of most of these pro-grammes is that the student body remainsnational, and therefore does not benefitfrom interaction with students or teachersfrom a mix of countries.

4 Global support structures

Very few universities have developedtruly global support structures; the train-ing programmes within globalised indus-tries are probably closer to realising thisaspect of global education. Whilst theInternet (in the form of electronic mail)provides a relatively trouble-free mecha-nism for administration, sending inassignments and receiving one-to-onetutoring, none of these features scale upto handling large numbers of students.Indeed, what we find most commonly, isthat any course or programme operatingglobally, is only doing so with very smallnumbers of students (typically under 100,and usually closer to 20). Frequently thecourse is administered and tutored by thecourse author – what can be categorisedas a one-man band!

In many respects, the technology is not inplace yet to manage large-scale globalteaching. The aspects of support requiredin global courses include:

• a system for many-to-many communi-cation, such that students can interactnot only with the tutor, but also withother students, and is accessible at rea-sonable speeds and reasonable cost ona global level

• a system for handling the electronicsubmission of assignments, both forthe purposes of annotating, comment-ing and marking by the tutor, but alsofor recording, monitoring and storingby the accrediting institution

• a system for marketing courses andprogrammes, handling registrations,fees and queries from around theworld, preferably electronically.

Examples of these systems operatingmore or less successfully do exist, butcertainly not on any large scale. MonashUniversity has a system for managing theelectronic submission of assignments[17]; a number of universities use variouscomputer conferencing systems on aninternational level for many-to-manycommunication [13] and a combinationof Web pages, Web forms and email,allow many on-line courses to be advert-ised electronically and for perspectivestudents to register and obtain furtherinformation. [7]

While the number of truly global coursesis quite small, there are many globalactivities and events taking place withinthe higher education and training sectors.These can be categorised as:

• global research projects, in which par-ticipants communicate, exchange dataand draft joint reports and books elec-tronically, using simple email, audio-graphics, videoconferencing andCUSeeMe on the Internet

• consortia development activities, inwhich perspective partners developmutual understandings which areintended to lead to global educationalpartnerships, or, as with the GlobewideNetwork Academy satellite videocon-ferences, which foster a climate forglobal educational activity

• conferences held electronically, usingthe Web, computer conferencing, list-servs or videoconferencing, either as asubstitute for any face-to-face meetingor as a method of extending access tothose unable to attend physically (twoexamples are Teleteaching 96, VirtualUniversity Web-based discussion athttp://www2.openweb.net.au/TT96University/, and [1]).

• global programmes in which theadministrators and/or the tutors com-municate electronically, but the stu-dents do not

• global videoconferencing in the work-place for such activities as presenta-tions to all employees, for product pro-motions, job interviewing, or interna-tional project work.

The point is that these activities promoteunderstanding and acceptance oftelecommunications media and developrelevant technical and social skills inhandling global communications.

Global institutions

What I have presented so far is a picture,not of a revolution in the higher educa-tion sector, but more of a sliding scalefrom ‘traditional’ distance education, tointernational distance education, to on-line courses to virtual universities andfinally edging to globalisation. Most ofthe operations are piecemeal, consistingof a good deal of flag-waving fromsenior staff, or idealistic visions of neweducational paradigms from educationaltechnologists, or financial officers rubb-ing their hands in expectation, but at theend of the day, a very few academicsactually delivering something that couldbe called global in parts.

I turn now to the question of who is lead-ing the globalisation movement. One


response can be given with certainty: it isnot the established institutions! This isnot surprising considering their invest-ment in physical buildings, their aca-demic traditions and their self-imposedguardianship of quality, standards andresearch. One then looks to the newerinstitutions as being more flexible, morein need of making their mark and moredesperate to find sources of income. Cer-tainly this sector has adopted computertechnology more wholeheartedly and isexperimenting more readily with variousforms of distance education. However, itfaces many internal barriers:

• lack of appropriate staff training inorder to teach with new technologies

• lack of an appropriate reward structureto attract staff to adopt new methods

• lack of resource to fund the develop-ment of global courses.

In fact, many of the ‘one-man band’global courses do come from institutions(usually teaching face-to-face) where oneor two early adopters are enthusiasticallygoing ahead with little institutional sup-port and much hard work, simply be-cause they are committed to the princi-ples or enjoy the actual practice.

The distance teaching universities areobvious candidates for leading the fieldin the globalisation movement. While theUKOU is one of the most prominent,many distance teaching institutionswhether dedicated or dual mode, havesome partially global courses – franchis-ing systems, arrangements with one ortwo other countries, or a few Web-basedcourses being piloted to test the globalwaters. Nevertheless, these institutionsalso have entrenched attitudes, bureau-cratic procedures and general inertia toovercome before launching themselvesas global institutions. Re-engineering aneducational institution to teach in newways, with new media, to new kinds ofstudents is not an overnight task.

If one institution finds it difficult to gath-er the necessary resources, train its staffand provide whole programmes ratherthan odd courses, what about consortia?In fact, many of these exist both nation-ally and trans-nationally to share deliverytechnologies, course development,accreditation mechanisms, marketing andregistration procedures and programmeplanning. Some of these are well-known,such as the National Technological Uni-versity; many others are still in the early

stages of consolidation and have notestablished a reputation. Without a doubt,consortia of all kinds will become thenorm in the long term development andlarge scale build-up of global education.The complexities of teaching, administer-ing, supporting and developing globalprogrammes makes this an obvious solu-tion for the university sector to adopt inresponding to the pressures for globalisededucation.

Another, slightly different solution hasappeared very recently, arising from theuniversity sector but based on the newcyber-speak. These are the virtual univer-sities, designed for a global platform andoperating purely electronically. Some areidealistic and visionary in aim, such asthe Globewide Network Academy (http://uu-gna.mit.edu:8001/uu-gna/index.html),although this originated in 1994, Spec-trum Virtual University (http://horizons.org) and Virtual Online University(http://www.athena.edu); others areavowedly ‘for profit’ such as PhoenixUniversity (http://www.uophx.edu) andIMLearn (http://www.imlearn.com). Thescale of these operations is variable:many of them on investigation prove tobe working from one office; others havesignificant numbers (e.g. 2000 registeredfor on-line course at Phoenix University),although very few students will be non-nationals. On the whole, these are notuniversities in the commonly acceptedsense of the term: they borrow academicsfrom other institutions rather than fundtheir own established full-time faculty;they do not cover a full range of disci-pline areas; they do not fund researchprogrammes and they do not supportwhat is usually called an academic envi-ronment. Nevertheless, they providecourses which people want; they are cap-italising on the phenomenal growth ofthe Internet and are trail-blazing theglobal pathways for others.

There exists one final model which mightbe said to be leading the field: the neweducational providers. This group con-sists of organisations whose primarybusiness is not, or has not been, educa-tion. Often they have services or productswhich have now become central to thedelivery of global education. The obvi-ous examples are the telecommunicationsproviders, whether satellites, cable, tele-phone or combinations. Other examplescome from the computer and softwareindustries. The advent of these newproviders offering professional updatingprogrammes, adult education courses,

life-long learning opportunities and just-in-time training resources to what hasalways been the market monopolised byuniversities and continuing educationunits, has caused ripples of alarm in allbut the most un-reconstructed universi-ties [18]. The natural advantages whichthese new providers have over the estab-lished education sector are:

• in many cases they control the meansof delivery

• they do not carry the academic ‘bag-gage’ of established institutions andare free from their bureaucratic andentrenched attitudes to education

• as ‘green field’ sites they can set upsystems geared specifically to a globalmarket, rather than having to adaptexisting procedures

• they hire content specialists from tradi-tional universities, but do not suffer theexpense of supporting a research pro-gramme from the teaching revenues.

These points raise many questions aboutquality, accreditation, and ultimately thenotion of graduate-ness and the purposeof universities. They are not issues spe-cific to global education, nor are theynew problems to the education sector.Several examples include:

• The Global TelecommunicationsTraining Institute is a brokerage ser-vice of distance learning courses tosupply training on-line for staff oftelecommunications companies world-wide. It is based in Geneva, and is cur-rently in the planning stages.(http://www3.itu.ch/VTC/gtu_gtti.htm)

• Microsoft Online Institute (MOLI) isan on-line education programme whichat the moment is used only for promot-ing Microsoft product training and cer-tification. Trainees download coursematerials from the Virtual CampusEnvironment, and can engage in realtime discussion and feedback sessionswith the tutor and other participants.The system is built around MicrosoftNetwork which ships with Windows‘95. (http://www.microsoft. com)

• Motorola University, Schaumburg, Ilis part of the growing trend for corpo-rate universities to move to the VirtualCampus model, by making strategicalliances with universities and on-lineproviders. For example, using the en-gineering courses offered by theNational Technological University(NTU), Motorola has over 300 engi-


neers in North America and Asia fol-lowing masters degree programmes.By acting as NTU’s main customer inAsia, they made it possible for NTU toestablish an Asian service, and nowother global companies such asHewlett-Packard have joined, makingit possible to justify one full transpon-der devoted to delivering NTU pro-grammes in Asia.(http://www.mot.com/)

• Mind Extension University is foundedby Glenn Jones to extend equality ofopportunity for higher education tothose unable to attend on-campusclasses. Using a satellite and cabletelevision network dedicated to dis-tance education, it offers many coursesand degree programmes through 30regionally-accredited universities andeducation providers. (http://www.meu.edu:80/meu/)

Technologies for teaching

at a distance

That computer technologies change andevolve more quickly than books aboutthem can be published, is obvious fact.But educational use of computer andtelecommunications technologies, or atleast widespread take-up by institutions,is much slower, so that it is possible todiscuss current delivery technologies andspeculate about trends with some degreeof certainty about the direction, if not thedetails, of global education deliverymedia. As I mentioned earlier, it is dis-tance education, as presently practisedusually on a national scale, which is theforerunner and first pioneer of globaleducation. It therefore makes sense tostudy the technologies in use today, inorder to assess the benefits and limita-tions of various media for the global edu-cation of tomorrow.

There are three broad categories withinwhich current technologies can be divid-ed:

• text based systems, including elec-tronic mail, computer conferencing,real time chat systems, MUDS/MOOs,and many uses of the World Wide Web

• audio conferencing and audio exten-sions such as audiographics, and audioon the Internet

• videoconferencing, one way and twoway, video on the Internet with prod-ucts like CUSeeMe, and other visualmedia such as video clips on the Web .

The implication of this list is that text,audio and video are discrete media.While this is partially true today, the evo-lution of all these systems is towardsintegration – of real time and asyn-chronous access, of resource material andcommunication, of text and video.

I have not included multimedia CD-ROMs in this list. Although they com-bine elements of text, audio and videoand have tremendous potential as stimu-lating learning resources, they lack themain ingredient of person-to-personinteraction. Furthermore, they are diffi-cult and costly to update, and problem-atic as a global distribution medium. Allof these disadvantages of CD-ROMs areovercome, however, in my fourth tech-nology, which is the rising star of globaleducation delivery:

• the World Wide Web, which integratestext, audio and video, both as pre-pre-pared clips and as live interactive sys-tems, both real time and stored to beaccessed later, and furthermore pro-vides text-based interaction as well asaccess to educational resources ofunprecedented magnitude.

I will discuss each of these four media inturn, looking at the main educationaladvantages and disadvantages and con-sidering their potential for global coursedelivery.

Text-based systems

Without a doubt the most commonlyused technology for communicating withstudents at a distance, is computer con-ferencing. Text-based interaction,whether many-to-many in conferences,or one-to-one in electronic mail, is prac-tised at most institutions of higher educa-tion, whether students are geographicallyremote, or actually on campus. In fact,with the change towards the new major-ity (students who are older, or have somekind of part-time employment, or havefamily commitments or other barriers toattending campus full-time), many face-to-face teaching institutions use text-based interaction as a means of commu-nicating with students, thereby reducingtheir dependence on physical attendanceon campus at specific times. The term‘close-distance education ‘ is used todescribe this new phenomenon.

Some institutions use standard electronicmail systems (which include the facilityfor sending messages to a group) to com-municate with students at a distance.

Those accessing from abroad usually usethe Internet; those living locally use amodem over telephone lines. The pri-mary use is for students to ask questionsof the tutor, but an additional use is theelectronic submission of assignments, asan attachment to a mail message. This isthe simplest and most accessible of allthe telecommunications technologies,with the possible exception of FAX.

More commonly in distance education, aproprietary computer conferencing sys-tem is used. FirstClass is a very success-ful product amongst educators, and LotusNotes is common particularly in BusinessSchools and in training organisations.Microsoft Exchange is a new systemwhich is expected to have very wide-spread use. Several others are also popu-lar, and a number of institutions havedesigned their own, with a purpose-builtinterface for the facilities they offer.Computer conferencing systems allowstudents on one course to share discussionareas, to have sub-conferences for smallgroups, and to have easy access to all thecourse messages throughout the length ofthe course. Computer conferencing sys-tems are slightly more complex thanemail: they may require the student tohave client software; a faster modem maybe necessary or at least highly desirable;and in the case of Microsoft Exchange, apersonal computer of a particular specifi-cation is necessary.

MOOs are a text-based virtual realitydevelopment from MUDs, an acronymfor Multi-User Dungeon, a class of pro-grams for playing dungeons and dragonsover a network. From these real time,text adventure games evolved the MOO,which stands for MUD, Object-Oriented.Unlike conventional chat rooms, MOOsallow the manipulation and interactionwith cyber-objects as part of the commu-nication with other people. When inte-grated into the learning experience,MOOs can be used to create an environ-ment in which students interact moredirectly with the course ideas than theywould in an unstructured discussion.

Text-based systems can be divided acc-ording to whether they are primarily syn-chronous or asynchronous in use. Moreaccurately, while the technologies usu-ally support both, in practice, one or theother is the primary intended use, andthis influences the design of the interac-tion features. Most text-based communi-cation systems are used primarily to sup-port students (with the contents of the


course delivered through some othermedium); however, some educators run‘on-line courses’ in which the primarycontent of the course is the discussionsand activities taking place amongst thestudents. As this technology has been inuse on a relatively large scale for nearlyten years, and as some of these uses haveinvolved students in many differentcountries, it is the most obvious area tostudy for issues arising from global edu-cation.

One of the major advantages of text-basedmedia is that they facilitate interaction forthose using their second language. Mostpeople are better able to write in anotherlanguage than speak. Furthermore, asyn-chronous systems allow time for readingmessages slowly and composing aresponse with the aid of a dictionary. Notsurprisingly, there are a range of very suc-cessful asynchronous text-based pro-grammes at an international level for sec-ond language teaching. Usually they pro-vide natural language practice withmother-tongue speakers, which is muchmore engaging and profitable education-ally than artificial classroom practice.

Another primary advantage of any text-based system for global education is thatmany people world-wide can access themusing a personal computer and telephoneline from their home. In fact, althoughthere are a few uses of real time chat oreven computer conferencing in whichstudents go to a study centre, campuscomputer room or training centre, mostuses of these systems are from the stu-dent’s own machine (whether at home orin the workplace).

My third advantage of text-based systemsis rather more contentious. Much hasbeen made of the equalising effects oftextual communication – the concentra-tion on what is said rather than who saysit. While it remains the case that the dis-abled and the disadvantaged can partici-pate without the usual judgmental reac-tions, text-based systems do not removebias and ‘advantage’, they merely shift itaround a bit. Clearly those who have reg-ular access (for example, from both workand home) or have no concern about thecost of access, are advantaged in terms ofbeing able to participate in discussionsmore easily than those who haverestricted access and cost considerations.Furthermore, those who have good writ-ing skills tend to dominate by the veryquality of their messages, in that less lit-erate participants defer to them or simply

are deterred from putting in messagesthemselves. Finally, the openness ofthese systems to anyone, anytime tomake their opinion known, to respond toother viewpoints and to engage in dia-logue, is true in theory, but in reality,messages not following the main thrustof the discussion (i.e. keeping up with theongoing conversation) tend to be ignor-ed. Abuses of the openness of the system,such as flaming, harassment, and anti-social behaviour, have driven participantsaway and generally damaged both theimage and the value of text-based com-munication (although this is much lessprevalent in educational than socialuses). So what was originally hailed as anew democratising medium, inherentlymore open than other modes of commu-nication has been shown in practice overtime to be as flawed as the human beingswho use it. Nevertheless, for somegroups of people, text-based interactionallows access to education in a formideally suited to their situation.

Audio systems

Straight audio conferencing using ordi-nary telephone lines is a ‘low-tech’ solu-tion to supporting students around theworld, due to the near ubiquity of thetelephone. Many print-based distanceeducation programmes use audio confer-encing to help motivate students, and ithas also been used for small group col-laborative work at post graduate level[4]. Nevertheless, there are few uses ofthis technology in group discussion mode(as opposed to simple student to tutorphone calls) in international programmesof distance education. Audio conferencesare difficult to manage with more thanhalf a dozen sites, although it is possibleto increase the number of participants byhaving groups of students at each site.

An extension of pure voice interaction isaudiographics: voice plus a shared screenfor drawing or sharing pre-preparedgraphics. Whilst this technology has hadmore extensive use in distance education,and examples of it being used betweentwo sites in different countries do exist, Iknow of only one international, multi-siteuse in an educational application [16]. Aswith audio conferencing, audiographicsuse with more than two sites requires anaudio bridge to connect all the linestogether. There is no technical barrier todoing this internationally, except cost.Some systems use the Internet to carrythe data; others combine both voice anddata on an ISDN line.

ISDN, which stands for Integrated Ser-vices Digital Network, is a set of interna-tional switching standards to whichworld-wide telecommunications provid-ers are recommended to adhere. How-ever, to date there is no universal agree-ment regarding the standards. As an inte-grated digital network, it can be used formore than one service, and in most edu-cational contexts, this involves telephony(voice) and data (graphics or movingimage).

Audio over the Internet is a developingtechnology which a number of educa-tional institutions are trialing. RealAudio,for example, is a product which allowsreal time lectures on a global scale(http://www.realaudio.com/), but interna-tional Internet connections are too slowto permit multi-way audio interaction.Many distance education systems haveinvolved sending audio cassettes out tostudents through the post. With audio onthe Internet products like RealAudio, it ispossible for large numbers of studentsaround the world to access these ‘broad-casts’ in real or delayed time.

The UKOU is developing the use ofRealAudio to deliver a global lecture anddiscussion session with a series ofexperts. The interface which has beendeveloped on the Web to support theevents, is called the KMi Stadium, andprovides facilities for slides associatedwith the lecture. Plans are in hand todevelop a range of other tools to providea greater sense of a global audience: indi-cations of how many concurrent partici-pants there are, buttons to show agree-ment or disagreement with the lecturersideas etc.

KMi Stadium is an experiment in verylarge scale telepresence. We areenhancing existing media and develop-ing new media intended to give partici-pants a sense of ‘being there’ at eventsof all kinds, including master classes,performances, tutorials, conferences,workshops, ceremonies, parties, jamsessions, recitals, industrial trainingsessions, university lectures, trainingon demand, town hall meetings,debates, and so on. For us, ‘beingthere’ is not primarily about VirtualReality per se, although VR can cer-tainly help. Rather, it is a question ofcapturing the right participativeaspects of audience presence (such asapplauding, laughing, shouting, askingquestions, whispering to neighbours)and harnessing those aspects to convey


as much of the mood of an event aspossible. We are interested in telepres-ence at both live events and on-demand replays, because we believethat both types of event are enhancedby a sense of the presence of others.(Eisenstadt, http://kmi.open.ac.uk/sta-dium/)

Videoconferencing

Some educators feel that video is notnecessary in supporting students at a dis-tance, and that audio, especially audio-graphics works better because it concen-trates attention on content rather than dis-tracting the learners’ attention with thevisual image of the speaker. For them,the significantly higher cost of providingvideo is not justified by the educationalbenefits. Others feel that we live in avisual age in which it makes no sense torestrict the learner to audio exchange.Video, when well used, contributes to themotivation of the student, makes thelearning environment more social, andfacilitates the delivery of exceptionallearning materials in almost every area ofthe curriculum (see [15]).

One way video (with two way audio)systems have widespread application inNorth American distance education andtraining in national and internationalcompanies. Many of these systems usesatellite delivery to extend coverage. Theeducational paradigm of most pro-grammes is the lecture at a distance, withstudents watching either from smallercolleges, in the workplace, or (most com-monly) later on recorded video at home.A very great deal of distance training iscarried out by videoconferencing, someof it internationally.

Two way videoconferencing over ISDNis beginning to take over from the oneway systems, and a number of multi-siteapplications are in use, for example, inAustralia [14]. Global interoperability ofISDN systems is still problematic due tothe lack of agreement about standards,but international point-to-point and evenmulti-point videoconferences do takeplace daily in the workplace and are rela-tively frequent occurrences amongst edu-cational institutions, although usually notdirectly for teaching. As with internation-al audio conferencing, the barriers toextensive use are primarily financial, nottechnological.

PictureTel has designed a videoconfer-encing system called Socrates specifi-

cally for the education market(http://www. picturetel.com/). Socrates isan integrated presentation station with atouch sensitive screen which allows thelecturer to see, on one window of thelectern, exactly what is being shown tostudents locally and remotely, and onanother window, to preview the visualaids prepared for the lecture or to browsethe remote sites.

Streaming video on the Internet is anoth-er developing technology, technicallyfeasible today, but restricted by the band-width available internationally. In theorythis allows the image to be downloadedby remote sites in real time. What ismore realistic today, is video clips inte-grated with text-based material, to illus-trate and highlight, rather than to deliverlarge amounts of course content.

The Web

There is little doubt that the Web is themost phenomenally successful education-al tool to have appeared in a long time. Itcombines all the media described above:text, text-based interaction, audio andvideo as clips, and, with somewhat lessrobustness, multi-way interactive audioand video. Its application in global edu-cation is unquestioned. Although accessto the Internet is hardly universal, andlarge segments of the global populationare more remote from access to it(whether through cost, or through un-availability at any cost) than they are toprint and post based systems of distanceeducation, nevertheless, vast numbers ofpeople world-wide do have access, manyfrom their home, and this access is grow-ing exponentially.

The Web is merely a collection of proto-cols and standards which define access toinformation on the Internet. The threedefining characteristics of the Web are:

• the use of URLs (Universal ResourceLocators) which provide the address-ing system

• the HTTP standard (Hypertext Trans-fer Protocol) by which the delivery ofrequested information is transacted

• the development of HTML (HypertextMarkup Language) through whichlinks between documents and parts ofdocuments are made.

Fundamental to the nature of the Web isits client-server architecture, whereby theclient (Web software residing on the

user’s machine) requests a particular doc-ument from a Web server (a programrunning on a computer whose purpose isto serve documents to other computersupon request). Having transmitted thedocuments, the server then terminates theconnection. This procedure allowsservers to handle many thousands ofrequests per day.

Exciting new Web developments whichadd features useful for students and func-tions appropriate for course delivery areannounced regularly. Some examplesinclude:

• functionalities for supporting anima-tion effects, thus making documentsmore dynamic

• facilities for handling forms, and inputto forms (useful for course registration,evaluation, and other kinds ofquestionnaires)

• support for presenting tables, footnotesand mathematical symbols

• features to simplify page developmentand editing

• improvements to the search facilities

• greater functionality to the communi-cation systems linked to Web pages.

I have referred to the possibilities ofaudio and video on the Internet. One ofthe mechanisms for implementing this isthrough the programming language call-ed Java and the HotJava browser:

HotJava interprets embedded applica-tion <APP> tags by downloading andexecuting the specified program withinthe WWW environment. The specifiedprogram can be an interactive game,animation or sound files, or any otherinteractive program. Also, when a fileor application requires particular view-ers, such as for video, Java anticipatesthis and calls these viewers up auto-matically. ([6], Interface 3)

Now that it is possible to download pro-grams from the Web along with data, andto receive the appropriate software tohandle the program automatically, theinstitution wanting to deliver coursematerial can manage the maintenanceand updating of any software required forthe course. The student no longer needsto struggle with the installation of mass-ive packages, and, furthermore, can use arelatively low cost machine – even aportable.


Synchronous versus asyn-

chronous distance educa-

tion

As is apparent in this description of dif-ferent technologies for delivering educa-tion at a distance, some of the systemsrely on real time interaction, while otherscan be accessed asynchronously. Thisdifference has major implications for thedesign and delivery of distance educa-tion, as well as for the study require-ments of the learner. There are advan-tages to both forms and in the end,personal learning styles and the largereducational context determine what ismost appropriate.

Collis [6] identifies four basic patterns ofcommunication in the learning environ-ment:

• telling, which in the asynchronousmode has traditionally been the printedtext, but increasingly is taking on anew form in hypertext Web pages, al-though many conventional linear texts,articles, reports and original works arealso available on the Internet

• asking, which can take place throughtext messages via email or computerconference, through real time text chatsystems, or through any of the audiosystems

• responding, which also is supported indelayed time through asynchronoussystems, and much more immediatelythrough synchronous systems

• discussion, or collaborative workamongst small groups of students,which can take place over an extendedtime period through computer confer-encing, or for much shorter periods viaaudiographics.

The following list details the major bene-fits of each mode in an educational con-text.

Asynchronous delivery

There are four crucial advantages to theasynchronous media and I have arrangedthem in descending order of significance:

• flexibility – access to the teachingmaterial (e.g. on the Web, or computerconference discussions) can take placeat any time (24 hours of the day, 7days a week) and from many locations(e.g. oil rigs)

• time to reflect – rather than having toreact ‘on one’s feet’, asynchronoussystems allow the learner time to mullover ideas, check references, referback to previous messages and takeany amount of time to prepare a com-ment

• situated learning – because the tech-nology allows access from home andwork, the learner can easily integratethe ideas being discussed on the coursewith the working environment, oraccess resources on the Internet asrequired on the job

• cost-effective technology – text basedasynchronous systems require littlebandwidth and low end computers tooperate, thus access, particularlyglobal access is more equable.

Synchronous delivery

There are four equally compelling advan-tages to synchronous systems, although Iam less confident of general agreementabout the order:

• motivation – synchronous systemsfocus the energy of the group, provid-ing motivation to distance learners tokeep up with their peers and continuewith their studies

• telepresence – real time interactionwith its opportunity to convey tone andnuance helps to develop group cohe-sion and the sense of being part of alearning community

• good feedback – synchronous systemsprovide quick feedback on ideas andsupport consensus and decision mak-ing in group activities, both of whichenliven distance education

• pacing – synchronous events encour-age students to keep up-to-date withthe course and provide a discipline tolearning which helps people to priori-tise their studies.

There are many distance teaching pro-grammes which are entirely asyn-chronous (for example, those using printplus computer conferencing, or thoseusing the Web for both course deliveryand interaction), and others which are(almost) entirely synchronous (for exam-ple, those using videoconferencing fordelivery and interaction). However, thetrend is very much towards combiningsynchronous and asynchronous media inan attempt to capitalise on the evidentbenefits of both modes. The various per-mutations of media use and the amount

of synchronous interaction included arealmost as varied as the number of institu-tions providing distance education. Hereare just a few examples:

• University of Twente, the Netherlands,an advanced level course about tele-learning developed and taught by Dr.Betty Collis, consisting of some face-to-face meetings as well as extensiveuse of the Web for resource material,for collaborative activities and for dis-cussion. In 1996, for example, thecourse had two students participatingfrom outside the Netherlands, and all33 students operating in their secondor third language. (http://www.to.utwente.nl/ism/online96/campus.htm)

• The Open University of Catalonia,Spain currently offers courses to over200 students in the Catalan region,beginning with programmes in busi-ness studies or educational psychol-ogy. Delivery mechanisms includeprint, multimedia CD-ROMs, videosand tapes; support media include elec-tronic mail and study meetings heldtwice a semester in resource centreslinked up by fibre optic cable forvideoconferencing. (The FinancialTimes, October 3, 1995)

• Nova Southeastern, Florida, has anM.S. and Ed.D programme in instruc-tional technology and distance educa-tion, using electronic mail and bulletinboards, and audio conferences andface-to-face ‘summer institutes’. Theyclaim to have graduates in 50 states ofthe US and 35 foreign countries.(http://alpha.acast.nova.edu/)

• George Washington University, Wash-ington D. C. is a member of the MindExtension University (ME/U), a cabletelevision network dedicated to dis-tance education. The universities affili-ated with ME/U offer undergraduateand graduate courses in a range of sub-jects, using cable and satellite tele-vision and text-based asynchronousconferencing. Students outside the USare sent video recordings. (http://gwis.circ.gwu.edu/)

• The Fuqua School of Business, DukeUniversity, North Carolina, offers aglobal executive MBA using multipleinternational programme sites for resi-dential meetings, as well as email, bul-letin boards and streaming audio onthe Web. Desktop videoconferencingand CD-ROMs are also used occasion-ally. (http://www.fuqua.duke.edu/pro-grams/gemba)


• CALCampus, New York is an interna-tional on-line learning centre offeringhigh school and vocational coursesthrough the Internet. Two formats areprovided: directed independent study(using text-based material sent elec-tronically or by post) and live format(using real time chat systems). (http://www.calcampus.com)

• School of Industry and Technology,East Carolina University, has an M.Sdegree in Industrial Technology usingreal-time chat, email, listservs, theWeb and CUSeeMe interactive video.(http: //ecuvax.cis.ecu.edu)

• The Virtual College, New York Univer-sity, offers a small number of coursesusing digital video and video for Notessoftware from Lotus, which can beviewed in real time or downloaded forlocal storage and viewing. [24]

• The Sloan School of Management,MIT, Massachusetts has combinedsatellite videoconferencing (to Singa-pore and China) and Web materialsand email communication with MITstudents and faculty. [10]

• OnLine Education, University of Pais-ley, Scotland, supplies course materi-als both in hard copy and on a com-puter, which along with a modem andprinter are delivered and set up in thestudent’s home. Live events take placeboth face-to-face and by teleconfer-ence, and students (in Hong Kong)communicate with their local tutor andwith students in Scotland by email.(http://www.online.edu/online/online.htm)

• Birkbeck College, part of London Uni-versity offers a global Web-basedcourse on the Principles of ProteinStructures which includes the use of aMOO to provide real-time interaction.(http://www.cryst.bbk.ac.uk/PPS/index.html)

It would be difficult to overstate theimportance of flexibility to most distanceeducation students. The pressures ofmodern life are such that most peopledemand programmes which allow themto fit their studying in and around manyother commitments. Nonetheless, it isequally the case that most people findsynchronous events very beneficial fortheir learning. Obviously synchronousevents raise even more problems inglobal education than in asynchronousdistance education, because of the vastdifferences in time zones world-wide.

Even though global videoconferences areheld with some participants accessing at‘unusual’ times, other solutions involvesomething which might be called ‘fastasynchronous’. For example, the Univer-sity of Phoenix operates an on-line text-based interactive system which is tightlystructured to provide daily flexibility butpacing and feedback similar to syn-chronous teaching:

Each of your class meetings and theaccompanying class discussions willbe spread out over a full seven days,giving you the weekends to get muchof your reading and papers completed.As discussions build throughout theweek, it is important to visit yourclassroom at least five days out ofevery week, but at times you chooseand that best suit your individual cir-cumstances. Before each week of classbegins, your instructor will typicallysubmit a lecture and review the assign-ments for the upcoming week. Then,throughout the week, he or she will beinvolved in the class discussions – pro-viding expertise, guidance, feedbackand answers to questions.

At the conclusion of each week, youwill be asked to provide a summary ofthe concepts covered. Based upon yoursummary and contributions to the classdiscussions, your instructor will letyou know how you are doing andrespond to any issues or concerns youmight have. ([7] p. 328)

Another solution might be called ‘pan-synchronous’ to describe global eventssuch as the KMI Stadium lectures, whichmany will access synchronously, yet oth-ers can access individually or formgroups to participate in on-demandreplays anytime thereafter.

Merging synchronous and

asynchronous learning

Much of the above discussion impliesthat learning itself happens syn-chronously or asynchronously. In fact,what we have been describing are tools,materials and events which aid the learn-ing process. What is at issue is how tomeet students’ needs, both for consider-able flexibility in when and from wherethey access course material, as well asfor the galvanising power of real timeevents to keep them on track. The fol-lowing scenario by Collis provides avision of how technologies are converg-ing to satisfy both these needs:

With convergent technologies, thismeans that the subscribers, throughtheir respective learn stations, may beable to see and talk to each other inreal time; to review a real-time discus-sion that occurred earlier and perhapsmake a non-real time follow-up contri-bution; as well as to participate infamiliar asynchronous computer con-ferencing discussions. And even thefamiliar asynchronous computer con-ference will benefit from the possibil-ity of multiple representational formsthrough the learner station – the tele-learner may choose to send her contri-bution by text, by voice, or by videoclip; any combination of forms can bestored as a incoming contribution to anasynchronous discussion, waiting to beread, heard, or watched when the otherparticipants come to participate in thediscussion. ([6] 10–10)

What is so remarkable about the Web,and undoubtedly accounting for its popu-larity with such a diversity of users, is itscapacity to bring together a range of oth-erwise disparate technologies, opportuni-ties for designing courses, and competingproviders of resources for learning. Itsversatility can be summed up in thenotion that anyone can publish andbroadcast on the Web and thus reachlarge numbers of intended and unintend-ed receivers. Users can choose to accesslearning materials, to communicate withfellow learners or to prepare their ownpersonal pages. It supports real time per-sonal interaction with its high telepres-ence through visual and auditory connec-tion, yet it also provides outstandingfacilities for asynchronous resource shar-ing and communication.

Regardless of how technology evolves inthe foreseeable future, the Web has defi-nitely set the benchmark against whichany new media for global education willbe measured.

Global technologies here

and now

If the Web is so powerful and so success-ful, and so appropriate for global educa-tion, why are other media still beingused? I will outline a range of reasonswhich apply to current conditions:

Inertia

Many institutions have been using othermedia for a number of years, have natu-


rally built up professional expertiseamongst the teaching staff, have develop-ed courses to work with those media andhave attracted a market to fill thosecourses. These are persuasive reasons forremaining with the status quo, while theWeb technologies begin to stabilise andthe early adopters have begun to developsome indicators of best practice. Thesearguments apply particularly to the largenumbers of organisations using computerconferencing. What will persuade themto move to Web-based courses is a con-ferencing system integrated with Webpages and offering the full range of facil-ities they have relied on in current propri-etary computer conferencing systems.While Web-based interaction, similar tocomputer conferencing, is already avail-able, the ‘look and feel’ of these systemsis much less friendly and conducive tointimate, collaborative, and small groupinteraction than current conferencing sys-tems. This situation will definitely havechanged within 12 to 18 months, and theadded advantage of using the Web fordelivery of course material, is bound toprevail.

Approach to teaching

Studies show that the approach to teach-ing of those currently offering on-linecourses by computer conferencing is verymuch a student-centred one [2]. Theenvironment of on-line text-based inter-action, especially in asynchronous mode,lends itself to a facilitative teaching role,rather than to a didactic or authoritarianmodel. On-line courses usually centrearound collaborative activities, simula-tions and problem-based learning. Thisparadigm of guided discovery learning isthe most obvious direction for Web-based courses as well, because of thecombination of interactive multimediacapability and the many-to-many com-munication facilities, most of which willbe used in asynchronous mode for sometime to come. Consequently, the earliestmigration onto the Web will be fromthose currently running on-line coursesvia computer conferencing, and perhapsto a lesser extent via real-time chat sys-tems.

Most of the courses currently operatingover videoconferencing systems arefounded on an expository approach toteaching. One of the key advantages ofthis medium particularly to large-scaleadopters is that it does not disturb the tra-ditional model of lecturers on the face-to-face teaching campuses [15]. While the

introduction of remote students at othersites does have some effect on the teach-ing strategy (the need to acknowledge theremote sites while lecturing, the extrawork involved in preparing visual teach-ing materials) and requires administrativeand technical support in managing stu-dents at other sites, on the whole thedelivery and presentation of the coursematerial need not alter. Translating a lec-ture-based course to the Web, evenassuming that clips of lectures wereincluded, would require a significant re-engineering of the course, from an essen-tially didactic model to a resource-basedmodel. Although many lecturers willwelcome the opportunity to make thistransition from videoconferencing to theWeb, some of those in areas of the cur-riculum which are based on the accumu-lation of factual information and under-standing, will find their old approaches toteaching much more difficult to adapt.

Development costs

Apart altogether from entrenchedapproaches and valued expertise, there isthe cost of developing Web-basedcourses. While it is common to hear aca-demics talking glibly about putting theirlecture course on the Web, and it is sadlyall too common to find teaching materi-als designed for a different deliverymedium, ‘dumped’ on the Web, in fact itis generally accepted that good Webmaterials need to be tailor made for themedium, and should exploit at least thegraphical if not the audio and video capa-bilities of the medium. Course contentneeds to be re-thought for the hypertextstructure, for the possibility of collabora-tive group work, and for the opportunityof interaction with the course materials.As with CD-ROM, this requires a teamapproach, because the skills demandedrange from content expertise and distancelearning educational technology, tographical design and broadcast skills.

It will take time for sufficient resourcesto be accumulated to produce high qual-ity programmes on the Web across arange of disciplines. Nevertheless, manyof these are in progress, and new coursesare being offered on the Web every day.

Administering global courses

Systems such as PerformTrac from Com-puter Knowledge International, havebeen developed to provide basic adminis-trative functions for computer ortelecommunications-based courses, par-ticularly those in the training field using

CBT (computer-based training). In addi-tion to on-line registration, schedulingand testing, the system allows the institu-tion to determine how many trainees areaccessing the CBT courses, how muchtime each individual spends workingthrough the material and whether theypassed the test at the end.

The kind of tracking which is importantin the education sector, is slightly differ-ent. Tutors using text-based interactivesystems need to know how often theirstudents are logging on, whether they arecontributing regularly, and who has readparticular messages. This information isuseful both on a daily basis throughoutthe course, and at the end to assess theoverall success of the course. If studentsare being marked on the quality of theircontributions to discussions, it is impor-tant for the tutor to be able to call up allthe inputs of an individual student. MostWeb-based conferencing systems do notprovide this kind of support; some pro-prietary systems do have at least some ofthese features. Email and listservs ofcourse have none of these.

Conclusions

I have provided an overview of the pre-sent state of global education and dis-cussed the delivery media currently inuse on distance education courses, divid-ing them into four categories accordingto whether they are based on text, audioor video. I have described the ways inwhich the Web integrates all three, soacting as the fourth category.

There is no perfect medium however, andI have tried to highlight the primaryadvantages and disadvantages of thesesystems, based on the often conflictingneeds of learners.

While it is still appropriate to talk aboutsynchronous and asynchronous systems,this difference is falling away, as newdevelopments aim to integrate the best ofboth worlds. The Web is the main tech-nology to which all the others are con-verging, and its success derives from thefact that it can be ‘all things to all peo-ple’. Except of course, if you do not haveaccess or are never likely to be able toafford it. There will continue to be aplace for ‘lower tech’ systems such ascomputer conferencing and electronicmail, and the length of this perioddepends on how the Internet continues tobe funded in the future. Equally, there


will continue to be a place for distancetaught courses, operating globally, butnot using any form of computer ortelecommunications technology. All ofthese technologies, even those which useunsophisticated computers or low band-width, are less flexible, more costly andmore complicated than the technology ofthe book and ‘snail mail’.

References

1 Anderson, T, Mason, R. Internationalcomputer conferencing for profes-sional development : the Bangkokproject. The American Journal ofDistance Education, 7, (2), 5–18,1993.

2 Berge, Z. Personal communication,study to be published early 1997.

3 Birkerts, S. The Gutenberg elegies :the fate of reading in an electronicage. New York, Fawcett Columbine,1994.

4 Burge, E, Roberts, J. Classroomswith a difference : a practical guideto the use of conferencing technolo-gies. Toronto, The Ontario Institutefor Studies in Education, 1993.

5 Collins, M, Berge, Z. Student eval-uation of computer conferencing in a(primarily) audioconferencing dis-tance learning course. In: Inter-nationalism in distance education :a vision for higher education. M.Thompson, (ed.). Pennsylvania StateUniversity, PA, 1996. (ACSDEResearch Monograph No. 10.)

6 Collis, B. Tele-learning in a digitalworld : the future of distance learn-ing. London, International ThomsonComputer Press, 1996.

7 Corrigan, D. The Internet university :college courses by computer. USA,Cape Software Press, 1996.

8 Edwards, R. Different discourses,discourses of difference : global-isation, distance education and openlearning. Distance Education, 16, (2),1995.

9 Evans, T. Globalisation, post-Fordism and open and distance edu-cation. Distance Education, 16, (2),1995.

10 Feller, G. East meets West – Online.Internet World, March, 48–50, 1995.

11 Field, J. Globalisation, consumptionand the learning business. DistanceEducation, 16, (2), 1995.

12 Gayol, Y. The role of culture in theintegration of distance education : theMexican perspective. In: Inter-nationalism in distance education :a vision for higher education. M.Thompson, (ed.). Pennsylvania StateUniversity, PA, 1996. (ACSDEResearch Monograph No. 10.)

13 Harasim, L et al. Learning networks.Cambridge, Mass., MIT Press, 1995.

14 Latchem, C, Mitchell, J, Atkinson, R.ISDN-based videoconferencing inAustralian tertiary education. In:ISDN application in education andtraining. Mason, R and Bacsich, P(eds.). London, The Institution ofElectrical Engineers, 1994.

15 Mason, R. Using communicationsmedia in open and flexible learning.London, Kogan Page, 1994.

16 Mason, R. Evaluation report oncourses over JANUS – vol. 1. Brus-sels, European Commission, 1994.(Delta Deliverable.)

17 Mason, R. Using electronic network-ing for assessment. In: Open and dis-tance learning today. Lockwood, F(ed.). London, Routledge, 1995.

18 Mason, R. Old world visits new.Innovations in Education and Train-ing International, 33, (1), 68–69,1996.

19 Moore, M. Is there a cultural problemin international distance education?In: Internationalism in distance edu-cation : a vision for higher educa-tion. Thompson, M (ed.). Pennsylva-nia State University, PA, 1996.(ACSDE Research Monograph No.10.)

20 O’Donnell, J. Teaching on theInfobahn. In: The Internet university: college courses by computer. Corri-gan, D (ed.). USA, Cape SoftwarePress,1996.

21 Open University. International strat-egy review. Internal Memorandum,1995.

22 June 1995. An international WWWcourse. OLS News, 1995.

23 Poster, M. The mode of information :poststructuralism and social context.Cambridge, Polity Press, 1990.

24 Reinhardt, A. New ways to learn.BYTE Magazine, March, 1995.

25 Spender, D. Nattering on the Net :women, power and cyberspace. Mel-bourne, Spinifex, 1995.

26 Talbott, S. The future does not com-pute. Sebastopol, CA, O’Reilly,1995.

27 Tait, A. From a domestic to an inter-national organization : the Open Uni-versity UK and Europe. In: Interna-tionalism in distance education : avision for higher education. Thomp-son, M (ed.). Pennsylvania State Uni-versity, PA, 1996. (ACSDE ResearchMonograph No. 10.)

28 Trindade, A. Globalization of dis-tance education : setting a trans-Atlantic policy for collaboration. In:Internationalism in distance educa-tion : a vision for higher education.Thompson, M (ed.). PennsylvaniaState University, PA, 1996. (ACSDEResearch Monograph No. 10.)

29 Umbriaco, M, Paquette, D. Coursesand graduate program developmentin DENAID (Distance Education,National and International Develop-ment). In: Internationalism in dis-tance education : a vision for highereducation. Thompson, M (ed.). Penn-sylvania State University, PA, 1996.(ACSDE Research Monograph No.10.)


82

The future of learningB Y A N T H O N Y W . B A T E S


It is argued that the rapidly changingcontext of adult learning requiresequally rapid and radical changes inteaching methods and methods of de-livery, with profound implications forthe organization of post-secondaryinstitutions.

The multimedia WWW-presentation1

on which this paper is based givesexamples of some of the new teachingstrategies that are being developedwhich are moving in the directionsneeded. As the technologies converge,and as access to computers and high-speed networks increase, it willbecome increasingly possible to pro-vide learners with any course, any timeand anywhere they need it.

It is critical for government, employersand educational institutions to worktowards this goal, because in an agewhere knowledge and information isthe basis of economic growth, thosesocieties that successfully harnessinformation technologies to the learn-ing process will become the economicleaders of the 21st century.

This paper was first presented at theMinister’s Forum on Adult Learning,Edmonton, Alberta, 30 November –1 December, 1995.

Alternative futures

In a speech to a Labour Party conferencein London on 17 October 1994, DavidPuttnam, the producer of ‘Chariots ofFire’ and previously head of one ofHollywood’s largest film companies, isreported as saying [1]:

“Multinational media companies arecreating education programmes whichleapfrog over schools and appealdirectly to children and their parents ...one senior executive has predicted thateducators would be paid more thanfilm stars by the end of the century ...but this international ‘edutainment’material would be driven by babycommercial values ... we risk losingout to a tidal wave of relatively lowquality and certainly low cost materialwith just enough educational content tomake it attractive to parents.”

The market analysts have targeted adulteducation and training as an even more

lucrative market. Is there an alternative toJapanese and American-originated ‘lowquality edutainment’? Is this the educa-tional equivalent of cheap textiles andcars? Does it matter if multinationalmedia organizations turn education into abusiness, and people buy their products?

While there are important roles to beplayed by private sector organizations inthe education and training of adults, thereare many reasons why a strong publicsector presence is needed in adult educa-tion. However, in order to win and main-tain public sector support, the public sec-tor will need radical changes in its app-roach to teaching and learning. This willrequire a clear and shared vision of thefuture that we want, a determined and co-ordinated policy approach, and partner-ship between government, employers andlabour, and the public sector institutions.This paper concentrates on a vision forlearning in the future, and attempts tosuggest some of the issues that need to beresolved in order to achieve such avision.

Technology change

Within the next 10 years, we will see thefollowing important technological devel-opments so commonplace that they willbe found in the majority of homes andbusinesses in most developed countries:

• Integration of computers, television,and telecommunications, through digi-tization/compression techniques

• Reduced costs and more flexibleuses/applications of telecommunica-tions, through developments such asISDN/fibre optics/cellular radio

• Increased processing power, throughnew micro-chip development andadvanced software techniques.

These developments, all available or cur-rently under development, will result in asingle, integrated entertainment/commu-nications/learning ‘box’ in each homeand business. This will provide a muchwider range of applications than nowavailable, such as television programmesand music, home shopping and banking,and education and training, all availableon demand.

Multimedia, through the integration ofhigh quality graphics, audio, video andtext, and more powerful editing and auth-oring software, provides a major en-hancement of computer-based learning.

The costs of hardware and the cost ofproducing multimedia materials are alsodropping rapidly. ‘Stand-alone’ com-puter-based learning will become evenmore powerful as artificial intelligenceand virtual reality develop.

However, while ‘stand-alone’ applica-tions of multimedia will continue to beimportant in education, a much more sig-nificant development is the application ofhigh-speed multimedia networks for edu-cational purposes. As well as the conver-gence of different media within a com-mon computer platform, we are also see-ing the convergence of the previouslyseparate technologies and industries ofcomputing, telecommunications and tele-vision. For instance, in April 1994, Sten-tor, an alliance of Canadian telephonecompanies, announced an $8 billion, 10year initiative, called BEACON, that willbring broadband, multimedia services to80 % – 90 % of all homes and businessesin Canada by the year 2004.

The implications for education and train-ing are immense. Learning can be inde-pendent of time and place, and availableat all stages of a person’s life. The learn-ing context will be technologically rich.Learners will have access not only to awide range of media, but also to a widerange of sources of education. However,the speed and extent to which these newtechnologies are being developed andapplied is both revolutionary and deeplychallenging to established educationalinstitutions.

The need for vision

“If you don’t know where you’regoing, any road will do.”The White Rabbit, in ‘Alice in Won-derland’ (Lewis Carrol)

The way educational technology is devel-oping somewhat resembles the freneticactivities of the White Rabbit in ‘Alice inWonderland’; because it does not seemto know where it is going, any road willdo. Educational technology applicationsthough should be driven by our vision ofeducation and training in the 21st cen-tury. It is necessary to develop a sharedvision of what we want the education andtraining system to achieve. That visionshould take into account the potential oftechnology, but not be driven solely bywhat is possible technologically; what itcan do may not be what we want it to do.

1 http://bates.cstudies.ubc.ca


The need for change

Those from the business sector are wellaware of the global economic changesthat are affecting our societies. One ofthese changes is the need for a muchmore highly skilled workforce to remaineconomically competitive and to sustaina prosperous society based on highwages [2]. This is making the training ofthe existing workforce a high priority,and this training must be continuedthroughout a person’s lifetime. Invest-ment in training is or will become asessential for company survival as capitalor plant investment [3].

It is hard to quantify the need for ‘work-force’ training, but assuming that a per-son will need to re-train at least fivetimes in a working life-time, and thatsuch re-training requires the equivalentof three months full-time learning (prob-ably a gross underestimate), then thecapacity of the current education andtraining market, public and private, needsto be doubled [4].

This is because of increased demandfrom two sources. The first is fromyoung people continuing into post-sec-ondary education. This demand will con-tinue to increase slightly in most deve-loped countries (between 2 % to 5 % perannum for another 10 years at least) asmore and more young people realise theimportance of further education for theirfuture prosperity. However, the majorincrease in demand will be from all of usin the workforce who need to continuelearning if we are to stay employed, andif our employers are to remain economic-ally competitive.

The requirements of this new market forlearning are very different from those ofthe youngsters the system has tradition-ally served. Much of the learning in theworkforce market will be initiated byindividuals as part and parcel of theirworking and leisure lives. It will beinformal (i.e. not leading to any formalqualification), self-directed, and piece-meal (broken into small chunks of learn-ing, some as small as a few minutes aday, to some several hours in length). Itwill be driven as much by short-termneeds as by any conscious plan of study.Thus, it will not be determined by somemaster instructor, but by the task at hand[5].

To see just how prevalent this kind oflearning is already, just ask yourself how

you have learned to use a computer. Howmuch of this was due to formal training,with an instructor, and how much waspicked up piece-meal by trial and error,with a bad manual, and help from coll-eagues? This is not to say that the learn-ing would not have been much moreeffective if it had been structured anddirected all through by a skilled tutor, butwhat drives such learning is not the con-trol of an instructor but the needs and themotivation of the learner.

Employers, however, cannot rely entirelyon the self-motivation of their em-ployees. This kind of learning, althoughwidespread and surprisingly effective, isnot efficient. Management consultantshave identified that most companies useless than 10 % of the capacity or capabil-ities of the computer technology in whichthey have invested. In blunt terms, mostemployees do not know how to use theirmachines properly. As companies in-creasingly look to higher skill levelsfrom their employees, not just in com-puter skills, but in the broader skills ofverbal and written communication, prob-lem-solving and management, theemployers themselves will need to takegreater responsibility for making thiskind of learning more effective, both tokeep valued staff, and to increase pro-ductivity.

Traditionally, large companies have donethis by establishing their own trainingcentres and programmes; small andmedium sized companies have reliedmore on ‘out-sourcing’ the training, toprivate training companies or to the pub-lic sector institutions. All of these meth-ods, however, are labour-intensive, andany increase in such activities will lead toproportional increases in cost, at a timewhen companies need to be more costcompetitive.

While employers will be faced with theneed to keep their current workforcehighly skilled and up-to-date in the fieldin which they are working, they will notbe able to rely on the public sector toprovide, free of charge, the training ser-vices they require, for several reasons. Inmany countries, the public sector institu-tions still concentrate mainly on pre-ser-vice education and training. Furthermore,governments in many developed coun-tries are in serious financial difficulties;they have borrowed too much. Merelyservicing the debt is taking away fundsthat otherwise could be used for publicservices such as education and training.

At the same time, these governmentshave reached the limit of tax elasticity;the public is unwilling to increase theamount they pay in taxes. In other words,governments are expected to do morewith less. Consequently, many publicsector institutions are looking on privatesector training as a cow to milk, ratherthan to feed.

Lastly, many employers and members ofthe public are growing increasingly criti-cal about the quality of education beingprovided through the public sector. Thereseems to be a mis-match between theskills taught and the requirements of thelabour market [6].

In some ways, this is an unfair criticism,since educational attainment of studentsin public schools has increased over thelast 20 years; the problem is that thedemands on the workforce have increas-ed at an even faster rate [7]. For instance,production line workers need greaterliteracy skills today to deal with writteninstructions, manuals, etc.; they nowneed more than just an arm and a leg tooperate the production machinery; theyneed intellectual skills as well. Similarly,with greater emphasis on teamwork andworker involvement and motivation, thelevel of communication and social skillsrequired from managers and supervisorshas increased. While a great deal ofattention is being paid to the gap betweenthe skills of those entering the workplaceand the needs of employers, less attentionis being paid to the much wider gap be-tween the skills of those already in theworkforce and the demands of the work-place. For instance, the older the worker,the lower the functional literacy level inmost developed countries.

Another area which needs more attentionis the gap between the way educationalservices are provided, and the needs ofemployers and working people. Workingpeople are unable or cannot afford togive up jobs or move house to becomefull-time or even part-time campus-basedstudents again.

Jobs are changing

While countris like Canada and Norwaywill remain dependent on primaryresource industries for their economicwell-being, the sources of employmentare rapidly changing, due to increasedmechanization, the need to diversify to‘value-added’ secondary industry (e.g.


furniture, paper), and the growth of newindustries and services not directlydependent on the primary resource indus-tries, such as communications, informa-tion technology, financial services andinternational trading.

Most of the new jobs are either in serviceindustries, in companies employing lessthan 20 people, or require highly skilledspecialists in the larger, resource-basedindustries, each ‘new’ employee oftenreplacing many existing staff. Many ofthe new jobs are on a part-time or con-tract basis, with at least two-thirds of thenew jobs going to women, and a majorityof new jobs are relatively low-paid [8].Nevertheless, nearly half the new jobscreated require graduates or people withthe equivalent of 17 years full-time edu-cation [9].

The traditional picture of work as a life-time commitment to a particular trade orinstitution, with a secure pension at theend, applies to an increasingly smallerproportion of the population. In particu-lar, secure middle management jobs of ageneral kind, requiring little or no profes-sional or technical expertise, are dis-appearing rapidly. A very small propor-tion of the youngsters leaving school willfind employment in the traditionalresource-based industries as unskilled orsemi-skilled workers; the majority ofthose already unemployed, and a goodproportion of those already working inlarge companies or in primary-resourceindustries, need to be re-trained in thenext few years.

The most significant development is thatmany of the new jobs will require a muchhigher level of skill than the jobs they arereplacing, especially in management andresource based industries; people willretain existing jobs only if they are re-trained to higher standards; even for themajority of new jobs that will be low-paid and require generally low skill lev-els, training or re-training will be neces-sary, especially in basic skills, just inorder to keep the job.

With respect to the skills needed in theworkforce, these have been well definedby the Conference Board of Canada [10]:

• Good communication skills (reading/writing/speaking/listening)

• Ability to learn independently

• Social skills: ethics; positive attitudes;responsibility

• Teamwork

• Ability to adapt to changing circum-stances

• Thinking skills: problem-solving; criti-cal/logical/numerical

• Knowledge navigation: where to get/how to process information.

Learning in the 21st

century

Modern learning theory sees learning asan individual quest for meaning and rele-vance. Once learning moves beyond therecall of facts, principles or correct pro-cedures, and into the area of creativity,problem-solving, analysis, or evaluation(the very skills needed in the work-placein a knowledge-based economy), learnersneed inter-personal communication, theopportunity to question, challenge anddiscuss. Learning is as much a social asan individual activity. However, forsomeone working in a small company,the nearest person with similar interestsand expertise may be somewhere on theother side of the country, particularly inleading-edge technologies.

Learners will interact with their desk-topor portable workstations in a variety ofways, determined by the nature of thelearning task, and their preferred style oflearning in the work situation. These pre-ferred styles will vary considerably, bothwithin a single person, depending on thetask, and, for the same task, between dif-ferent individuals.

The learning context will need to encom-pass the following:

• Working alone, interacting with learn-ing material (which may be availablelocally or remotely)

• Working collaboratively (and in anequal relationship) with fellow work-ers at different remote sites, either syn-chronously or asynchronously: boththese modes are likely to be multi-media

• As an ‘apprentice’ or ‘student’, work-ing with a more experienced worker,supervisor, or instructor

• As an instructor, supervisor or moreexperienced colleague for other lessexperienced colleagues.

The same persons may find themselvesin each of these roles within a single

working day. Learners will also need tobe able to work from home, or from awork-site, or while in transit. They willneed the following:

• Access to information (searching,downloading) from multiple sources inmultiple formats

• Selection, storage and re-ordering/re-creation of information

• Direct communication with instructors,colleagues, and other learners

• Incorporation of accessed/re-workedmaterial into work documents

• Sharing and manipulation of informa-tion/documents/projects with others.

Learners will need to access, combine,create and transmit audio, video, text,and data as necessary. If we take this asthe design requirement, there is then aneed to build systems that support thisform of learning, both for formal andinformal learning.

New models of teaching

A number of factors are leading manyCanadian post-secondary institutions toexperiment with new information tech-nologies for teaching. There is pressurefrom governments for greater efficiency,requiring institutions to take more stu-dents, while at the same time reducingfunding. Another influence is the use bygovernments of earmarked funds forinnovation (funds often established byholding back a proportion of ‘regular’funding).

Another factor is the increase in studentfees, leading to more and more studentsbecoming part-time, in order to worktheir way through university. Forinstance, almost half the students in theuniversities, and two-thirds of the collegestudents, in British Columbia now arepart-time. Also, the trend towards life-long learning and the need for re-educa-tion and training for people already in theworkforce is leading to a changing stu-dent population, with many more olderstudents, working and with families,returning to post-secondary education (orin some cases never leaving it). Thesestudents need greater flexibility in theprovision of learning, to fit it aroundtheir already busy and demanding lives.

It is not surprising then that many institu-tions are now turning almost in despera-tion to multimedia technologies as one


possible solution to their increasinglypressing problems. It is from these oftenprototype developments that we will seesome of the newer models of teachingand learning developing that will meetthe needs of the changing adult learningmarket.

For instance, at the University of BritishColumbia, almost $4 million of itsoperating grant has been held back overthe last two years by the government foran innovation fund for the university.The university has used approximately50 % of this for investment in technologyinfrastructure, such as fibre optic cablingacross the campus, computer networks inbuildings, computer labs, networkedservers, and video-conferencing facili-ties, and the remaining 50 % for technol-ogy applications. Almost all the applica-tion money has been used for projects (ineach of the 12 faculties) using eitherWorld Wide Web or CD-ROM technolo-gies.

This has led to a great deal of innovationacross the 50 or so projects that havebeen funded in this way, and someimportant lessons have also been learned.Also, some lessons learned in other con-texts, such as the use of technology fordistance education, have also proved justas relevant for on-campus use of technol-ogy. I will describe briefly some of thelessons learned and some principles toapply when creating multimedia coursesand products.

Two-way video-conferencing, using eith-er compressed or full motion television,has been the major new technology usedfor delivery in many universities and col-leges across North America. One of themost technologically advanced systemsof video-conferencing operates betweenthe University of Calgary and the Uni-versity of Alberta. Video-conferencinghas been slow to take off at UBC, but in1995/96 one fourth year undergraduateplant sciences course was delivered inthis mode, interestingly with the teachingoriginating from a community college,the University College of the Fraser Val-ley, because there were insufficient stu-dents at UBC in this specialist area (tem-perate fruit management) to justify alocal teacher. Because of the excellentteaching technique and careful advancepreparation of the instructor (Tom Bau-mann), this has been a very successfulcourse. Video-conferencing has theadvantage, from an instructor perspect-ive, of not requiring a radically different

approach from regular classroom instruc-tion, and appears quick and easy tomount. However, unit costs are higherthan regular classroom teaching, it doesneed considerably more preparation time,and students still need to be at a fixedplace at a fixed time.

The biggest explosion in teaching tech-nology in the last couple of years at UBChas been the WorldWideWeb (the Web),due to the development of relativelyuser-friendly hypercard technology suchas Netscape. This enables ‘creators’ ofmaterials to store primarily text andgraphics on ‘screens’ that provide linksto any other ‘screens’ located anywhereon the Web. Thus, by merely clicking ona highlighted or coloured section of text,a link can be made to another screen onanother computer anywhere in the world.

The main use of the Web to date hasbeen for accessing (and downloading)information. However, a number of insti-tutions have started designing coursesusing Web technology. For instance, atUBC several courses were designed in1995 using Web software and localservers for on-campus students. The firstof UBC’s distance education coursesusing Web software is in ContinuingStudies, which began offering a course‘Mastering Cyberspace’, primarily aimedat people who want to use the Web forbusiness purposes. In January 1997 UBCwill deliver several Web-based creditcourses to students off-campus.

Netscape software lends itself particular-ly to subject areas where there is a needfor extensive colour illustration in stillgraphic form. Thus one of the most suc-cessful applications at UBC has been inGeology 202, where the combination of astructured approach to teaching with stu-dents able to access UBC-generated data-bases of graphics such as rock slides androck formations, together with on-lineaccess to other geology Web sites acrossthe world, has considerably enhanced theflexibility of access for students.

At this stage there has been little evalua-tion of the use of the Web for the directdelivery of distance education courses.The UBC use is primarily that of inform-ation transmission, with some self-testingthrough multiple choice questions and inone case simulations. However, at thetime of writing (November, 1996) Websoftware lacks the interactive flexibilityof dedicated computer conferencing tech-nology such as CoSy or First Class, al-

though Hypertext provides a basic con-ferencing facility. Protocols also have tobe established to ensure that only thosewho have paid their fees can accesseither the content or the interactive partof the course.

The main constraint of using the Web fordirect teaching, once one goes off-cam-pus, is the speed (or rather the slowness)at which pages can be accessed or down-loaded. It is one thing to make materialavailable on campus over high-speed net-works and high-end modems; it is anoth-er to deliver a substantial amount ofcourse materials over public telephonenetworks, through low-speed modems tothe kind of computers that students arelikely to have at home. We will be moni-toring this aspect quite carefully. It mayin many cases be better to provide thematerials on a CD-ROM, with only up-dates being made available via the Web.

Computer-based multimedia

There has been a rapid development inrecent years of computer-based learning,due to technological developments bothin the computers themselves, throughfaster processing, greater storage capac-ity and ‘embedded’ software that enablesthe integration of graphics, sound andmoving visuals from external sources,and in the development of ‘peripheral’equipment, such as CD-ROM and video-discs. This has led to the development ofcomputer-based multimedia materials(defined as textual data, sound, graphicsand moving visuals integrated into a sin-gle computer platform).

Computers have for a long time beenused for drill and practice. However, thecombination of sound output, in the formof a digitized human voice and musicaltones, recognition and analysis of humanvoice input, and feedback of results, hasenabled the UBC Music department toapply off-the-shelf commercial softwarefor the teaching of music. Students areasked to recognize either sheet music ortones generated by the computer andmatch that with their own singing input.

Off-the-shelf software, such as Super-card, can also be used for constructingteaching materials, making it much easierfor instructors to develop materials for aparticular context. There have been awide number of projects at UBC in thisarea. Mark Broudo and colleagues in theFaculty of Medicine have developed


teaching materials from scratch to assiststudents develop skills in the area ofcardiovascular measurement, using ani-mation, integrated full video and sound,and text, with multiple choice-type ques-tions providing feedback. The material isstructured in modules, with students ableto access the material when they need it.

Brian Holl, a specialist in turf grass man-agement at UBC, has developed a ‘vir-tual’ office and laboratory to allow stu-dents to diagnose and treat turf disease.Students can click on a number of ‘tools’in the office and laboratory to accessinformation. In addition, there is a virtualworkbook that allows students to entertheir own diagnoses and treatments,which is automatically transmitted to theinstructor for checking. The interfacealso includes links to the World-WideWeb, and e-mail and messagingcapabilities.

One of the more exciting uses of multi-media is the development of expert sys-tems that allow the user to test hypo-theses and simulate different conditionsand see the outcome. One such moduledeveloped by Dr. Hamish Kimmins atUBC allows users to test different app-roaches to forestry management. Themodule shows the relationship betweendifferent forestry practices in terms ofrevenue yield, sustainability, bio-diver-sity and the aesthetic features of the land-scape, and shows the impact of a particu-lar decision over the various time periodsof five, ten and forty years.

The increased user-friendliness of multi-media development tools enables eventhe learners themselves to create theirown instructional materials. Liz Ham-mond-Karreemaa, a college instructor atMalaspina College on Vancouver Island,has developed a multimedia package onkiller whales, based on local wildliferesources, in such a way that learners cancreate their own reports from the materi-als she has assembled.

There are several lessons to be learnedfrom these experiences. With the excep-tion of the video-conferencing example,the construction of learning materialsfrees the instructor from the need fordirect ‘real-time’ delivery of the teachingmaterial. Once created it can be used anytime, anywhere by the learners. Theexamples also progressively move awayfrom using the technology just for in-formation transmission. There is somesimple feedback and testing of students,

and some provide skills development, butsome of the examples, particularly theturf grass and forestry managementexamples, go further and develop prob-lem-solving techniques which studentscan practice. Brian Holl’s and Liz Ham-mond-Kaarreemaa’s approaches requirestudents to develop research skills. Someof the examples also enable students toprovide open-ended responses using nat-ural language, and all provide learnerswith much greater control over theirlearning, but within a clearly defined andstructured learning environment. Suchapproaches are particularly valuable foradult learners in the new learning con-text.

New models

However, these examples by no meansexhaust the range of new teaching mod-els needed. These examples still reflectthe early stages of a new paradigm forlearning, rather like Max Sennett’s Key-stone Cops were merely a necessary steptowards Jurassic Park.

UBC has not for instance moved yet intothe area of on-line collaborative learning,using computer mediated communica-tions such as listserves, computer confer-encing systems such as FirstClass (usedextensively at the Open Learning Agencyin BC), or the Virtual University model,based on integrating new tools with Websoftware, being developed at SimonFraser University. These models encour-age learners to work together from differ-ent sites and asynchronously on commonprojects, to construct new knowledge.

One interesting use of the Web is thedesign of courses around material public-ly available on the Web, where studentsare guided or encouraged to seek inform-ation through the Web, and incorporate itinto assignments. Some courses at theOpen Learning Agency in BritishColumbia are planning to use this strat-egy.

These technologies now enable us toconsider new approaches to teaching andlearning that meet the needs already de-fined. For instance, it is now possible tooffer students truly international courses,with faculty and other students drawnfrom across the world. Millions of dollarsare spent by many countries in sendingstudents to Canadian universities and col-leges for extensive periods. The technol-ogy now allows students to come for

shorter periods and study the rest of theprogram at home. This will allow manymore students to benefit from such inter-national programs. The benefits to Cana-dians of being able to make contacts withand understand the business practices andculture of different countries, and the dif-ferent perspectives brought to commonproblems, are obvious.

Thus, one interesting use of the Web hasbeen for networking post-graduate stu-dents in different specialist areas. Forinstance, the School of Architecture atUBC, together with Schools of Architec-ture at the University of Sydney in Aus-tralia, the University of Hong Kong,MIT, and several other universitiesaround the world, have been using theWeb to allow graduate students to pre-sent and critique each others’ design pro-jects.

Resource-based tutoring

Another model is aimed at students seek-ing accreditation within a particular sub-ject area who already have a good found-ation, but who are working towards amore advanced level of study and a per-sonally relevant area of expertise. In thismodel, the learner is put in touch with atutor with specialist expertise who canguide the learner to sources of informa-tion and pre-prepared multimedia learn-ing materials relevant to their interests. Inthis model, the tutor helps the learnernavigate remote databases, or institution-al libraries, which contain the multimediainstructional materials needed by thelearner, sets and evaluates relevant learn-ing tasks such as project work, and putsthe learner in contact with other learnersand experts with similar interests. Thismodel is not very different from learner-centred teaching practised in British pri-mary schools, except that it operates at adistance via telecommunications.

Just-in-time training in theworkplace

It is not difficult to build a convincingportrait of learning at the work-place. Wecan envisage a computer software de-signer or television animation artist, call-ed Wayne, probably working from home,needing information on a certain tech-nique or approach, or advice on how bestto create a certain effect. From previousexperience and contacts, or on the adviceof a colleague, he has the name of some-one half-way across the country (Sue).


From his work-station, Wayne calls Sue,talks about the problem, and Sue loadsup some software which she ‘shares’with Wayne via the network. Wayne asksa few questions, tries a couple of thingson-line while Sue watches and com-ments, then downloads the software. Sueand Wayne are both registered with aneducational institution that has been setup to enable the exchange of commerc-ially sensitive material for learning pur-poses. Wayne’s work-station has auto-matically displayed the cost per minuteof consulting Sue, and the cost of rightsfor downloading the software. However,Wayne was also able to give Sue someinformation, and this is charged back toSue’s account. Wayne now not only hasthe software he needs, but also can con-tact Sue (on a chargeable basis) any timehe has a problem with the software. Thelearning context has been established.Note it is fragmented, on demand, andcharged at cost.

Lastly, the whole question of delivery oflearning has been implicit in the discus-sion of new models of teaching. Adultsneed flexibility. They have families, andwork commitments around which theyhave to fit their learning. This means thatlearning should as far as possible beaccessible at any time and at any place.This means making use of delivery to thehome, to local learning centres, such asthe Community Skills Centres beingestablished in small towns in BC, in theworkplace, and into other institutions.

One project with an innovative methodof delivery in British Columbia isSkillPlan. This project was developed asa result of a partnership with constructionindustry unions. The unions found thattheir members, many of whom weredoing contract work at different sites, andoften had some spare time between jobs,were needing to up-grade their basicliteracy, numeracy and communicationskills. The Open Learning Agency identi-fied a computer managed learning systemdeveloped by the Jostens Corporation inthe USA that provided adult basic educa-tion courses aimed at adults who havenot completed high school graduation.Workers who need to improve their read-ing and writing skills can ‘drop in’ at theAgency’s local centres, and use the sys-tem when it suits them. The system keepstrack of each individual’s progress, andenables learners to carry on where theyleft off the last time they were able todrop in.

Many other new curriculum models canbe suggested (see [11], for instance). Thechallenge for educators and trainers in aworld where information is multiplyingso fast that even specialists in a particularfield cannot keep pace is to ensure thatlearners have the learning skills and thepreparation to go beyond accessing andnavigating information to making senseof that information, and applying it inways that are relevant to their lives andtheir work; in other words to movebeyond information to knowledge. Thismeans being able to construct newknowledge, by finding new ways to dothings, based on knowledge alreadyacquired, and to communicate that to oth-ers.

Many would argue that this is exactlywhat a traditional liberal arts educationdevelops in learners. Employers on theother hand argue that workers are notcoming to them with the necessary skills,and students should be trained to be morejob specific [6]. There is not necessarily acontradiction here. It can be seen from theexamples given that good teachingapproaches can develop both good genericskills and be applied at the same time.

However, it is important to note thatproblem-solving and decision-makingapproaches such as those found in theturf and forestry management examplesdepend on expert systems, in other wordsknowledge bases developed from re-search, which are then applied to a par-ticular problem. Secondly, in a worldwhere jobs change for most people everyfive years, it is extremely dangerous totrain for just the skills large companiessay are needed here and now. The newjobs are less and less with large industrialor resource-based companies, and muchmore with the small entrepreneur. Theseneed skills that allow for adaptation tochanging conditions, not narrowly focus-ed skills that may become redundant in afew years.

What is required then is a teaching andlearning environment that more closelyintegrates basic research, navigation andunderstanding of information, applicationof knowledge, and skills development.

The challenge for public

institutions

Daryl Le Grew, the Vice-President (Aca-demic) at Deakin University, Australia,has pointed out that many post-secondary

institutions “are moving to reconstructtheir infrastructure, redesign policy andrealign external relationships to gaincomparative advantage in the Informa-tion Superhighway environment”. Heargues [12] that there is a transformationtaking place (a ‘paradigm shift’) in post-secondary education, characterized byTable 1.

In particular, he argues that the new tech-nological environment “opens access tostudy across sectoral, disciplinary andcultural boundaries”, which “will quicklyerode traditional ideas of the course ofstudy”: selective access, sequenced andcarefully integrated content, and level-based progress rules that are pre-deter-mined by the institution. On the contrary,he argues, “curricula will be increasinglydisassembled, modularized and custom-ized to suit a wide range of clients requir-ing flexible delivery ... What constitutesa course will be increasingly negotiatedbetween provider institutions, studentsand client groups.”

If he is correct, this will require somemajor changes in the organization ofpost-secondary institutions.

Who’s in charge?

Managing multimedia

projects

The most common procedure adopted byfaculty when embarking on multimediaprojects is to develop a proposal, usuallyaround a particular technology or soft-ware (e.g. video-conferencing orNetscape). Once funding is received, fac-ulty order equipment and software, andfind some willing, interested and com-puter-skilled post-graduate student tohelp with the development of the com-

Industrial society . . . . to . . Information society

technology peripheral . . to . . multimedia central

once-only education . . . to . . lifelong learning

fixed curriculum . . . . . . to . . flexible/open curriculum

institutional focus . . . . . to . . learner focus

self-contained . . . . . . . . to . . partnerships

local focus . . . . . . . . . . to . . global networking

Table 1


puter-based material. This I call the LoneRanger approach, with the post-grad asTonto.

There have been several advantages inthis approach. First, it has brought a lotof faculty into exploring the use of newtechnology who would not otherwisehave done so. Secondly, by choosing acheap and (apparently) easy approach,faculty are able to make the most ofmaterials already prepared, such as lec-ture notes and student handouts, whichare then converted into multimedia mate-rials. Most important of all, faculty them-selves retain control of the project, andlearn, usually the hard way, that there ismore to multimedia design than they hadanticipated. In fact, the experience ingeneral is that faculty become excited bythe potential, underestimate the work andproblems they eventually have to face,and finally recognize the need for morehelp.

Another approach, and one used extens-ively in the private sector, is to developproject teams to create multimedia mate-rials. Thus, as well as faculty, therewould be an instructional designer, whowill help with choosing the most app-ropriate technologies, with structuringthe material and defining the teachingstrategies and learning activities to bedeployed in the multimedia materials,recommend student assessment strate-gies, and design the piloting and evalua-tion of the product.

Secondly, it is important to construct a‘screen’ environment that allows thedesigners to maximize the learning fea-tures of high quality multimedia materi-als. This means designing an interface sothat learners can interact in a variety ofways with the learning materials. Thisrequires an interface designer to workwith the subject expert and instructionaldesigner. There is also need for someonewith specialized multimedia graphicsdesign skills to make screens look pro-fessional.

Someone is also needed who has theskills to obtain video and audio record-ings, and the ability to edit and convertthese recordings into digital format, forincorporation in the learning materials.And someone is needed who can developa working script, integrate all thesedifferent elements into a well-edited andwell-constructed final format, and whocan choose appropriate software to facili-tate the various production processes.

Lastly, each project will need a manager.The project manager sets up meetings,breaks up the design and developmentprocess into discrete stages with estimat-ed amount of work and sets deadlines foreach piece of work, monitors work flowand deadlines, manages the budget, andworks on delivery and student support.The project manager may also handleliaison with external contractors and pub-lishers, and marketing activities.

The importance of teamwork

Some people will be able to combinemore than one role. However, each roleneeds a high level of expertise. Team-work, and a recognition by each memberof the team of their own limitations, andthe contribution that other members ofthe team can make, are critical tosuccess.

Project management and a team of expe-rienced specialists will initially seem likean expensive way to produce material,but in the end, it is almost always morecost-effective than just ploughing on witha multimedia project, and learning as onegoes. In particular, it can reduce the timespent by faculty in mastering skills thatthey do not necessarily need, thus freeingthem up for other activities. A good pro-ject team will also substantially reducethe time needed to complete the project,compared with a faculty member and aresearch student soldiering on alone.Lastly, if a higher quality product is pro-duced, there are more possibilities of costrecovery from other applications than theoriginal ‘targeted’ use.

Faculty often criticise multimedia devel-opment as too expensive or too demand-ing of faculty time. It is also very chal-lenging, requiring faculty to try methodsand technologies with which they are notfamiliar or comfortable. These are allvalid concerns. Multimedia teaching willbe more expensive, time-consuming andthreatening if it is merely added on to allthe other duties of faculty, within anorganizational environment that has notbeen modified to support such activities.Just as automation required majorchanges in industrial manufacturingmethods, so too is there a need to re-structure universities and colleges to en-able the benefits of the new technologiesto be realised.

Institutional change

Consequently, a number of issues need tobe addressed. Most universities and col-leges are aware of the need to put in acomprehensive technology infrastructureon campus. However, probably the mostimportant step to be taken is to providesupport to the people who will be mostaffected by technological innovation:faculty and students.

Human resource development

Bluntly, faculty need training, not just inhow to use the technology, but moreimportantly, in understanding how learn-ing takes place, and how to design teach-ing approaches based on that knowledge.Without this fundamental understandingof the teaching and learning process, it isalmost impossible to design high qualitymultimedia learning experiences.

There are at least three practical stepsmanagement can take in this regard. Thefirst is to organize regular workshops onteaching practice and the use of technol-ogy (which in some institutions willmean creating or strengthening facultydevelopment departments). The second isto provide rewards, in terms of tenureand promotion criteria, for successful,innovative teaching. Thirdly, whileworkshops are important, there is a needfor more comprehensive and systematiccourses or programs aimed at teachers ofhigher education. These are minimumrequirements. An even more radical stepwould be to require successful comple-tion of a higher education teaching quali-fication for tenure appointments; unfor-tunately such courses, if they exist at all,are not available in a manner that makesit practical for most faculty, i.e. part-timeand at a distance, even if faculty associa-tions could be persuaded into acceptingsuch a policy.

The convergence of on-campusand distance teaching

However, while such strategies areimportant, they do not necessarily pro-vide any solution to faculty overload.This has to come from re-organizing theway that teaching is provided. One of theimpacts of technological innovation oncampuses is the convergence of on-cam-pus and off-campus or distance teaching.Once materials are developed for on-campus students, they can, if designedwith this in mind, be just as easily avail-


able to off-campus learners. Also, manymore learners can access the same mate-rial, once it is created. Thus, the use oftechnology can allow for more studentsto be accommodated for the same num-ber of faculty, with scope for re-alloca-tion of resources to free up faculty for thedevelopment of teaching materials.

Furthermore, is a student registered asfull-time who opts to take most of thecourse from home a distance student oran on-campus student? Does it matter?The introduction of multimedia coursesopens up questions about the relevanceof residential and prior qualifications forentrance; if space is not a limitation, whyrestrict students who want access? Whatbecomes clear is that once courses arecreated in ways that allow for open andflexible learning, accessible anywhereand at any time, careful considerationneeds to be given as to who needs accessto a campus, and for what reasons.

Student access to technology

There are though two large assumptionsbeing made about the use of multimedia.The first is that students will have accessto computers, Internet, CD-ROM players,etc. Student access to computers is in-creasing, but there is also a need for aclear policy to help students access thenecessary technology being used forteaching and learning. While many stu-dents now arriving in universities alreadyhave their own computers (about 40 %),many do not. In any case, there is a needfor common standards; multimedia mate-rials require medium- or high-range per-sonal computers and fast networks.These are often not available off-campusor in students’ homes.

Again, there are several strategies thatcan be adopted. Institutions can negotiateleasing arrangements with suppliers, sostudents can pay a modest rental fee withthe option for low-cost purchase at theend of the leasing period. A college canensure there are ports on campus to allowstudents to plug in their own computers(much cheaper than providing computerlabs). Capital funds for buildings (and carparks) can be re-allocated to support stu-dent’s off-campus access to courses,thereby reducing the demand for campusfacilities. Institutions can form consortiato block-buy long-distance Internetaccess throughout a province or state forstudents (the Open Learning Agency inBritish Columbia has recently negotiated

such a deal, allowing students local callaccess to the Internet and OLA’s servicesfrom most parts of the province). Gov-ernments may even install or buy leasednetwork facilities for educational use, asin New Brunswick, Canada.

Comparative costs of multi-media and conventionaleducation

A more fundamental issue is the changein cost structures resulting from thedevelopment of multimedia courses. Inthe conventional higher education model,costs have tended to increase with stu-dent numbers, or quality drops. In otherwords, if the student/teacher ratio is keptconstant, more teachers are needed. Ifmore teachers are not recruited, classsizes increase, interaction with the teach-er drops, and hence quality declines. Wehave seen over the past 10 years in manycountries access to higher educationbeing increased by increasing the stu-dent/teacher ratio, or by the hiring ofyoung Teaching Assistants scarcely morethan a couple of years ahead of the stu-dents they are teaching.

However, the cost structures for pre-pre-pared multimedia materials are quite dif-ferent. They cost a good deal of moneyup-front to create, but once created, theycan be used by as many students as nec-essary with minimal increases in costs.This becomes even more pronouncedwhen the life of the course materials aretaken into account.

Original multimedia materials createdfrom scratch have a higher ‘starting’ orfixed cost than classroom teaching, butafter production, the only additional costsare the cost of making and distributingcompact discs (assuming that studentshave their own computer and CD-ROMplayer). Classroom or lecture hall teach-ing, however, increases roughly in pro-portion to the number of students. In fact,costs increase in steps, depending on theassumption about maximum class orgroup size before the level of teacher/stu-dent interaction, and hence the quality ofthe teaching, is considered to drop. Thus,as student numbers increase, extra ses-sions or classes have to be opened, andadditional teachers or instructors found.Figure 1 shows the relationship betweenthe costs of classroom teaching and thecosts of pre-prepared multimedia.

Most universities and colleges haveabsorbed extra numbers of students inrecent years by either increasing classsize, or by using more and more low-paidand inexperienced teaching assistants (orboth); in other words, the quality ofteaching has been reduced.

Quality comparisons betweenconventional and multimediateaching

The second assumption is that the qualityof interaction with pre-prepared multi-media materials is the same as betweenindividual students and a teacher. This infact is not the case. Well-designed multi-

$

Multimedia materials(assumes students have own computer)

Conventional education(fixed student/teacher ratio)

No. of students

Figure 1 Conventional classroom teaching vs. Multimedia materials


media materials can both present inform-ation and provide a large amount of theinteraction and feedback previously pro-vided by teachers. This frees up the timefor the teacher then to concentrate just onthose areas where person-to-person inter-action is critical. However, computers arenot smart enough to anticipate all thequestions, misunderstandings, and moreimportantly, original and creative outputsthat students can generate. Thus, there isstill the need for some form of provision,not only for student/teacher interaction,but even more importantly for interactionamongst students. This can be providedeither in the traditional face-to-face smallgroup seminars, or ‘on-line’ and at a dis-tance by using e-mail or computer.

Basically, with multimedia, the teachingfalls into two parts: presentation andinterrogation of information, some feed-back, and some skills developmentthrough the use of interactive, pre-prepared multimedia materials; andcounselling, student guidance, knowl-edge building, argument and creativethinking, hypothesis development andtesting, inter-personal skills develop-ment, and group work through either on-line or face-to-face group contacts. Theaim is to use the most valuable and costlyresource – that of the teacher – more effi-ciently, by allowing teachers to concen-trate their time with students on whatthey can do best, namely encouragingand responding to the great variety ofcontributions that students make to thelearning process.

Comparing cost and quality

Figure 2 not only indicates the cost ofproviding on-line student/teacher andstudent/student interaction (via computerconferencing – arrow C), but the cumula-tive cost of multimedia materials pluscomputer conferencing (arrow D – see[13] for a methodology for costing edu-cational technologies).

It can be seen in this model that forsmaller numbers of students, convention-al classroom teaching is likely to be lesscostly than multimedia and computerconferencing combined. However, asnumbers increase the new media becomeincreasingly more cost-effective.

The need for cost-benefitanalysis

It must be stressed that this is a hypo-thetical model. There is a great lack ofhard data on the actual costs, not only ofthe new media, but even of conventionalteaching in post-secondary education.Thus we do not know (a) if the modelwill hold in real-life, or (b) even if itdoes, where the critical point ‘y’ is, i.e.the number of students where the newmedia become more cost-effective. Whatis needed is at least some cost-benefitanalyses of the actual applications ofthese new technologies in post-secondaryeducation. There is no guarantee thatthese new technologies will be morecost-effective. Nevertheless, it can beseen that, in theory at any rate, the new

multimedia technologies offer the hopeof maintaining or increasing the qualityof teaching as student numbers increase,at relatively less cost than conventionalclassroom methods.

Conclusions

It is more than unfortunate that at a timewhen there is a need for major changes inthe structure and operation of the post-secondary education system, governmentsacross Canada are threatening reductionsin resources. This is unfortunate for tworeasons. Now is not the time to reduceactivity in adult education. There is amajor new market, those in the work-force, and students coming from highschools need new approaches too. Nor isthere any clear evidence that the applica-tion of new technologies and methods ofteaching will save money; indeed they arelikely to cost more, and there is certainlya substantial cost of change. The secondreason is that the threat of cuts is likely tolead to defensiveness and a retreat to ‘thegood old days’.

At the same time, many of the new activ-ities in adult learning do not need to bedirectly funded by the state. People whoare working and in good jobs, andemployers who want to keep ahead of thecompetition, get direct benefits from re-education and re-training. We are likelyto see then increasing entrepreneurshipfrom the public sector, and increasingcompetition with the private trainingsector. Indeed, competition generally islikely to increase, as the provincial andnational borders of education are power-less to stop educational offerings throughelectronic networks. For instance, thereare four out-of-province MBAs availableat a distance in BC already. The mainthreat though to the public sector institu-tions will come not from out-of-provinceuniversities and colleges, but the largemultinational media organizations, whosee adult learning as a major new busi-ness for them.

Lastly, it is easy to get hung up on thetechnology, which is exciting, challeng-ing and not without major risks and haz-ards. However, technology is not theissue. There is more than enough technol-ogy available already for us to teach inmore or less any way we like. The issue isthe changing needs of adult learners, andthe need to find new ways of teaching andlearning that will prepare them for theuncertainties of the 21st century.

$

Multimedia materials(assumes students have own computer)

Conventional education(fixed student/teacher ratio)

No. of students

Computerconferencing y

A

B

C

D

Figure 2 Conventional classroom teaching vs. Multimedia materialscombined with computer conferencing


12 Le Grew, D. Global knowledge :superhighway or super gridlock.Applications of Media and Technol-ogy in Higher Education. Chiba,Japan, National Institute of Multi-media Education, 1995.

13 Bates, A W. Technology, open learn-ing and distance education. NewYork, Routledge, 1995.

References

1 MacLeod, D. Educational films ‘willflood Britain’. The Guardian, 18October, 1994.

2 Porter, M. Canada at the crossroads :the reality of a new competitive en-vironment. Ottawa, The BusinessCouncil on National Issues/Govern-ment of Canada, 1991.

3 Reich, R. The real economy. TheAtlantic Monthly. February, 1991.

4 The Open Learning Agency. Lifelonglearning and human resource devel-opment. Burnaby, B.C., The OpenLearning Agency, 1992.

5 Weimer, B. Assumptions about uni-versity-industry relationships in con-tinuing professional education : a re-assessment. European Journal ofEducation, 27, (4), 1992.

6 British Columbia Labour ForceDevelopment Board. Training forwhat? Victoria, B.C., Ministry ofSkills, Training and Labour, 1995.

7 Drouin, M-J. Workforce literacy : aneconomic challenge for Canada.Montréal, The Hudson Institute,1990.

8 Kunin, R. Canadian BusinessReview, Summer, 1988.

9 Canadian Labour Market Product-ivity Centre. The linkages betweeneducation and training and Canada’seconomic performance. QuarterlyLabour Market and ProductivityReview, Winter, 1989.

10 Conference Board of Canada.Employability skills profile : the crit-ical skills required of the Canadianworkforce. Ottawa, Ont., The Confer-ence Board of Canada, 1991.

11 Bates, A W. Educational aspects ofthe telecommunications revolution.In: Teleteaching. Davies, G,Samways, B. Amsterdam, New Hol-land, 1993.


Special

93


1 Introduction

Among telecommunications people the term conferencing hasbeen associated with audio conferencing, that is audio only, orvideo conferencing, that is audio and moving image.

The term conferencing is also used by data communicationpeople. Data conferences may be synchronous, that is real-timeconferences, or asynchronous, that is messages are sent to theother participants of the conference.

The introduction of ISDN and multimedia communication haveled to a need to add data communication functionality (e.g. textor graphics) to audio and video conferencing. In the same waypeople would like to add audio and moving images to data con-ferences.

Thus, the term multimedia conferencing is introduced. Fur-thermore, there is a need for multiparty multimedia confer-ences. The ITU-T T.120 series recommendations are definingthe data communication service for use in a multipoint multi-media conferencing environment.

This article presents an overview of the existing and plannedITU-T recommendations, and highlights some technologicaland market trends.

2 The basic principles

2.1 Area of applications

The ITU-T T.120 series of recommendations primarily address-es multiparty conferencing.

The initial application was audiographic conferencing, i.e.speech and data/graphics. The introduction of video codingalgorithms enabling the transmission of moving images both onthe ISDN and recently on the PSTN, led to an expansion of theapplication area. The recommendations now address multimediaconferencing, i.e. simultaneous transmission of speech, videoand data/graphics.

Single medium subsets (data conference or audio only confer-ence) are of course possible.

To set up a multiparty conference an MCU (Multipoint ControlUnit) is used as described in Figure 1. Point-to-point connec-tions are established between each conference participant andthe MCU. The protocols can also be used in point-to-point com-munications between two users, however, some of the protocolsare addressing multiparty functionalities.

The protocols also cover conference scenarios where two ormore MCUs are used to establish the conference as described inFigure 1.

The protocols are designed to set up and control multimediaconferences. The functions and protocols defined are free fromplatform or network dependence. The present versions areaddressing the data part of a multimedia conference and theconference control, while the audio and moving image parts ofthe multimedia conference need to be predefined similar to thepresent video conference service offered by a number of serviceproviders.

The following types of conferences can be described:

• Data conference*

• Audio conference

• Audiographic conference (i.e. audio and data/graphics)*

• Video conference (i.e. audio and moving image)

• Multimedia conference (i.e. audio, moving image anddata/graphics)*.

The T.120 protocols can offer control functionalities, e.g. chair-man control, to all these types of conferences, and in additionoffer data communication functions to those types indicated byan asterix.

2.2 Structure

The T.120 protocols can be separated into three groups as de-scribed in Figure 2. These groups are:

• Network related protocols

• Protocols addressing multipoint communication service andconference control

• Application protocols.

2.2.1 Network related protocols

ITU-T Recommendation T.123 specifies the network relatedprotocols. Protocol stacks are defined for:

• ISDN.This protocol stack is based on an MLP data channel estab-lished by the audio-visual framing and in-band signalling sys-tem defined in ITU-T Recommendation H.221. All terminalsthat are conforming to ITU-T Recommendation H.320 supp-ort this in-band signalling system.

• Line-switched data networks (X.21).The MLP channel shall be established by the audio-visualframing and in-band signalling system defined in ITU-T Rec-ommendation H.221.

95

Protocols for multimedia multiparty conferencing – a presentation of the ITU-T T.120 series recommendations

B Y T R O N D U L S E T H


Terminal Terminal

MCU MCU

Terminal Terminal

Terminal

TerminalMCU

Figure 1 A conference set-up using three MCUs

• Packet-switched data networks (X.25).

• The analogue telephone network (PSTN) viamodems.

• Other networks or leased lines where audio-visual framing and in-band signalling systemdefined in ITU-T Recommendation H.221 isused.

• ATM-based networks (B-ISDN).

• Local Area Networks.

The protocol stacks covered by this recommenda-tion are described in Figure 3.

The protocol stacks for ATM-based networks andLAN are in the 1996 version of the recommenda-tion.

By using these protocols conferences between usersattached to different networks can be set up. Theseaspects are further elaborated in Section 2.4.

It is worth noting that the standardised protocolstack for ISDN communication requires use of theframing and in-band signalling system defined inITU-T Recommendation H.221. The 1996 versionof ITU-T Recommendation T.123 describes somealternative profiles,

• Based on ITU-T Recommendation Q.922 (i.e.without the MLP data channel)

• Based on the telematic protocols defined inITU-T Recommendation T.90

• Based on start/stop characters as defined inITU-T Recommendation V.120.

None of these alternatives are fully standardised inthe Recommendation. Furthermore, there is no pro-cedure to identify the profile used apart from theprocedures described in the referenced recommend-ations.

Multimedia applications will normally use theH.221 framing and in-band signalling and will thushave no problems. But there is a problem for bothdata conferences and applications where there is nosimultaneous audio and moving image communica-tion.

2.2.2 Protocols for multipoint communi-cation and conference control

The Multipoint Communication Service is de-scribed in ITU-T Recommendation T.122, and theassociated protocols are defined in ITU-T Recom-mendation T.125.

ITU-T Recommendation T.124 (Generic Confer-ence Control – GCC) provides a set of services forsetting up and managing the multipoint conference.It provides access control, chair control functionali-ties and control of priorities for the data communi-


T.127 (MBFT)

User application(s)(Using both standard and non-standard application protocols

User application(s)(Using std. appl. protocols)

Nodecontroller

User application(s)(Using non-std. protocols)

T.126 (SI)Application protocol entity

ITU-T T.120Application protocolrecommendations

Generic Conference Control (GCC)T.124

Multipoint Communications Service (MCS)T.122/125

Network specific transport protocolsT.123

......

Non-standard applicationprotocol entity

Figure 2 Structure of the T.120-series protocols

X.224/0

null + SCF

Q.922

H.221 MLPISDN

T.125 Multipoint communication service

X.224/0

null + SCF

Q.922

H.221 MLPX.21/X.21bis

X.224/0

X.25

X.21X.21 bis

X.224/0

null + SCF

Q.922

Modemstart/stop

X.224/0

null + SCF

Q.922/AAL5

I.361I.432

X.224/0

TPKT

LAN

LANMedium

Figure 3 Protocol stacks defined in ITU-T Recommendation T.123

cation between the participants in the conference. Four levels ofpriority are defined; the highest level is reserved for the confer-ence control based on GCC.

2.2.3 Applications

So far, two applications are standardised:

• Still Image (ITU-T Recommendation T.126).The Recommendation includes JPEG encoded colour stillimages, JBIG encoded black/white still images and facsimilegroup 3/4. Five profiles are defined. Procedures for creatingdrawings and the control of pointers in a multipoint environ-ment are defined,

• Multipoint Binary File Transfer (ITU-T RecommendationT.127).The Recommendation defines procedures for file distributionand file retrieval. Multiple files can be transferred simultane-ously to a number of addresses that can be defined for eachfile.

There will be few application recommendations among theITU-T T.120 series of recommendations. The objective is tostandardise the principles for the communication between non-standard applications and the standardised protocols for the ser-vice. ITU-T Recommendation T.121 describes a model forapplication protocols. The recommendation is presented in Sec-tion 2.3.

2.3 Model for application protocols

To assist application developers ITU-T Recommendation T.121Generic Application Template (GAT) presents a generic modelfor T.120-conforming applications.

Figure 4 describes the principle of the generic model.

The model consists of two functionally distinct parts:

• The Application Resource Manager (ARM) which is respons-ible for managing GCC and MCS resources

• The Application Service Element (ASE) which providesapplication-protocol-specific functionality.

All standardised application protocols are based on this model.It is recommended to use this model when developing non-standard applications.

The use of this model ensures that all non-standard applicationshave equal structure. It will simplify the communication be-tween non-standard applications developed by several compa-nies, but it cannot guarantee the communication between theseapplications; the guarantee can only be achieved by standardis-ing the applications.

2.4 Network independence

The T.120 series recommendations support a broad range oftransport options, including the Public Switched Telephone Net-work (PSTN), Integrated Switched Digital Networks (ISDN), orLocal Area Networks. Conferences might be set up between


Node Controller

GenericConference

Control

User application(s)

ApplicationResourceManager(ARM)

Multipoint Communication Service

ApplicationService

Elements(s)(ASE)

Figure 4 The Generic Application Template Model

terminals of diverse capability, on multiple different networksand across administrations. An example is described in Figure 5.

Apart from the ITU-T standardised protocol stacks, DataBeamoffers solutions for communication on Novell Netware IPX net-works and on TCP/IP networks (e.g. Internet or Intranet) as de-scribed in Figure 6. The scenario described in Figure 5 cantherefore be extended to include Novell networks and TCP/IPbased networks. Since the version of ITU-T RecommendationT.123 approved at WTSC was completed by ITU-T SG 8,ITU-T has decided to accept normative references to IETF doc-uments (i.e. RFCs). The next version of T.123 will thereforeinclude a protocol stack for TCP/IP.

The network bandwidth restriction may create problems. Thebandwidth of an ISDN B-channel is 64 kbit/s, while the band-width when using standardised modems for the PSTN is re-stricted to maximum 33.6 kbit/s when using the recently app-roved extension to ITU-T Recommendation V.34. However, themodem bitrate depends on the quality of the connection; itmight be lower. On Local Area Networks the delay may createproblems for real-time communication.

As stated previously the present versions of the recommenda-tions focus on data communication and conference control.Bandwidth variations or delay introduces no serious problemsfor that part of a conference, but the impact on the audio andmoving image parts of a multimedia conference may be dra-matic. Furthermore, different video coding algorithms may cre-ate problems, transcoding between different video coding algo-rithms might be expensive and introduces a significant increaseto the delay of the connection.

2.5 T.120 implementor’s guide

Even though all possible effort has been made to avoid errorsand/or phrases that can be interpreted incorrectly, such prob-lems can hardly be avoided. ITU-T has no procedures for cor-recting errors in an approved document; a new edition has to beprepared.

The approval process of a new edition takes sometime. For the benefit of the users ITU-T has issuedan ‘Implementor’s Guide for the ITU-T T.120 Rec-ommendation Series Data Protocols for MultimediaConferencing’. This document is a compilation ofreported defects. It is a ‘living’ document freelyavailable. The document resolves defects in the fol-lowing categories:

• Editorial errors• Technical errors such as omissions or inconsis-

tencies• Ambiguities.

In addition, the document may include explanatorytext found necessary as a result of interpretationdifficulties apparent from the defect reports.

The document does not address proposed additions,deletions or modifications to the recommendationsthat are not strictly related to implementation diffi-culties in the above categories.

As already mentioned this is a living document.The latest version can be obtained at IMTC’s ftp-site at the address

ftp://ftp.imtc-files.org/imtc-site/IMPGUIDE

or ITU-T’s Web-site

http://www.itu.ch/itudoc/itu-t/com8/implgd.html.

Usually, new versions are available on the IMTC ftp-site earlierthan on the ITU-T Web-site.

3 Market actors and products available

3.1 Participants in the standardisation activities

North-American companies have been the major contributors tothe standardisation of the T.120 protocols. Mr. C.J. Starkey,Vice President of DataBeam and co-founder of this company,has been a key person and was rapporteur for the questionaddressing the T.120 protocols in CCITT (now ITU-T) for 6years until the middle of 1993.

Other major contributors have been AT&T, Bellcore, BJ Com-munications, BT (British Telecom), France Telecom, Picture-Tel, PolyCom and VideoServer. Other contributors are Intel,BNR (Nortel), ConferTech, Multilink and VTEL. MicroSoft hasbeen participating in 1996.

Apart from BT and France Telecom, European companies havebeen less active. GPT Video Systems has attended most of themeetings addressing the T.120 protocols. Other European com-panies attending some of the meetings are Teles, KPN (DutchPTT) and Austrian PTT. Other European network operators aswell as Japanese and South Korean companies have been pre-sent at a few meetings.

IMTC (The International Multimedia Teleconferencing Con-sortium, Inc.) is an important actor that is founded to facilitateand promote the creation and adoption of international standardsfor Multipoint Document and Video Teleconferencing, specific-


LAN A

LAN GatewayRouter

ISDN

MultiportTerminal

MultiportTerminal

Terminal

MCU

Terminal

MCU

Terminal

PSTN

PSTN

PSTN

ISDN

ISDN

Public Networks

TerminalTerminal

LAN B

LAN GatewayRouter

TerminalTerminal

ISDN

Figure 5 Example of a mixed-network conference topology

X.224/0

null + SCF

Q.922

COM.DRV

T.125 Multipoint communication service

X.224/0

null + SCF

Q.922*

NWIPXSPX.DLL

RFC 1006X.224/0

WINSOCK.DLL

X.224/0

null + SCF

Q.922

H.221 MLP

PSTN IPX TCP/IPISDN

Figure 6 Protocol stacks supported by DataBeam

ally the ITU T.120 and H.320 standards suites. The work onsupporting the T.120 series recommendations was initiated byCATS (The Consortium for Audiographics TeleconferencingStandards, Inc.). In 1994 CATS and MCCOI (The MultimediaCommunications Community of Interest) were merged to formIMTC.

Most of the CATS members were North-American companies,while the membership of MCCOI was based on both North-American and European companies. Telenor was a member ofMCCOI, and is now a member of IMTC.

In December 1995, IMTC merged with PCWG (the PersonalConferencing Work Group), and the major network operators,suppliers of video conferencing equipment as well as compa-nies like MicroSoft and Intel are now participating in the workof IMTC.

It is worth noting that, unlike other fora, IMTC is no alternativeto the standards organisations, the fundamental goal is to bringall organisations involved in the development of multimediateleconferencing products and services together to help createand promote the adoption of the required standards.

3.2 DataBeam

3.2.1 Introduction

DataBeam is one of the active companies in the work on theITU-T T.120-series recommendations. Co-founder and VicePresident of DataBeam, Mr. C.J. Starkey, was rapporteur for thework in CCITT, and is currently president of the IMTC.

DataBeam has established agreements with several partners onjoint activities. Examples are:

• Agreement with MCI to deliver new real-time services formulti-point collaboration via the Internet

• Relationship with Cisco on new Internet server with confer-encing capabilities

• Agreement with MicroSoft to facilitate the development anddeployment of real-time multi-point data applications in cor-porate Intranets and the global Internet

• Strategic co-operation between Cisco, MCI, MicroSoft andSun on development and marketing of neT.120

• Agreement with AT&T Paradyne and Multilink on develop-ment and marketing of technology for multiparty audiograph-ic conferencing.

The products offered by DataBeam are:

• CCTS (Collaborative Computing Toolkit Series), a develop-ment tool

• FarSite, a software product

• neT.120 Conference server, a data conferencing server.

3.2.2 CCTS

CCTS is an integrated set of software development tools basedon the T.120 standards. It is organised in 6 modules whichcorresponds to one or several T.120-series recommendations.

A toolkit on application sharing has been launched, but is nowwithdrawn, probably because ITU-T now is working on a rec-ommendation on application sharing (T.SHARE).

The modules are:

• Multi-Point Communications Application Toolkit (MCAT).MCAT is the basis module which provides a general-purposefacility for establishing, servicing and managing the simul-taneous delivery of data between multiple users connected viaLAN, WAN, and dial-up networks. MCAT is based on theITU-T Recommendation T.122/125 and T.123.

• Generic Conference Control Application Toolkit (GCCAT).This module offers facilities for establishing and managingmeetings, as well as controlling the communications re-sources required by the conference applications.

• Node Controller Toolkit (NCT). This module streamlines theestablishment and management of conference connections.NCT furnishes a generic standard for an undefined portion ofthe T.120 infrastructure.

• Session Manager Toolkit (SMT). This module offers a flex-ible abstraction of the session management functions com-mon to all T.120-based applications. SMT is based on ITU-TRecommendation T.121, and secures a standardised commu-nication between the applications.

• Shared Whiteboard Application Toolkit (SWAT). This mod-ule allows developers to rapidly add powerful T.126 compli-ant document conferencing or whiteboarding functions totheir applications. SWAT is based on ITU-T Recommend-ation T.126,

• File Transfer Application Toolkit (FTAT). This module offersa simple way to embed multi-point background file transferfunctions into an application. FAFT is based on ITU-T Re-commendation T.127.

CCTS can be considered as an implementation of the principlesdescribed in ITU-T Recommendation T.121, and is probably theonly commercial available tool for development of T.120-appli-cations.


Non-standardapplications

SMT (T.121) Node controller

SWAT (T.126)

FTAT (T.127)

GCCAT (T.124)

MCAT (T.122, T.123, T.125)

Figure 7 CCTS modules

3.2.3 FarSite

FarSite is a software which offers the tools needed for remotecollaboration based on the T.120 protocols. FarSite providesboth application sharing and shared whiteboarding, in point-to-point connections and in a multipoint conference.

Multipoint Binary File Transfer conforming to ITU-T Recom-mendation T.127 is offered. The whiteboarding function in thepresent version (FarSite 3.0) is probably not fully conforming toITU-T Recommendation T.126, although this is the objective ofDataBeam.

FarSite is available as a separate product and is included in sev-eral multimedia conferencing terminals and systems.

3.2.4 neT.120

The neT.120 Conference Server is a data conferencing serverbased on T.120 technology. It can be connected to the networksidentified in Figure 6, that is ISDN, PSTN, IPX-based networks(Novell) and TCP/IP-based networks (Internet and Intranet).However, the present marketing of neT.120 indicates thatDataBeam strategic decision is to offer products for dataconferencing on Internet and Intranet. The agreement withCisco, MicroSoft and Sun clearly documents this decision.

A software-only server, the neT.120 Conference Server lever-ages existing web browsers and desktop applications. The mini-mum requirement for participating in a neT.120 based confer-ence is an Internet browser such as MicroSoft Internet Explorer,Netscape’s Navigator or NSCA Mosaic.

By using terminals conforming to the T.120-series recommend-ations more functionalities can be offered.

3.3 Suppliers of T.120-based products

In the past standardisation has not been given high priority bycomputer and software manufacturers. Usually, the marketforces, strategic alliances, etc. have been used to establish defacto standards. In the development of video conferencing term-inals several suppliers have developed alternatives to the stand-ardised solutions. However, the last year a trend to choosestandardised solutions can be observed. Examples are:

• Intel ProShare 2.0 which conforms to ITU-T Recommenda-tion H.320, and offers T.120 functionality. Although not offi-cially confirmed, Intel will probably use MicroSoft developeddata conferencing functionalities. Intel’s development will berestricted to the audio and video part. Co-operation betweenIntel and VideoServer is established, and the VideoServerMCU is compatible with Intel ProShare 2.0.

• Apple’s video conferencing system, QuickTime is also con-forming to ITU-T Recommendation H.320. Apple has alicence agreement with DataBeam, and plans to introduceQuickTime with T.120 functionality in 1996. The licenceagreement with DataBeam allows the QuickTime users todevelop T.120-based applications without an extra licensingagreement with DataBeam.

• MicroSoft has indicated that support of T.120 protocols willbe a part of the successor of Windows 95, and that the T.120

protocols will be used in other MicroSoft products (e.g.MicroSoft NetMeeting). MicroSoft and DataBeam have sign-ed an agreement on development of T.120 application for useon Internet.

• PictureTel, one of the major suppliers of video conferencingequipment, is active in the ITU-T work on the T.120 recom-mendations. PictureTel is also active in the interoperabilitytesting events organised by IMTC (Event 120).

• The major supplier of video conferencing MCUs,VideoServer, is actively participating in the ITU-T work onthe T.120 recommendation, and offers some T.120 function-ality in software version 5 for their MCUs.

4 Trends

4.1 Market trends

No doubt the T.120 protocols will be a success. Several observ-ations supporting this statement can be made:

a) Important market actors such as MicroSoft and Intel are nowactively participating in the standardisation work, and areactively promoting these protocols.

b)Computer Supported Co-operative Work, teleworking andmultiparty multimedia conferencing have become importantstrategic elements for a number of companies.

Another aspect is the introduction of the T.120 protocols in theInternet environment. So far, real-time multiparty video confer-ences can be set up by using MBONE, a virtual network usingIP multicast on Internet. The information is distributed to allhosts who have expressed interest in becoming a member of acertain multicast group. Like the present version of Internet,there is no Quality of Service guarantee. By the use of the TimeTo Live (TTL) value the distribution of the information can berestricted.

The introduction of neT.120 is the first step towards multimediaconferences on the Internet that include a well-defined group ofparticipants. New functions such as chairman control can beoffered. Extensions to the present IP protocols which offerguaranteed bandwidth are available, and the Internet Engineer-ing Task Force is working on a new version of the Internet Pro-tocol (IPv6) which can guarantee Quality of Service. When thenew version of IP is implemented, world-wide multimedia con-ferences offering functionalities and Quality of Service similarto those offered on public networks, can be set up. This servicewill at the beginning be introduced on corporate networks(Intranets). Multimedia conferences set up on Intranets wheresome participants may be attached via public networks will bean important market segment.

From a technology point of view multimedia conferencingincluding moving images is the most exciting application. Fromthe users’ point of view audiographic conferencing and dataconferencing might be more interesting on the short term.Today, video conferences where the quality of the movingimage is acceptable for professional use requires connections setup via ISDN or networks offering higher bitrates. Audiographicconferences or data conferences can be set up on PSTN, Internetor a combination of PSTN and Internet (e.g. the audio connec-tion on the PSTN, the data connection on the Internet). The


required investment in terminal equipment will usually be low,a telephone, a modem and a PC are required. New modemstandards offering simultaneous transmission of voice and dataon a single telephone line may also be an alternative.

Another important trend is the evolution from the businessmarket to the domestic market. In the future multiparty gamesmight be an interesting market segment.

4.2 Network platforms

As already described standardised protocol stacks for publictelecommunications networks (PSTN and ISDN) and for publicdata networks are available. Protocol stacks for LANs andTCP/IP networks are also available.

ITU-T is currently working on a recommendation on audio-visual terminals on LANs without a guaranteed Quality of Ser-vice (Recommendation H.323).

A recommendation on videophones attached to the PSTN isapproved by ITU-T (Recommendation H.324). However, thesize of the screen as well as the audio and video quality areprobably not acceptable for business users. Real-time videotele-phony on PSTN or Internet has for the time being several short-comings, due to time delay as well as available bandwidth.Today, the only realistic network platform for multimedia com-munication where moving images are included, is the ISDN. Itis possible to set up conferences where both audio, movingimages and data are transported on two ISDN B-channels, i.e.128 kbit/s. For more demanding applications the bandwidth canbe increased to e.g. 384 kbit/s (6 B-channels).

Both data conferences and audiographic conferences can be setup on the ISDN. However, the standardised protocol stack forISDN communication is based on the in-band signalling systemspecified for audio-visual communication. Data conferencing onthe ISDN not using this in-band signalling system is therefore aproblem for the time being.

These considerations lead to the following scenarios:

1. Multimedia on the ISDNThis scenario is requiring video conferencing terminals,usually a PC, with T.120 software. Another option is a desk-top videophone or a video conferencing terminal with a PCinterface. Some videophones are equipped with a PC inter-face, but few support the necessary function on the data linklayer to enable T.120 communication.

2. Audiographic conferencesThis scenario may be based on ISDN communication. Thereis a standard specifying a wideband (7 kHz bandwidth) ISDNtelephone where up to 14.4 kbit/s can be reserved for datacommunication. The in-band signalling system specified inITU-T Recommendation H.221 is used. The terminal must beequipped with an interface to a PC, similar to that of desktopvideophones.

A more realistic short term scenario is to set up two PSTNconnections, one for audio, the other for data. A telephonemay be used for the audio communication; usually a hands-free solution is preferred. An alternative might be a headset.

3. Data conferencesThis scenario will be based on PSTN or Internet communica-tion.

Combinations are of course possible. One alternative is to useInternet for the data part of a multimedia or audiographic con-ference.

Where guaranteed Quality of Service can be offered Internet/Intranet is a realistic alternative for multimedia conferences too.

4.3 Interoperability testing

Experiences gained by the introduction of new protocols such asfacsimile and videophony indicate that the risks for misinterpre-tations of the requirement of the recommendations, errors in therecommendations or shortcomings of the recommendations aresignificant. There is always a need for a verification of a stand-ard or recommendation, as well as a need for verifying thatequipment produced by different manufacturers is able to inter-work.

IMTC is addressing these issues. An interoperability testingevent called Event 120 is organised by IMTC. The first eventwas held in Santa Clara, California, 25–27 March 1996. 25manufacturers participated. Approximately 6,000 test sequenceswere conducted, covering equipment connected to ISDN,PSTN, LAN and Internet.

A second session was held in Gaithersburg, Maryland 18–20 Sep-tember 1996. At this session 3,000 test sequences were performed.

It is claimed that 95 % of these tests were successful. Althoughthese results are promising, it is worth pointing out a fewfavourable conditions:

• One is the leading role of DataBeam. Most of the implement-ation was based on the DataBeam products or DataBeam’sdevelopment toolkit. When software or development toolkitfrom other suppliers is available, the situation could be worse.

• Another is that the tests were predefined. It is therefore likelythat attempts have been made to keep clear of problem areas.

The initiative to start Event-120, however, is positive, and animportant condition for interoperability between terminal equip-ment produced by different manufacturers.

In Europe, EV (European Videoconferencing); an organisationwhere European network operators are participating (amongthese Telenor), has carried out both interoperability testing andconformance testing of H.320 terminals.

4.4 Standardisation

4.4.1 Applications

As already stated, there will be few recommendations on appli-cations among the ITU-T T.120 series of recommendations.Some proposals for application recommendations have beenforwarded. One of these is to expand to T.190 series recom-mendations on Co-operative Document Handling (CDH) toinclude multiparty scenarios. A proposal for organisation of thework is described in Figure 8.


Work in ITU-T SG 8 on XAPI (Extended Applications Pro-grammable Interface) has so far been addressing point-to-pointCDH. The extension to a general use in a T.120-environment isnow considered.

Other applications related issues are security and encryption inan T.120 environment.

The most important proposal, to develop a recommendation onapplication sharing in a T.120 conference (T.SHARE), is pre-sented by MicroSoft, PictureTel and PolyCom. The proposal issupported by a number of other North-American companies.The proposal is based on technology used in MicroSoft Net-Meeting and PictureTel LiveShare Plus.

The principles of this recommendation are described in Figure 9.

The application runs on a host. The screen information is trans-mitted to the other participants of the conference. The otherparticipants can control the application by using mouse and key-board. Windows is not required. However, if Windows is used,all participants may switch between applications as if the appli-cations were running on their PC. The only restriction is thatinformation from the clipboard of other PCs than the host can-not be copied into the application(s).

The transmission of the screen information and control informa-tion will use traditional T.120 channels, and full T.120 function-ality is available.

T.SHARE will in principle specify two protocols:

• The main protocol is the protocol which will be maintainedand further processed


• In an annex a protocol which secures compatibility withMicroSoft NetMeeting and PictureTel LiveShare will bespecified. This protocol will not be maintained, and will bewithdrawn after a period to be defined (e.g. three years).

Several observations can be made on this proposal:

a) It is proposed to standardise a new application, a logicalevolution to secure interoperability between applications, butin conflict with the argument that applications are a subjectfor competition, not standardisation.

b)One of the sources of the proposal is MicroSoft, a companywhich is large enough to establish de facto standards.

c) It is proposed to standardise principles which are already usedin Internet products. This is an indication that in certain areasstandards designed for traditional telecommunications appli-cations will also be used in products designed for Internet. Inother words, the communication between traditional telecom-munications networks and Internet will be simplified.

d)The proposal, supported by a number of North-Americancompanies, is to some extent in conflict with the decisionmade by ITU-T SG 8 to develop CDH protocols for multi-party conferences.

e) Intel has accepted that a protocol which is not compatiblewith their application sharing protocol, is standardised. Fur-thermore, Intel has declared that there will not be further re-vision of the present application sharing software, the newsoftware will be based on T.SHARE.

f) An underlying question is – should protocols covering all rel-evant scenarios be standardised, or should attempts be madeto standardise protocols which are simpler and which arefunctioning although they may not cover all possiblescenarios.

By tradition the Internet environment has prepared their ownstandards, RFCs. However, some standards developed by ISOor ITU-T on algorithms for coding of audio and moving imagehave been used. There is also a trend to remove the border be-tween RFCs and formal standards developed by ITU-T andother standardisation organisations.

T.127T.126

Extend Applications Programmable Interface(XAPI)

Applications

SDPADPJSET.121

T.120 Multipointprotocol stack

Point-to-pointprotocol stack

Responsibilities of multipoint group

Responibilities of document transfer and CDH group

JSE : Joint Synchronous EditingADP : Asynchronous Document ProductionSDP : Sequential Document Production

Figure 8 Proposal for organisation of the work on CDH

Host

A

C

Host

B

Screen information Screen information

Moucekeyboard

Moucekeyboard

Figure 9 Principles of T.SHARE

4.4.2 Network interworking and applications inter-working

Network interworking is usually no problem for data conferenc-es and for audiographic conferences where two separate connec-tions are set up. The present standards for multiparty video con-ferences are based on the condition that all participants whoreceive the moving image have to have the same video codingalgorithm, and to transmit/receive the coding image, audio anddata at the same speed. In most cases these participants are con-nected to the same network. However, gateways do exist be-tween LANs and ISDN which support the principles of the ITU-T H.320 standard, and thus offer the user a possibility to partici-pate in a multiparty conference from a terminal on the local net-work. Participants may also be connected to the MCU via leas-ed lines. These conferences where the video coding algorithmand the transmission speed are identical are identified ashomogenous conferences.

Conferences where the participants are connected to differentnetworks (ISDN, Internet, ATM, etc.), different video codingalgorithms are used, and where also graphics and audio mightuse different coding algorithms, are described as heterogeneousconferences.


User applications

Realtime

audiodevice

Realtimevideodevice

Network specificmappings (T.131) Node controller

Remote Device Control(T.RDC))

Non-standardapplication

protocol entities

Audio-visual-control(T.132/T.133) GCC (T.124)

MCS (T.122/125)

Network specificcontrol entity

Network transport protocols(T.123)

Network multiplex

controlvideostream(s)

audiostream(s)

control data

ITU-T Standardapplication

protocol entities

Figure 10 Relations between the T.120 series recommendations and the planned T.130 series recommendations

Standards addressing these issues are on the way. Initially, asingle recommendation identified as T.AVC was planned. Nowthe objective is to prepare a set of recommendations identifiedas the ITU-T T.130 series. The structure of Figure 2 can beextended as indicated in Figure 10.

The number of recommendations in the T.130 series is not yetfixed. There will be an introductory recommendation, T.130.There is work on

• T.132 – Audio Video Control: Infrastructure Management• T.133 – Audio Video Control: Service Management.

Drafts are available. It is planned to approve these recommenda-tions at study group level in 1997. However, there are still anumber of complex, unsolved issues; the plan may fail.

A network related recommendation, T.131 – Protocol for Net-work Specific Mapping of Real Time Audio Video Elements, isplanned to be approved in 1998.

One of the issues discussed is whether control of remote devices(cameras, microphones, VCRs, etc.) should be an integral part

of the already planned recommendations, or be a separate rec-ommendation (T.RDC).

Most of the functionalities specified in the T.130 series arebased on the use of an MCU or a server. Scenarios where the in-band signalling is based on ITU-T Recommendation H.221(equipment connected to ISDN or leased lines), or ITU-T Re-commendation H.245 (PSTN, ATM or LAN) will be covered.

A control protocol for network interworking will be defined.Control mechanisms for transcoding (e.g. between H.261 andMPEG) as well as mechanisms enabling the user to specify andcontrol Quality of Service, will be prepared.

Whether T.131 and T.132 should be merged or not is anotherissue. A decision to merge these recommendations will probab-ly result in a delay.

ITU-T recommendation T.133 will cover

• Video switching• Video processing (e.g. Continuous Presence)• Mixing of audio information• Identification of audio and video sources• Chairman control.

An example is described in Figure 11. A Continuous Presencefunction presents images from several sites (e.g. five sites). Oneof the images is larger than the others. Additional information(video, documents, etc.) can also be presented.

The present Telenor video conference service (TelenorMøtepunktet®) offers a Continuous Presence function whereimages from four sites are presented on one screen. However,when the new recommendations are in force more flexibilitycan be offered. Some of the functions overlap with functionsspecified in the T.120 series recommendations. One solution isto extend ITU-T Recommendation T.124 to cover T.130-basedfunctionalities.

5 Closing remarks

The importance of standards can be illustrated by the success ofthe facsimile. Prior to the creation of the facsimile standards, afacsimile machine could only communicate with other machinesmade by the same manufacturer. The consequence was a few,expensive machines on the market. Today, people take it forgranted that when they send a fax, it will work no matter whatbrand of machines are sending and receiving. The price of a faxmachine has reached a level where it is interesting to homeusers as well as business users.

In much the same way, the T.120 series of standards defines ameans to communicate multimedia across various networks. Ascompanies develop software products based on these standards,they will be able to communicate with each other. The T.120series of standards are also creating a bridge between severalnetworks, PSTN, ISDN, Internet/Intranet, ATM and LAN. TheT.120 series standards will be an important element in the mul-timedia applications that will be developed in the coming years.


Site 1 Site 2 Site 3 Site 5

Site 3Document

camera

Site 3 stream 2Site 3 stream 1

Figure 11 A Continuous Presence presentation

1 Introduction

Developing the ATM networking concept to its current state ofmaturity has for several years been a major effort in researchand standardisation. In that process, most important aspects ofthe concept have been studied thoroughly and in depth.

Charging and billing are traditionally operator internal respons-ibilities. For this reason, and perhaps also for other reasonsrelated to the competitive nature of the topic, the charging andbilling aspects of ATM have received less attention than otheroperational issues. Now that large scale commercial deploymentis imminent, achieving insight into charging and billing of ATMservices is of vital importance for both customers and operators.Generating such understanding over a broad range of questionsrelated to charging and billing of ATM services constituted themain motivation for forming the European ACTS projectCA$hMAN. The project consists of several European compa-nies, including Telenor R&D and Ericsson from Norway.

To develop a good and sound billing system many aspects con-cerning both operators and customers have to be taken intoaccount. For a network operator it will be essential to achievecustomer satisfaction and good network utilisation, for therebyto increase the revenue. This means that a network operator hasto put requirements on how the charging scheme should operateand be implemented. Below is mentioned some of these require-ments [7]:

• Cost-recovery (the pricing should reflect the network cost foran average over a range of services and customers)

• The total billing system should be cost effective and tech-nically feasible to implement

• The pricing structure should be understandable by customersand salespersons

• Pricing should be flexible – ability to tailor to customer andmarket segments

• Provide economic incentive to upgrade network sufficientlyto provide good QoS

• Encourage responsible use of network by customer.

For the customer the main goal is service satisfaction. Even asthe operator puts requirements on the charging schemes, thecustomer will put requirements regarding charging aspects ofthe services offered. These could be:

• Clear, predictable and traceable call charges

• Comparison with competitors’ prices possible

• Clear relationship between cost and service delivered

• Possibility to trade quality against cost.

This article will give an overview of some different methods ofcharging/pricing of ATM services. Since our work on theseissues was initiated by the CA$hMAN project, we will link gen-eral thoughts and considerations about pricing policies for ATMto the work undertaken by the project.

To give a short summary of the article, some introductoryremarks regarding the special features of ATM relevant tocharging follows this introduction. Thereafter follows a shortpresentation of the CA$hMAN project. In Chapter 2, some

methods for ATM charging is presented, from the most simplesolutions to more sophisticated ones. The chapter is summed upby introducing the special charging algorithms the CA$hMANproject has decided to investigate in more depth. In Chapter 3 isoutlined a general discussion about the billing system as a com-plete system for charging ATM services. The discussion coverswhich components in the network would be involved in billingissues, how different services could be reflected in the billingsystem, and considerations about the reliability of a total billingsystem. In Chapter 4 we again turn the focus to the work under-taken in the CA$hMAN project, where investigation of differentcharging algorithms through user experiments are stressed. Thechapter describes the experiment platform used, what questionswe try to answer by conducting experiments, and the results wehave achieved so far in the project. In Chapter 5 some conse-quences for the network operators are discussed, and finallyChapter 6 gives some concluding remarks and some ideas offuture work to be done.

1.1 Some aspects of ATM relevant to charging

From the initial idealistic ideas of creating a simple and univer-sal system for asynchronous transfer, ATM [8] has developedinto a highly sophisticated networking concept. Although beingperhaps the most universal standardised networking conceptdeveloped until now, the intended simplicity vanished a longtime ago. It should not be surprising then that the sophisticationof the networking concept may also imply sophisticated charg-ing. The purpose of this section is to identify and outline aspectsof ATM that may have bearing on questions related to charging.

In ATM, fixed size units are switched based on the principle ofvirtual connections. Being asynchronous, the switching andtransfer rely on statistical sharing of network resources. Virtualconnections and statistical sharing of resources are as such oldconcepts, but a quite revolutionary property of ATM is thatthese concepts are combined with contractual guarantees on ser-vice qualities. This observation has several far-reaching impli-cations, one of them being that the network operator must close-ly watch ingress traffic, and if necessary impose traffic filteringon the traffic entering the network in order to be able to fulfilthe contractual obligations.

When a customer sends a service request for a new connectionto the network, a proposal is at the same time made for a trafficcontract. There is no guarantee that the network will actuallyaccept the proposal, guarantees apply only after the contract ismade. The process of evaluating such service requests is de-noted Connection Admission Control (CAC). A possibility (atleast theoretically, this is not standardised) may be that the cus-tomer at the same time requests one of several possible tariffsfor the service (see Section 2.4.1). Other elements of the con-tract proposed will be the type of service to be used for the con-nection, i.e. the ATM Transfer Capability (ATC), and theparameters of the policing mechanism (Usage Parameter Con-trol (UPC) parameters).

The UPC functions and the transfer capabilities are defined in[2]. The transfer capabilities relate to what kind of resourcesmust be allocated in the network to provide a service of a cer-tain kind. The type of policing functions to be used is alsoclosely related to the choice of ATC. Using ITU-terminology(ATM-forum have their own definitions, also with slightly dif-

105

Charging of ATM servicesB Y R A G N A R A N D R E A S S E N , Ø Y V I N D B R E I V I K , E I N A R E D V A R D S E N ,

T H O R E S K E D A L , N I N A K L O S T E R , O L A V Ø S T E R B Ø


fering semantics), four ATCs are defined: Deterministic BitRate(DBR), Statistical BitRate (SBR), Available BitRate (ABR) andATM Block Transfer (ABT). In the CA$hMAN project, onlythe first three of these have so far been considered.

For the DBR ATC, network resources must be allocated to sus-tain sources sending at a continuous peak rate. Accordingly, theUPC parameters are only related to the peak rate of the offeredtraffic, i.e. restricted to the two parameters defining the peakrate policing: Peak Cell Rate (PCR) and Cell Delay VariationTolerance (CDVT).

For sources that may not need the same amount of resourcesallocated, but still require quality guarantees, the SBR ATC isdefined. For such connections, the traffic contract will basicallycontain the two peak rate parameters, and two additional para-meters related to the foreseen burst level variability of thesource, named Intrinsic Burst Tolerance (IBT) and SustainableCell Rate (SCR). The choices made on IBT and SCR will influ-ence the amount of resources that must be allocated in the net-work.

The last ATC considered in the CA$hMAN project is the ABRATC, which is basically a ‘best effort’ service defined withfeedback rate-control mechanisms. The ATC is defined with aMinimum Cell Rate (MCR), denoting a guaranteed lower boundon available capacity for the source, a Peak Cell Rate, and anAllowed Cell Rate (ACR) that will vary dynamically betweenthe peak rate and the minimum rate, reflecting the ‘best effort’component of the service.

Other quality parameters that may be agreed for a connection ina traffic contract, ‘orthogonal’ to the ‘service plane’ spanned bythe ATCs, may be related to loss probabilities and delay proper-ties of transmitted traffic, and the ‘priority level’ of the traffic.All these parameters will influence the amount of resources thatmust be allocated in the network to provide a service, and maythus also be possible discriminators in pricing schemes based onresource usage.

1.2 The CA$hMAN project

The ACTS project CA$hMAN(Charging and AccountingSchemes in Multi-Service ATM Networks), is a consortiumconsisting of companies from several European countries. Net-work operators, hardware manufacturers, universities and soft-ware developers are all co-working to solve different tasks relat-ing to charging of ATM services. Since some underlying strate-gies of charging are operator confidential, charging schemes arenot an issue for recommendations. The main goal is, as stated inthe CA$hMAN project summary for 1996, to “study, develop,implement, verify and compare charging and accountingschemes for ATM networks”. The outcome of the project will,however, expectedly impact on standardisation for the structureof accounting models, on the TMN architecture related toaccounting management and requirements on the processingand monitoring capabilities to support the charging mecha-nisms. To do this, experiments and field trials with real usersare highly prioritised, and the use of ATM National Hosts forconducting them has been stressed. SUPERNET in Norway,TRIBUNE in the Netherlands and EXPERT in Switzerland areall research networks which the CA$hMAN project hope to be

able to take advantage of. The CA$hMAN project deals withcharging strategies at the ATM level.

The CA$hMAN consortium consists of skilled people bothwithin teletraffic theory for developing charging algorithms, andwithin exploration of how charging impacts on network man-agement architectures. Experienced hardware constructors arealso part of the CA$hMAN team to build prototype equipmentable to support different charging algorithms. Hence, the con-sortium consists of people with many different skills, who to-gether form a group that is able to evaluate the chargingschemes from different angles.

2 Charging algorithms for ATM

Generally, there may be a rather broad range of possible charg-ing algorithms for ATM depending on the parameters and themeasurements they depend on. The simplest one is obviouslythe flat rate, e.g. the subscriber pays a fixed amount per month.This simple algorithm gives no incentive to use the networkresources in a particular way. Contrary to the flat rate pricingscheme, which is independent of the actual use, one reasonableprinciple is that the charge in a reasonable way should reflectthe actual resources a connection occupies in the network. Suchschemes are commonly denoted usage based charging (UBC).There may be some different ways of achieving this. The UPCfunction (with parameters negotiated at call set-up) will upperbound the traffic conveyed by the network over a connection.One possibility could be to charge by algorithms derived fromthe traffic contract and the time duration. Another possibility,which is investigated in CA$hMAN, is to charge according tosome traffic measurements. In CA$hMAN these measurementsare obtained by introducing the important notion of effectivebandwidth, which reflects the actual usage of networkresources. Some important properties and implications of effec-tive bandwidth is mentioned in Appendix I.

The different possible charging schemes may be listed accord-ing to the issues mentioned above starting with the simplestones. We must, however, not forget that the different ATCs maybe charged according to schemes which may result from otherconsiderations than pure traffic aspects. As an example it seemsreasonable that ABR services will be charged lower than DBRor VBR (for the same traffic volume) because of the lack ofguaranteed bandwidth.

Another important issue is the implementability of the chargingschemes. If the extra cost for implementing a complex chargingscheme is very high the benefit in terms of better performance isof less value.

In the following some general aspects of some possible and rele-vant charging schemes will be briefly discussed. In Figure 1 isillustrated how charge and mean rate relate for some of theseschemes.

2.1 Algorithms based on flat rate

As illustrated in Figure 1a), this charge is only based on a flatrate (e.g. per month). It will mainly depend on the type ofaccess link chosen including the bitrate. There may, however,also be other parameters defining classes of subscribers. It may


be worth mentioning that although the charge is independent ofthe characteristic of a particular connection, both UPC and CACare applicable to guarantee the QoS.

The flat rate does not give any incentives to use the networkresources in some particular way, the charge is independent ofthe number of connections, the traffic characteristic, and thetime duration. Although this charging scheme is simple, thesubscribers will benefit from the fact that they always will beable to predict the cost of their communication services.

Flat rate policies are inherently applicable to the Permanent Vir-tual Channel (PVC) services that are now being offered. Theseservices form direct rivals to the traditional flat rate leased lineservices with the asset of offering more flexible capacity agree-ments. With the advent of Switched Virtual Channel (SVC) ser-vices more elaborate schemes may be introduced.

2.2 Algorithms based on traffic contract and callduration

For DBR and SBR services the contract parameters play animportant role for the traffic management. For DBR the relevantparameters are PCR and CDVT and for SBR the parameters arePCR, CDVT, SCR and IBT. From these parameters it is possi-ble to calculate an equivalent capacity and the charge could beproportional to that capacity multiplied by the call duration. Fora given set of such policing parameters, the charging algorithmis still insensitive to traffic volume, and Figure 1a) applies alsoto this kind of charging.

The main drawback with a charging scheme based only on traf-fic contract is that the users must pay for an assumed amount oftraffic, whether it is used or not. So this scheme gives no reduc-tion in the charge if the traffic sent is substantially lower thanthe traffic contract. On the other hand, this scheme gives astrong incentive to choose a contract that closely matches thegenerated traffic.

2.3 Algorithms based on traffic volume

A purely volume based charging scheme is illustrated in Figure1b). The main objective to charge according to traffic measure-ments is that users should pay an amount that somehow reflectstheir real usage of network resources. To obtain this goal differ-ent measurements are possible, ranging from plain cell counts tosophisticated estimations of different traffic parameters.

The main objection against only using measurements for charg-ing in ATM networks is that this type of algorithms are inde-pendent of the traffic contract. This means that users have noincentive to choose a traffic contract that tightly matches thesource characteristics. It is likely that the CAC procedure to agreat extent will depend on the contract parameters, so for theATCs bounded by some CAC algorithm (mainly DBR andSBR), the requested resources at call set-up should somehow bereflected in the charge. For the ABR services the situation is dif-ferent, where apart from the minimum guaranteed cell rate(MCR), the ABR users may share the unoccupied resources inthe network. So for ABR the traffic beyond the MCR could wellbe charged according to only traffic measurements such as cellcount.


Figure 1 The effective bandwidth and charge as a function ofthe mean bitrate for some different charging schemes

Cost

Mean rate

Cost

Mean rate

Cost

Mean rate

Figure 1a. Volume insensitive algorithms

Figure 1b. Volume-based algorithm

Two different tariffsfor traffic with agiven peakrate

Effectivebandwidthcurve

m1 m2

1tariff

2tariff

Figure 1c. Kelly's charging algorithm

peakrate

2.4 Algorithms based on traffic contract, callduration and traffic volume

As mentioned above it is likely that a satisfactory chargingscheme for ATM will contain all the three aspects mentionedabove. One part of the algorithm will be a flat rate independentof the traffic sent into the network. In addition, there will be atraffic dependent charge based on both traffic measurements,call duration and traffic contract parameters. It is also possiblethat other traffic parameters (which are not part of the trafficcontract for the different ATCs), such as an estimated mean

rate, will be a part of the charging scheme. Such approaches,based on the concept of effective bandwidth, have been investi-gated in the CA$hMAN project where two charging schemesare implemented in the testbeds. These are Kelly’s chargingalgorithm and the Two tax-band algorithm, both describedbelow.

2.4.1 Kelly’s charging algorithm

This algorithm is based on cell count and time duration plus apossible extra set-up charge. The cost of a connection is thengiven by the formula:

COST = CTT + CVV + CA

where T is the time duration and CT is the charge per time unit,V is the traffic volume and CV is the charge per volume of traf-fic carried, and CA is a fixed charge per call. The tariffs CT, CVand CA will depend on the traffic contract parameters negotiatedat call set-up, but they could also depend on other traffic param-eters, for instance an estimate of the actual mean cell rate of aconnection. One possible choice of the parameters is derived byKelly [1] and is given in Appendix I. For this particular choiceof parameters the tariff will encourage the users to give the bestestimate of the mean rate m to get the cheapest charge. How tar-iff and mean rates relate in the Kelly charging algorithmscheme is illustrated in Figure 1c): Users submitting traffic withmean rate close to m1 should prefer Tariff 1 to Tariff 2.

The charging scheme above may also be a good choice for ABRservices. The essence in this approach is that the traffic up toMCR is charged at a flat rate per time unit, and the volumeabove the MCR is given a rate per volume which should be lessthan the corresponding rate per volume for DBR and SBR traf-fic. Anyhow, the result will give a charging function which ison the same form as above, but the tariff parameters CT, CV andCA must be chosen differently for ABR services.

2.4.2 The two tax-band charging scheme

A charging scheme based on Kelly’s algorithm will only dependon the duration of a connection T and the volume V. This meansthat two connections with the same duration and volume will becharged equally, so there will be no extra charge for burstinessby applying this formula. For example, an on/off source withdifferent burst length will be charged the same if it has the samepeak and mean rate, in spite of the fact that the source with thelonger bursts will be more difficult for the network to carry.

One way to take burst length into account is to implement somekind of burst detection by sorting the traffic flow into two cate-gories – one for high intensity and another for low intensity.Then, by charging the first category higher than the other one,we are able to better identify and charge bursty traffic accordingto resource usage.

The above charging scheme is implemented by defining a smallmeasuring interval t, over which we measure the number ofcells carried Vt, and classify the interval of being of type I or IIaccording to Vt ≤ K or Vt ≥ K where K is a constant dependingon the interval t. Then the two tax-band charging formula takesthe form:

COST = CT1T1 + CT2T2 + CV1V1 + CV2V2 + CA

where T1 and T2 are the total duration of intervals of type I andII; (T1 + T2 = T). Similarly, V1 and V2 are the total volumes ofcell generated in type I and II; (V1 + V2 = V). The value of K ischosen to be less than ht (which is the volume generated by thepeak rate h in the measuring interval of length t, and one possi-ble choice is to set K = 1/2ht). The tariffs CT1, CT2, CV1, CV2and CA may be given as functions of the contract parameters,but also additional parameters may be included. One set of tar-iffs may be obtained by a similar procedure as for Kelly’s algo-rithm; however, the expressions get more involved and it alsorequires that the users must specify their estimate of the meanvalues in each tax-band and also an estimate of the portion oftime in each tax-band.

2.4.3 The CA$hMAN approach to usage basedcharging

Charging and tariffs are not present only to collect money fromthe customers, but can also be actively used to derive informa-tion suitable to optimise the network utilisation. For instance, bytailoring the tariffs to the choice of traffic contract parameters,the users are encouraged to choose tariff and contract para-meters that matches his traffic profile (see Section 2.2). Withoutsuch a linkage, the users do not have incentives to give adequatesuggestions for contract parameters. By making reasonablygood choices, both parties may profit. The users by havinglower bills, and the network operator by using the informationto optimise network utilisation. However, contract parametersonly define the upper limits for the traffic a customer is allowedto send, and do not reflect the traffic he actually will send. Anidea worth following, is therefore to investigate whether it ispossible to derive more information from the users – informa-tion that can be used to further increase network utilisation. Aspointed out in Section 2.4.1 the mean value of the traffic hasbeen proposed to be such a useful parameter.

It is not obviously clear that a user is able to estimate the meanvalue of his intended traffic. But on the other hand, whateverknowledge the user posesses may be of value to the operator inoptimising network utilisation. The user may also be motivatedto improve his estimates if he receives the right incentives andhe may learn by experience, possibly aided by suitable “intelli-gent” functions implemented in his communication equipment.The task to be solved is to implement a charging scheme thathas the necessary incentives for a customer to provide the net-work operator with adequate information. The aforementionedalgorithms (Section 2.4.1 and 2.4.2) have the needed propertiesthat may encourage the user to give good estimates.

As indicated, there are many uncertainties present in the mostadvanced implementation of UBC, and there are a number ofopen issues to be investigated. For instance, we do not know

• Whether the potential gain to the network is large enough

• Whether we can load the customers with more ‘duties thanalready done’

• Whether the incentives for the customers are large enough

• Whether the incentives can be presented in an understandableformat

• Whether it is technically possible to predict mean rate esti-mates good enough to be used in CAC in order to increasenetwork utilisation.


To answer these questions the CA$hMAN project has imple-mented an experimental platform with the most fundamentalfunctionality included. The platform incorporates the minimummanagement functions and the capability to present on-linecharging information to the user. The experiments are describedin more detail in Chapter 4.

3 Aspects of billing support

In order to introduce and conduct any usage based chargingscheme, a system that supports the operation and the manage-ment of the entire billing process is required. The billing systemis a vital part of an operator’s total network solution as it pro-vides the means for securing revenue. The billing system is alsoessential in the handling of the customer care relation, e.g.traceability and logging of charging events are important indefence of customer dissatisfaction with possible subsequentinvoice complaints. Operators strive to meet customer demands.New requirements on billing options are important to satisfy asthe bill forms an important interface between the operator andthe customer. With the introduction of new services, one suchcustomer requirement could be a preference to receive one con-solidated bill that covers the whole range of used services. Fur-thermore, marketing analysis (e.g. who are the big spenders andwhat services are they using?) and tariffing forecasting tasks(e.g. modelling of the revenue distribution of introducing newtariffs) may be alleviated by means of functions of the billingsystem.

A complete and comprehensive billing system is not yet avail-able for ATM. From a commercial point of view the technologyis relatively new, and a flat rate policy has initially been adopted– a flat rate charging scheme does not necessarily require anextensive and complicated billing system. With the introductionof more sophisticated charging schemes a more elaborate sys-tem is required. Increased competition resulting from the de-regulation of the telecommunication market may create a driv-ing force for the development and deployment of more sophisti-cated charging methods. A system that supports such deploy-ments may represent an important competitive edge towardscapturing market segments and new customers.

In general and in brief, the billing process in a telecommunica-tions system may involve

• Measuring the usage of offered and utilised services

• Collecting and transferring the measured usage to an appro-priate tariff system for charge calculations

• Associating calculated charges to the rightful customers, and

• Forwarding charges to an invoice system for subsequent pay-ment processing that eventually turns into revenue.

This process is also applicable to ATM. The following sectionsdiscuss some of the functionalities and components required atdifferent phases of the billing process relating to usage basedcharging schemes for ATM.

3.1 Components for realising usage basedcharging of ATM-Services

An ATM network comprises a set of interconnected switchesthat route the ATM traffic streams to their various destinations.

ATM bearer services are offered to the customers at the fringesof the network; with access points that are typically attached tosmall edge switches. Within the core of the network a smallerset of high capacity backbone switches allow traffic to be con-centrated and aim to assure reliable transport and interconnect-ivity across the network.

The provisioning and control of ATM bearer services as well asthe management of the ATM network infrastructure require aset of support systems. The functions that are required may beclassified into two categories: those that address the operationsof the network and those that address the related business pro-cess. The former category is often termed the Operations Sup-port System (OSS); the latter the Business Support System(BSS). The classification of these functions target two of themain areas of concern for the operator in the context of serviceprovisioning and customer care, namely, how to automate thehandling of the technology in question (e.g., fault managementand configuration management) and how to automate parts ofthe business process (e.g., customer handling, tariffing andinvoice processing, marketing analysis applications).

A total billing solution for an ATM network must be basedupon the usage measurement functions provided by the networkelements. It must also be integrated with the OSS for surveil-lance and configuration management and with the BSS for post-processing and tariffing purposes. Some central problems thatwill be encountered in assembling such a system are addressedbelow and outlines of possible solutions are discussed.

To handle billing, among the features that are required is a sys-tem that collects the usage data from the network elements andwhich adapts and relays the collected information to the post-processing systems contained within the BSS domain. Such abilling accessory is essential in interfacing the diverse usagemeasurement capabilities of the network elements. The mea-surements and the subsequent data generation (DG) of billinginformation is performed at the switches. For each active ATMconnection, billing data is associated with and represented inCall Detail Records (CDR). These are accumulated in files,foreseen to be at 15 minutes or hourly intervals. One of thetasks of the billing auxiliary server alluded to is to collect thesefiles from the switches at regular intervals. Connectivity to andfrom the billing server with the network elements is thereforerequired – a virtual billing network takes form, as depicted inFigure 2. The server must be able to collect different file andCDR formats and handle different decoding rules to conduct thefunction on a possible plethora of different switches containedwithin the network. A proprietary solution, coupled to a specificvendor switch element, will not do. As ATM networks evolve,both in size and complexity, and as new and unforeseen tech-nology is introduced, novel switch billing capabilities will un-doubtedly enter the picture. In this context, the billing serveradaptability is paramount. It must be able to handle the varia-tions and the mixtures of old and new network equipment in aflexible manner. The alternative, replacement or even altera-tions of billing components, as new switch technologies areintroduced, is unquestionably an expensive and arduous option.Adaptability is also important in the context of interfacing theBSS. Existing operators that already offer other services (e.g.telephony) have their own post-processing system that has beenexhaustedly exercised and well-carved to meet the particularneeds of the operator. Integration with the existing post-process-ing system may very well be desired. The implication is that the


billing server must transform or convert the collected CDRs to aformat and use encoding rules that are appropriate for the partic-ular system. There are numerous such post-processing systemsavailable and the billing server must therefore be able to offer ageneric solution to this mediation function.

3.2 The bandwidth diversity aspect

ATM is a broadband technology, promising high capacity trans-mission services. Collecting the measured usage of these ser-vices pose a challenge to the capacity and to the storage capabil-ities of the entire billing system. The traditional telephony callis typically of short duration, in the order of minutes. The mainunit of usage measurement is the time period in which a calloccupies network resources. In addition to the usual time zonesand distance tariffs, this time period historically forms the basisfor usage based charging in telephony. The story may be morecomplicated for ATM.

In ATM, the granularity of the measurement unit for a volumebased charging scheme1 – the cell – is next to infinitesimalcompared to the bandwidth that is available for use. ATM per-manent virtual connections, which are normally held for a longduration, often months at a time, will soon exhaust the switchaccumulators that keep track of the volume usage if such a vol-ume based charging scheme is deployed. There is a need torelieve the switches and drain the accumulators at regular inter-

vals. The billing server cannot simply collect the desired CDRonce at the time a connection is released, which could be thecase in plain telephony systems. Instead, partial complete CDRsmust be collected regularly (e.g. every 15 minutes or at hourlyintervals) until the connection is taken down. Additional and tosome extent superfluous CDRs will therefore be generated; andat some intermediate stage an assemble function must be intro-duced to merge the related, although partial CDRs.

With the advent of switched virtual connections a new twist ofthe capacity problem takes form. The duration of switched vir-tual connections are found at the other end of the time scale topermanent virtual connections. It may be envisaged that switch-ed virtual connections will have holding times that span periodsof perhaps a few seconds instead of months. Noting that the fre-quency of the CDR generation for these types of connectionshas an inverse proportion to the small duration, it becomesapparent that the number of CDRs that need to be collected willput a serious workload on the billing system. The scene istroublesome but not entirely unmanageable. By means of net-work planning, peak hour predictions, and by estimation of thecustomer arrival process a certain upper threshold on the busyhour call attempts may be approximated. Clearly, the billing


1 Measured volume may be one of the components in a usagebased charging scheme.

Mediation System (MS)

• Collector system• Queueing system• Transformation handling• Filtering and correlation

DG

DG

DG

CDR

CDR

CDR

ATM Network

An ATM network offers services at thefringes of the network. Usage measure-ments are performed at the switches.The Data Generation (DG) part gathersthe usage data pertaining to utilisedservices into Call Detail Records (CDR).These are accumulated into files andlater collected by the Mediation System.

Mediation System

The Mediation System (MS) function asa collector of the CDR files originatingfrom multiple service sources andmultiple switches with various fileformat and detail combinations. Theinformation is consolidated, trans-formed, and later relayed to theBusiness Support System domain forsubsequent processing.

Support Systems

Two categories of support systems assistin the provisioning of services: The Oper-ations Support System (OSS) and theBusiness Support System (BSS). TheOSS supports management of the serviceand billing network technology, whereasthe BSS supports the business process.The BSS includes tariffing systems.

Operations Support System (OSS)

• Fault management• Configuration management• Performance management• Security management• Accounting management

Business Support System (BSS)

• Customer contract evaluation• Customer directory assistance• Tariffing and invoice system• Customer behaviour analysis• Sales and marketing analysis

Management control

Consolidated billing data

Figure 2 A Billing System

system must be designed to sustain this threshold rate. Further-more, it must cope with exceptional conditions. That is, eventsthat cause the threshold to be exceeded. Such conditions cannotbe excluded since an estimate of the threshold is just that, anestimate; and there will be variances and uncertainties bound toit. On the average, however, the frequency of CDR generationshould be below the threshold as peak hours is not the norm.Processing overload conditions may be effectively controlled bymeans of buffering and queuing mechanisms: During overloadconditions CDR files are placed in a queue and during idletimes those that have been waiting are served. This mechanismforms a trade-off between capacity and buffer storage at theexpense of delay in the billing processing.

3.3 The heterogeneous service aspect

Unlike telephony, ATM offers various bearer services with dis-similar characteristics and quality assurance. The different ser-vices give rise to different needs concerning the location of theusage measurements and consequently the CDR generation for aconnection. It is also very much dependent on the charging pol-icy of the operator. Take for instance the case of charging,based upon usage, for a VBR service. One approach is to mea-sure the traffic that enters the network, i.e. at the ingress. SinceATM connections are bi-directional, and may even be asymmet-rical with respect to bandwidth requirements, two measurementpoints must be considered, one at each entry into the network,as depicted in Figure 3a. This will produce two distinct and dis-tributed CDRs, one for each direction, that must be collected,correlated, and assembled to enable the details of the bi-direc-tional connection to be associated with a billed party. Why notmeasure the usage at one point only so that just one CDR iscreated, thus avoiding the need to perform the additional pro-cessing? This could lead to an inconsistent charging policy:charging based upon what enters the network for one directionand charging based upon what exits the network for the otherdirection seems like an odd course to take. Suppose that the twosources involved in a connection transmit identical traffic. Bymeasuring the two streams at one point only (see Figure 3b),one may end up with a bizarre situation in which the two charg-es will differ (since cells may be lost due to congestion and themeasured usage will therefore be dissimilar for the two direc-tions). Measurements for a VBR service should therefore be dis-tributed, and either take place at ingress or at egress of the net-work depending on an entry or an exit charging practice. Amore elaborate policy might be devised to differentiate thecharge for cells of different cell loss priorities. Cells that aretagged at the entry of the network (and hence are given lowerpriority than those that are not) are not assured deliverance tothe destination in case of network congestion. One policy is tocharge only for those cells that are delivered. Measuring only atthe ingress (after the UPC function) is insufficient in this case.In such a scheme one may require that all four measurementcombinations be taken into account (see Figure 3c). For theVBR service category, low priority cells are treated in a besteffort service fashion. There is also a separate service categoryfor best effort traffic, namely UBR, that treats all cells with thelowest priority. Given that all cells are treated equally, it is onlyrequired to measure at the two egress points to handle a schemethat aims to charge for the delivered volume. There are stillother cases. If one considers multicast services the number ofCDRs that must be collected proliferates. And that is not the endof the story. There is also the case of performing measurements


CPE1

Measurements of usage from the CustomerPremises Equipment (CPE) at ingress andat egress and at both access locations.

CPE2

CDR=Ingress CPE2Egress CPE1---


CPE1

Measurements of usage from the CustomerPremises Equipment (CPE) at ingress only.The data generating part of the accessswitches accumulates informationpertaining to the service in partial CallDetail Records. To address the billedparty and consolidate the usage inform-ation gathered, the partial records mustbe correlated and assembled.

CPE2

CDR=Ingress CPE1---

CDR=Ingress CPE2---

Correlation and assembly

CPE1

Measurements of usage from the CustomerPremises Equipment (CPE) at ingress andat egress, one location only.The data generating part of the accessswitches gathers the usage data in a CDRwhich is later accumulated in a file.

CPE2


Figure 3a Usage measurements at ingress only, two locations

Figure 3b Usage measurements at ingress and egress traffic,at one location only

Figure 3c Usage measurements at two locations, both ingressand egress and at two locations

Correlation and assembly

at the interface between operators for traffic that is in transit.There is a variety of possibilities with various degree of sophis-tication; and they all consume capacity of the billing system. Itis important, however, that the billing system does not constrainthe choice of charging policy. Ideally, there should be supportfor all combinations by enabling the desired and appropriateconfiguration to be set from the Accounting Management func-tion of the OSS.

3.4 Reliability aspects

The reliability of such a system is vital: if the billing systemmalfunctions, the operator will be unable to collect paymentsfrom the customers for services offered and carried. The reli-ability does not merely depend on the mediation functions andthe BSS but, since it depends on the virtual billing network, alsoon the ATM network. For instance, one can easily imagine ascenario in which one of the network links goes down. If such abroken link was used to carry billing information streams, fromone or more switches, the function of the billing system is dis-rupted. There are various technology solutions to this one sce-nario, among them is protection switching. In a network withprotection switching, nodes that detect severe link failures willre-route the traffic using other available links. But suppose thenthat the billing mediation function breaks down and that thereare no backup solutions. What policy should the operator pur-sue? One could argue that, since the operator cannot get paid forthe traffic being carried, the result should be revocation of theservices, until the billing system is operational again. The alter-native is to offer the services free of charge in the recoveryperiod. The latter might seem like the most obvious policy, butthere is a pitfall: customers might grow suspicious of the opera-tor’s billing system; and the very last thing an operator wants isbreeding customers that question the operator’s invoice func-tions. These customers will soon enough shun the particularoperator and move their business to other providers2. Fault situ-ations are not the only events that may endanger the operationalstate of the system. For instance, bringing new functionality orequipment into service must be performed in such a way as notto jeopardise the ongoing billing process. The key issue is thatonce a billing system is established and made operational itmust remain operational, lest the operator faces the possibilityof loss of revenue and customer dissatisfaction. The depictedscenarios aim to stress the importance of reliability and reliabil-ity on all components introduced to handle billing functions.

As is the case for any new technology system reliability cannotbe guaranteed until initial unforeseen problems have been eradi-cated. ATM is a relatively new technology and ATM networkshave yet to be exhaustively exercised in commercial environ-ments. With heavy investments in the introduction of novel ser-vices using new ATM technology, a provider must clearly avoiddissatisfied customers. New ATM services may promise newrevenue and more customers, but if the billing system is unreli-able or even unpredictable both revenue and customers will belost. Reliability is crucial in any telecommunication systemcomponent, but perhaps even more so in the context of billing.The novelty of the ATM technology makes this a difficult area.

3.5 Charging and billing of composite services

The discussion so far has been confined to ATM in isolation.Taking a wider perspective, tomorrow’s broadband networkswill offer other and more intelligent services (e.g. Frame Relayand ATM Interworking, LAN Emulation, Video Conferencing,Shared Multimedia Applications, Voice over ATM, and so on),in addition to the pure ATM access and transmission function-ality. The networks will also reach out and extend into otherrelated technologies (e.g. residential broadband access). Indeed,ATM might very well become the common transport technol-ogy that other services and higher layer protocols will make useof, but these latter will be charged differently. The introductionof these value added services will require additional systemcomponents, such as access devices, application servers, gate-ways, and interworking units, that may have their own chargingcapabilities. A tremendous challenge is to integrate these com-ponents in a common billing system framework.

3.6 Perspective

A system to support billing is complex and expensive. The situ-ation is no different for ATM. This chapter has given a briefoverview of some of the issues that must be considered in con-structing a billing system for usage based charging of ATM ser-vices. There are other topics that are closely related and to someextent coupled to the billing domain that have not been broughtforward. One such topic is fraud detection, another is customersubscription handling. Security mechanisms is yet anotherimportant subject. Billing spans across many complicated areas.Billing systems, in the context of ATM, are still immature and alot remains to be explored. This is the crux of the problem, andyet, it is what makes it so important. There may be good com-mercial opportunities in this field for early vendors of ATMcharging and billing systems.

4 Experiments in CA$hMAN

In Chapters 2 and 3 we have discussed different charging meth-ods and considerations which have to be taken into account inconstructing a billing system based on the ATM technology.Constructing a total billing system is a major undertaking.Within the CA$hMAN project limits only pieces of such a bill-ing system would be implemented. The aim of the project, asstated earlier, is to implement different charging schemes andinvestigate their feasibility through conducting user experi-ments. By doing this the projects gain experience of the imple-mentation costs involved by implementing different chargingschemes, how it effects the hardware construction and networkutility and how users react to different charging policies. In thischapter we summarise what kind of experiments have been con-ducted in the first round of the project, what our goals for theexperiments were, and the results obtained within the first pro-ject year. A brief overview of the experimental platform is alsogiven to describe the experimental environment.

4.1 Background

In the CA$hMAN approach the charge of a call is due to timeand volume usage based on a given traffic contract. By usingthe experimental platform developed, the user has an opportu-nity to also offer an estimate of the mean bitrate of traffic to be


2 Billing is the main source for customer dissatisfaction. Oftenas much as 50 % of customer complaints stem from billingfunctions.

generated. This mean rate estimate actually constitutes a choiceof tariff for the call to be made, so within one set of traffic con-tract parameters, the user may choose among many mean rates,i.e. different tariffs. A typical call can go like this: The userdecides what kind of traffic the call will generate onto the net-work and chooses a tariff that he thinks will give him the lowestcharge. If the user has chosen an expensive tariff, he should beable to renegotiate the choice of tariff. This gives the user theability to influence on the charge for every call by selecting tar-iffs based on e.g. QoS limits, bearer capability, contract para-meters, bandwidth, CDV restrictions, etc. Since the user earlierdid not have this opportunity, the proposed charging conceptintroduces many questions of how a user understands the newconcept and how he reacts to it.

In the user experiments, we wish to examine user behaviour andthe possibilities an operator has to influence the user in order toachieve good network utilisation. Some of the questions we tryto answer regarding user behaviour are:

• How does the user understand the relation between a tariffand the traffic he is going to send/receive? What does the userknow about traffic characteristics, i.e. usage of bandwidth?

• How does the user understand the relation between time dura-tion cost versus cost per volume of data? Is it cheaper to sendan amount of data with short time duration and high band-width, or vice versa?

• Does the proposed charging scheme give the user a strongincentive to choose a tariff that gives the network an optimalutilisation?

• How willing is the user to trade quality against cost?

As pointed out earlier we also want to investigate the implica-tions the charging system has on network utilisation, and howthe charging algorithms and billing management system operateas a billing architecture. Questions we try to answer in theseexperiments are:

• What cost reductions can be gained by influencing the trafficprocess, i.e. shaping?

• How well does the proposed charging algorithm reflect thecost of network resources?

• What knowledge have we gained regarding implementationissues of the experiment platform?

• How well does the charging algorithm reflect the networkcost of different traffic characteristics?

Both types of experiments are vital to carry out. The experi-ments regarding network utilisation and architecture can, andshould be, conducted in a controlled environment. These experi-ments have therefore been carried out within a laboratory en-vironment. The user trials, however, are more difficult to con-duct. To get good and sound results many “ordinary” none-ATM knowledgeable people should be experiment users, andtheir choice of tariff should affect the payment of a service.Today, this is difficult to achieve. Also the user behaviour ex-periments have therefore taken place inside a laboratory en-vironment. One possible way to conduct real user trials is tomake use of ATM National Host networks. Both in Norway,Switzerland and the Netherlands testbeds have been installedwith the ability to connect to the National Host networks. Theaim would be to interconnect a group of experimental users at

each site, who are willing to take part in the trials. Vital condi-tions for carrying out these user trials are that participating net-work operators are willing to allow these kinds of trials on theirnetworks, and that representative and willing customers may befound for experiments.

4.2 The CA$hMAN charging experimentalplatform

An important idea in the formulation of the CA$hMAN projectwas the requirement to validate the implementation feasibilityof the theoretical charging models developed in the project.Among the evaluation criteria for any charging model, the tech-nical implications with regard to charging management will bean important factor. Making prototyping of the charging man-agement systems will also make possible the investigations ofuser behaviour and user acceptance. General sources of inform-ation for the system described in this section are [9] and [10].


Manager

Informationrequired

to issue bill

Charging management centre

Call detailretrieval

primitives

Tariff dataManager

Manager MIBAgent

Configurationprimitives

ManagerEPCU

Low levelresource

usageinformation

Low levelconfiguration

primitives

Configuration dataCall detail recordsin standardisedformats

Operator terminalwith graphical

"Advice of charge"user interface

Networkelement

User trafficto be charged

Figure 4 The CA$hMAN charging management system

Motivated by the above, a substantial effort has been made inthe project to realise prototype charging management systems.The functional structure of this system is shown in Figure 4.

Describing the system ‘bottom up’, the user generated trafficenters the ‘Ericsson Policing and Charging Unit’ (EPCU). Thisunit carries out the ‘low level’ operations involved in the charg-ing. For the simplest charging algorithm, these functionsamount to cell counting and time measurements. Such measure-ment functionality would be a natural part of a switch in a realsystem, but in the CA$hMAN project the functional entity isrealised in a separate piece of equipment developed from thegeneral Ericsson exchange terminal unit ET-34 (see photographin Figure 5). For the 34 Mbit link rates used for the transmissionthrough the EPCU, implementing the time critical chargingoperations by general off-the-shelf processors is sufficient tofulfil real-time requirements. The EPCU essentially consists oftwo exchange terminals linked together through a specialisedbackplane. By combining two exchange terminals, the EPCUthus forms a bi-directional integral ‘low-level’ charging device.

Management systems work on abstract data models representingthe network, so an abstract model of the charging unit is con-structed and realised as a ‘Management Information Base’(MIB). A set of primitives are defined, operating on this base.These may typically be defined as ‘Set()’ and ‘Get()’ operationson the elements of the base. The ‘Agent’ receives these primi-tives from the ‘Network management centre’, and must ensurethat the state of the MIB is consistent with the actual state of theEPCU. To this end, a specialised ‘low level’ control interface isdefined between the Agent and the EPCU. As the Agent isimplemented on a general purpose computing platform, separatefrom the EPCU, it may be described as a ‘proxy’ agent. For rea-sons of simplicity, consistency and dependability, a preferredsolution would have been to integrate the two.

Overall control of the charging system is executed from the‘Charging management centre’ via the operator terminal. From

the operator terminal, it is possible to configure charging andpolicing parameters for a connection, choose between chargingalgorithms and select tariff profiles. A ‘Manager’ is responsiblefor maintaining the tariff data base, and to handle the communi-cation with the proxy agent. In the CA$hMAN implementation,the operator terminal is equipped with functionality to performcontinuous advice of charge, and also graphically display run-ning billing information for a connection. As the complete man-agement system is implemented in a Unix environment runningthe X-protocol, the operator terminal may be displayed on anyX-server, e.g. also the X-server on which the user application islocated and executed. In this way, the user may get the runningcharging information displayed, simulating an ‘advice ofcharge’ service. In the development of the project, a more real-istic advice of charge service based on user/network signallingis under consideration.

The above described management network is of course a simpli-fication of a realistic charging management network. As pointedout previously, there will be a large number of charging agentsto be controlled, charging information for a single bi-directionalconnection will be generated at different locations and must beassembled consistently, and the configurations of charging andpolicing parameters for connections must be consistent. Thediversity of services offered in ATM networks will thus givenew challenges for management network designers. Not theleast of these challenges will be the dimensioning and traffichandling in the charging management network itself! Even so,the efforts made in the project have contributed towards demon-strating the feasibility of constructing a management model forusage based charging, and also shown that the single prototypecomponents required to implement the charging model may berealised within the limits of a moderately sized project team.

4.3 Experiment results

Results from the first round of experiments in the CA$hMANproject are documented in [11]. In these experiments the simplecharging scheme by F. Kelly was evaluated against a backdropof general and more detailed questions as stated in Section 4.1.The questions addressed the following areas of study:

• Properties of the scheme in relation to implementation

• Relation of the scheme with network operation

• Perception of the charging by user and customer.

Within all three areas useful information has been gathered.

In relation to the first and second bullet, developing and experi-menting with the tools provided a number of insights intoimplementation issues. Some of the issues evaluated were

• Speed of calculation and transfer between hardware and soft-ware components

• Implementation complexity within hardware and software

• Construction design of a user friendly graphical user inter-face.

How the charging scheme reflects network operation and utili-sation has been investigated by simulation with real traffictraces. Different experiments involving shaping and multiplex-ing of connections have been carried out, which verify the


Figure 5 The EPCU (Ericsson Policing and Charging Unit)

theoretical calculations. An interesting result is the reduction ofcharge caused by shaping, thus demonstrating that the chosencharging algorithms provide an incentive for users to improvecharacteristics of submitted traffic.

Regarding the third bullet above, many interesting results wereattained. The users participating were non-project colleagueswilling to run the CSCW (Computer Supported Co-operativeWork) application Isabel and give us feedback regarding theirunderstanding and feelings about the CA$hMAN approach tocharging ATM services. The main results from these experi-ments were

• A user has a strong desire to adopt his traffic characteristicsto suit the charge he is willing to pay, e.g. by means of shap-ing.

• A user prefers tariff choices explained by words in an under-standable way, instead of mathematical figures.

• Some services are more difficult to predict than others as towhat would be the best tariff choice.

• A user needs some knowledge of the traffic characteristics ofthe application he is running.

In the next round of experiments the project will continue toexamine many of the same issues taking into account differentcharging schemes, more elaborate network management for thebilling system and also more automation regarding advice to thecustomer about choice of tariffs.

5 Consequences for the networkoperator

Charging policy is one of the most important tools available fora network operator to control the utilisation of network invest-ments. Tariffs may be used to differentiate between customers.For instance, if he wants to attract the big market of householdcustomers, he creates a tariff that matches their situation asregards price profiles. High volume users or service providerswill have other demands, and must be approached by other tar-iffs. We do not intend to give answers as regards what pricelevel or tariff structure should be used in special cases. Themain factor that establishes prices and tariffs, is the market.

The choice of charging policy depends upon a number of fact-ors. For instance:

• Tariffs offered by other competitors

• The present utilisation of the network

• A general wish to attract/exclude special kinds of usage

• Coverage of investments and network maintenance

• The run for even higher profit

• Improve utilisation of network resources

• Avoid cannibalisation

• Protection against resellers

• Network stability

• Network flexibility

• Cost of implementing and running billing/charging systems.

The charging algorithms mentioned in Chapter 2 influence theabove factors in various ways. Maybe none of them match thecomplete set of requirements a network operator may have.There is, however, a fundamental difference between flat ratecharging (FRC) and the other charging schemes. While FRCdoes not require any measurements or logging functionality, allthe others do. This leads to a dilemma. If one believes that theadvantages of the FRC’s simple scheme more than compensatesfor missing properties (Chapter 4.1), one may decide to build anetwork without charging functionality. However, charging pol-icy depends on the market situation. What one actor does,influences immediately on the others. Consequently, to be ableto respond to changing market impulses, one has to be preparedfor shifting views of charging. This means that one has to makea choice flexible enough to respond to each demand from themarket. If first excluded, it may be impossible to implementusage sensitive charging schemes at a later stage, due to techni-cal limitations.

If the technical infrastructure for a complete billing system mustbe implemented whatever the choice of charging policy mightbe, then the main difference between FRC and the otherschemes is much less apparent.

6 Concluding remarks and future work

Charging and accounting in ATM networks is a challenging andimportant field. A goal is to obtain a stable and well performednetwork, both from the user’s and from the network operator’sperspective. To achieve such a goal it is necessary to gain ex-perience both from theory and experiments. In the CA$hMANproject these objectives are strongly focused. The results obtain-ed so far supports the idea of usage based charging (UBC). Twoschemes have been implemented in the laboratory testbeds justi-fying that UBC may well be introduced in a commercial ATMnetwork. Due to the lack of paying experimental users the effecton the revenue for the network operators and on the networkperformance is difficult to predict, however, we believe thatboth the customers and the network operators can benefit fromcharging based on UBC.

Still, there are open and yet unsolved questions to be addressed.One example is how to charge the ABR services. Another areaof interest is the question of making some advice of tariffs forthe users. The network may possibly guide the user to obtain thebest tariffs depending on his traffic profile, and thus raise thequestion of renegotiating of traffic contract and tariffs for on-going connections.

The questions raised above, which are important both from theoperator’s and customer’s perspective, are among the issues thatwill be examined in the future CA$hMAN work.

References

1 Kelly, F P. Tariffs and effective bandwidths in multiservicenetworks. In: The Fundamental Role of Teletraffic in theEvolution of Telecommunications Networks. Proceedings ofthe 14th International Teletraffic Congress-ITC14, p.401–410. Laboutelle, J, Roberts, J W (eds.). Amsterdam,Elsevier. Antibes Juan-les-Pins, France, 1994.


2 ITU. Traffic control and congestion control in B-ISDN. ITURecommendation I371. Geneva, 1995.

3 Hui, J Y. Resource allocation for broadband networks. IEEEJournal on selected Areas in Communication, 6, 1988,1598–1608.

4 Gibbens, R, Hunt, P. Effective bandwidths for the multi-type UAS channel. Queueing System, 9, 1991, 14–28.

5 Guerin, R, Ahmadi, H, Naghshineh, M. Equivalent capacityand its application to bandwidth allocation in high-speednetworks. IEEE Journal on selected Areas in Communica-tion, 9, 1991, 968–981.

6 Courcoubetis, C, Weber, R R. Buffer overflow asymptoticsfor a switch handling many traffic sources. Journal ofApplied Probability, 33, 1996.

7 Songhurst, D. Charging schemes for multiservice networks.In: Proceedings of the thirteenth UK Teletraffic Symposium(IEE), 1996.

8 Prycker, M. Asynchronous Transfer Mode ; solutions forbroadband ISDN, third edition. Prentice Hall, 1996. ISBN0-13-342176-6.

9 CA$hMAN Consortium. Design specification of testingplatform for round 1 equipment. ACTS Project AC-039,Deliverable d04, 1996.

10 CA$hMAN Consortium. First years result on pricing mod-els. ACTS Project AC-039, Deliverable d06, 1996.

11 CA$hMAN Consortium. Evaluation of experiments forround I and report on selection of user groups. ACTS Pro-ject AC-039, Deliverable d07, 1996.

Appendix I. Effective bandwidth

Because ATM allows statistical multiplexing the networkresources may be divided between the connections (users) in astatistical manner. This means that each connection on averageneeds less capacity than the peak rate; on the other hand, anallocation after mean rates will be too optimistic (in almostevery case, also for CBR sources). The concept was introducedby Hui [3] for bufferless multiplexing, and further elaborated byGibbens and Hunt [4], Guerin et al. [5] and Courcoubetis andWeber [6] to also cover multiplexers with finite buffers.

The idea of effective bandwidth is to assign a measure ofresource usage which adequately represents the trade-off be-tween sources of different types, taking account of their varyingcharacteristics and also the requirements on QoS.

Based on mathematics, the effective bandwidth is definedthrough the arrival process of a source. If X(t) represents thework generated by a source (i.e. the volume of bits arrived) inan interval of length t we define the transform

α(z,t) = 1 / zt logE[ezX(t)]

The transform above has some nice features which are import-ant in practical use. First of all, the effective bandwidth definedabove is linear, which means that the effective bandwidth of asum of two or more independent cell streams is the sum of theeffective bandwidth of each individual stream. Secondly, theeffective bandwidth lies between the mean and the peak arrivalrate of a source (up to the time t). The time parameter t andspace parameter z are left undefined in the definition above andwill be decided upon by the resource parameters such as capac-ity, buffer space and scheduling policy. To see this we refer tothe following important result concerning the cell loss L(c,b) ona link of capacity c using a cell buffer of length b [6].

Let

αT (z, t) = Σi

αi(z,t)

be the sum of the total effective bandwidth on the link. Then thecell loss is approximated by the following expression:

logL(c,b) supt

infz

[ztαT(z, t) – z(ct + b)]

Let z*,t* be an extremating pair in the expression above, then aQoS demand that the cell loss is below e-γ will limit the effect-ive bandwidth in the following simple way:

αT(z*,t*) c* = c + 1 / t* (b – γ /z*)

Stated in words: The sum of the effective bandwidths must bebelow an equivalent link capacity.

Although the expression seems to be linear in the sum of theeffective bandwidths the relation is not quite easy to use as aCAC algorithm, since one needs to re-calculate a new pair ofz*,t* for each new call arrival.

Another fundamental issue is the problem of estimating αT (z,t)for real traffic sources. One possible solution could be to makean on-line estimate of the effective bandwidth from the formulaabove. This needs a large number of calculations of exponentialterms and seems for the time being not to be realistic. Anotherpossibility which may seem to be appropriate is to estimate theeffective bandwidth by considering the worst case traffic obtain-ed from the traffic contract. Assuming negligible CDV theseparameters are PCR for DBR type traffic, and PCR, SCR andIBT for SBR type traffic. The main drawback with this app-roach is that there might be a large difference between the con-tract and the actual use of network resources for a source.

To reflect the actual usage Kelly has proposed to apply a simpleon/off model to estimate the effective bandwidth. We let h bethe peak rate and m the mean rate of an on/off type source. Thenthe effective bandwidth will read:

α(m,h) = 1/s log[1 + m /h(esh – 1)]

where s = zt.


≈

≤

The nice thing with this simple formula is that it is easy to esti-mate the effective bandwidth by counting cells, and therebyestimating the mean m. (We assume that the peak rate h isknown.) This gives rise to the famous Kelly’s charging formula[1].

Kelly’s charging algorithm

Consider a SIR source and suppose that the mean value isstochastic and given by a variable M with expectation m, thenKelly has shown that the effective bandwidth formula aboveremains unchanged (where m is now interpreted as the mean ofthe stochastic variable M). Suppose that the network at call set-up requires the user to announce a mean value m and thencharges the connection of an amount f (m,M) where M is themeasured mean rate for the connection. In [1] it is shown thatthe best choice of the tariff is to take the tangent of the curveα(M,h) at the point M = m. By this assumption one gets the fol-lowing tariff:

f (m,M) = a(m) + b(m)M

where

b(m,h) = esh – 1 / s[h + m(esh – 1)]

and a(m,h) = α(m,h) – mb(m,h).

List of abbreviations

ABR Available Bit Rate

ACR Allowed Cell Rate

ACTS Advanced Communications Technologies andServices

ATC ATM Transfer Capabilities

ATM Asynchronous Transfer Mode

BSS Business Support System

CAC Connection Admission Control

CBR Constant Bit Rate

CDR Call Detail Records

CDV Cell Delay Variation

CDVT Cell Delay Variation Tolerance

CSCW Computer Supported Co-operative Work

DBR Deterministic Bit Rate

DG Data Generation

EPCU Ericsson Policing and Charging Unit

IBT Intrinsic Burst Tolerance

LAN Local Area Network

MCR Minimum Cell Rate

OSS Operations Support System

PCR Peak Cell Rate

PVC Permanent Virtual Channel

QoS Quality of Service

SBR Statistical Bit Rate

SCR Sustainable Cell Rate

SVC Switched Virtual Channel

UBC Usage Based Charging

UBR Unspecified Bit Rate

UPC Usage Parameter Control

VBR Variable Bit Rate

VCI Virtual Channel Identifier

VPI Virtual Path Identifier


118

The frequency assignment algorithm used in the mobilenetwork planning tool MOBINETTB Y R A L P H L O R E N T Z E N


1 Introduction

Telenor Research and Development is developing the EDP-based mobile network planning tool MOBINETT. Based oninput data concerning traffic, field strength, interference,candidate base stations, available frequencies and cost data,MOBINETT proposes which base stations should be establishedand which frequencies should be assigned to which basestations. The objective is to minimize total cost. MOBINETTcontains a frequency allocation module based on integer pro-gramming. In this paper the integer programming model and thecorresponding solution algorithm are described.

The input to the frequency allocation module is a set of admiss-ible frequencies (which need not be contiguous) and a given setof base stations where each base station must have assigned to ita required number of frequencies. For each base station certainfrequencies may be preset and certain frequencies may beexcluded. In addition, a compatibility matrix is given, which foreach pair of base stations indicates to what extent they interfere.Based on this input the frequency allocation module in MOBIN-ETT tries to allocate frequencies to base stations such that thefrequency requirements are satisfied and the total interference isminimized. The frequencies are partitioned into groups. To thegreatest possible extent frequency groups rather than individualfrequencies should be allocated to the base stations.

The admissible frequencies have the form K0 + K1 × f where K0and K1 are constants and f is an integer. We shall refer to theindividual frequencies by their integer f.

For certain base stations certain frequencies or frequency groupsmay be preset. Furthermore, for certain base stations not all fre-quencies or frequency groups may be admissible.

2 Notation and basic concepts

The following notation is used:

F Number of admissible frequencies

f Subscript denoting frequency

g Subscript denoting frequency group

Ag Number of frequencies in frequency group g

g(f) The frequency group which contains frequency f

B Number of base stations

b Subscript denoting base station

Fb Number of frequencies required on base station b

C Compatibility matrix with dimensions B × B

The matrix C is normally obtained from the output of a fieldstrength prediction program. The elements cbb’ of the matrix Care either in the interval [0, 1] or equal to 2 or 3.

cbb’ < 0.1 indicates that the base stations b and b’ do not inter-fere in practice.

0.1 ≤ cbb’ < 1 indicates that the base stations b and b’ do inter-fere in the sense that ideally b and b’ should not use the samefrequency. If the same frequency is used on b and b’ a penalty isincurred which increases with increasing cbb’.

cbb’ = 1 indicates that if frequency f is assigned to b and frequ-ency f’ is assigned to b’, then we must have |f – f ’ | ≥ 1.

cbb’ = 2 indicates that if frequency f is assigned to b and frequ-ency f’ is assigned to b’, then we must have |f – f ’ | ≥ 2.

cbb’ = 3 occurs if and only if b = b’. It indicates that if both fre-quencies f and f’ are allocated to b, then we must have |f – f ’ | ≥ 3.

When cbb’ = 1 we say that b and b’ e-interfere, and whencbb’ = 2 we say that b and b’ n-interfere.

Based on the matrix C we establish a series of undirectedgraphs, namely GN, GE and G1, G2, ... , G9. The nodes in allthese graphs represent the base stations. In GN there is an edgebetween b and b’ if and only if b and b’ n-interfere. In GE thereis an edge between b and b’ if and only if b and b’ e-interfere. InGi there is an edge between b and b’ if and only if i / 10 ≤ cbb’ < (i + 1) / 10.

In order to obtain a tight linear programming relaxation of theinteger programming model we operate with odd holes, EN-pairs and cliques.

An odd hole with cardinality ≥ 5 in GN is denoted by HN.An odd hole with cardinality ≥ 5 in GE which is not in GN isdenoted by HE.

An EN-pair (E,N) consists of two sets of base stations E and Nwith |E| > 1 where all b in N n-interfere with all b in E ∪ N, andwhere the b-s in E e-interfere with each other.

For an EN-pair (E,N) we define s(E,N) to be the subset of Nwhich results when we for all EN-pairs (E’,N’) with E’ ∪ N’ ⊃ E ∪ N remove from N every b in N’.

A clique in GN is denoted by CN.

A clique in GE which is not a clique in GN, and which cannotbe written as E ∪ N where (E,N) is an EN-pair, is denoted byCE.

A clique in Gi is denoted by Ci.

3 The integer program

The frequency assignment problem is formulated as an integerprogram.

First we define the variables:

1 if frequency f is used on base station bxbf =

0 otherwise

1 if frequency group g is used on base station bxbg =

0 otherwise

sb frequency deficit on base station b

1 if the clique constraint associated with frequency f| and clique Ci is violated

z f , Ci = 0 otherwise


For realistic cases the number of variables is reasonably small(~10 000), but the number of constraints can be quite large(> 150 000). We use a heuristic where we repeatedly need tosolve the linear programming (LP) relaxation of the integer pro-gram. Most LP packages perform better solving the dual of suchproblems, i.e. linear programs with a reasonable number ofconstraints and a large number of variables. In the dualized pro-blem it will be the dual variables which have to be integer. Wetherefore formulate the dual of the integer program and describea heuristic to get a solution where the dual variables haveinteger values.

4 The dual of the integer program

The dual of the integer program can be formulated as follows:

minimize

(13)

subject to

(14)

≥ –cbf for f admissible on b

+ π f ,KLCE∋b∑ + π f ,KN

CN ∋b∑ + π f −1,KN

CN ∋b∑

+ π1f ,EN

E , N( ); N >1,b∈L{ }∑ + π1

f ,ENE , N( ); N >1,b∈N{ }

∑ + π1f −1,EN

E , N( ); N >1,b∈N{ }∑

+ π 2f ,EN

E , N( ); N >1,b∈L{ }∑ + π 2

f ,ENE , N( ); N >1,b∈N{ }

∑ + π 2f +1,EN

E , N( ); N >1,b∈N{ }∑

+ π 2f ,EN ,b'

E , N( ),b' ;b∈E∪ N ,b' ∈s E , N( ){ }∑

+ π f ,HEHE∋b∑ + π f ,HN

HN∋b∑ + π f −1,HN

HN∋b∑

πb + πbf + πb, f −1 + πb, f −2

+ π1f ,EN

f ,( E , N )∑ + π 2

f ,ENf ,( E , N )

∑ + π1f ,EN ,b'

f ,( E , N ),b' ∈s( E , N )∑

+ HE −1( ) / 2[ ]f ,HE∑ π f ,HE + HN −1( ) / 2[ ]

f ,HN∑ π f ,HN

Fbπbb∑ + πbf

b, f∑ + π f ,CE

f ,CE∑ + π f ,CN

f ,CN∑

Then we formulate the constraints:

(1)

(2) xbf + xb,f+1 + xb,f+2 + xb,g(f) + xb,g(f+1) + xb,g(f+2) ≤ 1

(3)

(4)

(5)

for EN-pair (E,N) with N > 1

(6)

for EN-pair (E,N) with N > 1

(7)

for all EN-pairs (E,N) and for all b’ in s(E,N)

(8)

(9)

(10)

(11) xbf ≥ 0 and integer, xbg ≥ 0 and integer, sb ≥ 0, zk,KL ≥ 0.

xbf and xbg are only defined if f and g are admissible on b.

Finally, we state the objective function to be minimized:

(12)

Here, the coefficients cbf and cbg are normally 0 initially. Aswill be seen, they may change during the solution process. Thepenalties cCi are chosen to increase with increasing i, and thepenalties cb are chosen to be large numbers.

min cbf xbfb, f∑ + cbg xbg

b,g∑ + cbsb

b∑ + cCiz f ,Ci

f ,Ci∑

xbfb∈Ci∑ + xb,g( f )

b∈Ci∑ − z f ,Ci ≤ 1

xbf + xb,g( f ) + xb, f +1 + xb,g( f +1)( )b∈HN∑ ≤ HN −1( ) / 2

xbf + xb,g( f )( )b∈HE∑ ≤ HE −1( ) / 2

xb' ,g( f −1) + xb' , f +1 + xb' ,g( f +1) ≤ 1

xbf + xb,g( f )( )b∈E∪ N

∑ + xb' , f −1 +

xbf + xb,g( f ) + xb, f −1 + xb,g( f −1)( )b∈N∑ ≤ 1

xbf + xb,g( f )( )b∈E∑ +

xbf + xb,g( f ) + xb, f +1 + xb,g( f +1)( )b∈N∑ ≤ 1

xbf + xb,g( f )( )b∈E∑ +

xbf + xb, f +1 + xb,g( f ) + xb,g( f +1)( )b∈CN∑ ≤ 1

xbfb∈CE∑ + xb,g( f )

b∈CE∑ ≤ 1

xbff

∑ + Ag xbgg∑ + sb = Fb


5.2 Fixing one variable at a time

If we decide to fix one variable at a time, we proceed asfollows: If there are fractional dual variables corresponding totype (15) constraints, we choose one with the highest fractionalvalue and fix it temporarily to 1. If all these dual variables arealready integer, we choose a fractional variable correspondingto a type (14) constraint with the highest fractional value and fixit temporarily to 1. If this fixing causes too many frequencies tobe allocated to a base station, the variable is permanently fixedto 0. If not, and the fixing does not lead to an increased Σb sb,the fixing is made permanent. Otherwise, we try a temporaryfixing to 0. Then we fix the variable permanently to 1 or 0 suchthat we get the least possible increase of Σb sb.

5.3 Fixing more than one variable at a time

If we allow for fixing more than one variable after each solutionof the relaxed LP, we choose a threshold > 0.5 and proceed asfollows:

First we sequence the fractional variables corresponding to type(15) constraints in descending order. Then, one by one, we fixthem permanently to 1 if

• Their value is greater or equal to the threshold,

• The fixing, together with earlier fixings, does not allocatemore frequencies than required to a base station.

Then we treat the fractional variables corresponding to type (14)constraints in the same way.

If no fractional variable exceeds the threshold, we fix one vari-able at a time as described in 5.2.

5.4 Fixing variables permanently

Fixing variables permanently is done, as explained below, byremoving constraints from the problem.

If we want to fix permanently a fractional dual variable to 1 wefirst try to do this by deleting constraints from the problem.

We delete constraints of type (14) or (15) representingcombinations of base stations and frequencies or frequencygroups which are incompatible with the fixed combination.Furthermore, if fixing a variable to 1 implies that the frequencydemand for the corresponding base station is satisfied usingfixed variables only, all constraints of type (14) and (15) per-taining to this base station whose dual variable is not fixed areremoved from the problem.

If we cannot find any constraints to delete, we change the righthand side element corresponding to the dual variable to ∞. (Thiscorresponds to setting the cost coefficient of the correspondingprimal variable in the primal problem to –∞.)

If we want to fix permanently a fractional dual variable to 0 wesimply remove the corresponding constraint from the problem.

(15)

≥ –cbg for g admissible on b

(16) πb ≥ –cb

(17) πf,CE ≤ cCE

(18) All π-s ≥ 0 except πb which is unconstrained.

(19) The dual variables xbf and xbg corresponding to constraints(1) and (2) are required to be integer.

5 Solution method

5.1 General

The solution method is heuristic and solves the dual problem.The dual variables associated with the constraints (14) and (15)in the dual program correspond to the primal variables xbf andxbg respectively in the primal program. We repeatedly solvelinear programming relaxations of the dual problem usingdynamic variable generation and deletion where wesuccessively fix one or more of relevant fractional dual vari-ables to 1 or 0 until they all are integer.

First we generate data structures external to the linear programwhich hold all cliques in GE, GN and Gi, all EN-pairs and allodd holes up to a specified size. These data structures are usedto dynamically identify variables which price out negativelyrelative to the current basic solution and thus can be added tothe linear program. Variables corresponding to cliques, EN-pairs and odd holes which become non-basic during the solutionprocess are deleted from the linear program.

Of course, the problems of generating all cliques, EN-pairs andodd holes are NP complete, but our graphs are sufficiently smallto allow us to do this generation in reasonable time provided werestrict ourselves to odd holes of small size.

In the variable fixing phase we can choose between fixing onlyone or possibly more than one variable to an integer value aftereach solution of the LP relaxation.

+ π f ,CECE∋b, f ∈g

∑ + π f ,CNCN ∋b, f ∈g

∑ + π f −1,CNCN ∋b, f ∈g

∑

+ π1f ,EN

E , N( ); N >1,b∈E{ }, f ∈g∑ + π1

f ,ENE , N( ); N >1,b∈N{ }, f ∈g

∑

+ π1f −1,EN

E , N( ); N >1,b∈N{ }, f ∈g∑ + π 2

f ,ENE , N( ); N >1,b∈E{ }, f ∈g

∑

+ π 2f ,EN

E , N( ); N >1,b∈N{ }, f ∈g∑ + π 2

f +1,ENE , N( ); N >1,b∈N{ }, f ∈g

∑

+ π 2f ,EN ,b'

E , N( ),b' ;b∈E∪ N ,b' ∈s E , N( ){ }, f ∈g∑

+ π f ,HLHE∋b, f ∈g

∑ + π f ,HNHN∋b, f ∈g

∑ + π f −1,HNHN∋b, f ∈g

∑

Agπb + πbff ∈g∑ + πb, f −1

f ∈g∑ + πb, f −2

f ∈g∑


5.5 Fixing variables temporarily

If we want to fix variables temporarily, we do not want to deleteconstraints which we later would have to recreate. Instead weimplement temporary variable fixing by manipulating righthand sides.

If we want to fix temporarily a fractional dual variable to 1 wedo this by changing the corresponding right hand side to ∞, andif we want to fix temporarily a fractional dual variable to 0 wedo this by changing the corresponding right hand side to –∞.

6 Use of GRASP

For the case where we fix one variable at a time only, we haveincorporated an option to use GRASP (Greedy RandomizedAdaptive Search Procedure) with parameter γ. In this case wesave the solution of the LP relaxation we obtain before variablefixing starts. Starting from this solution we carry out γ solutionprocesses with variable fixing as described above. In each ofthese processes we choose the variable we want to fix tempor-arily to 1 randomly amongst the n largest fractional dual vari-ables, where n is a parameter chosen by the user. This gives usin principle γ distinct solutions to our problem, and we choosethe best one.

7 Partitioning base stations into subsets

Even if the described heuristics are used to solve the frequencyallocation problem, computer time may be prohibitive for largecases. We have therefore implemented an option for partitioningthe base stations into subsets. The optimization is carried outand frequency allocations are fixed for the subsets one by one.We have to pay for this of course by a possible reduction insolution quality.

When we have solved the problem for some subsets, then theoptions for frequency allocations in the remaining subsets arereduced because of interference between base stations in subsetsalready treated with base stations in subsets not yet treated.

Ideally, we want to partition the base stations into subsets suchthat

• There is as little interference between subsets as possible

• The size of the subsets is such that the workload for theoptimization is equally shared between the subsets.

The problem of minimizing interference between subsets ofgiven sizes is a variant of the graph partitioning problem. Wehave implemented a simple heuristic for this problem where westart from a partition based on the sequential ordering of thebase stations by the user and then exchange base stations be-tween subsets until we obtain a local minimum.

Our method for choosing the subset sizes so that the optim-ization workload is evenly distributed is based on some verybold assumptions and approximations. One of the assumptionsis that all frequencies are admissible for all base stations.

We introduce some additional notation:

n number of n-interferences

e number of e-interferences

r average frequency requirement per base station

Bi number of base stations in subset i

s number of subsets

For a randomly selected base station the expected number of n-interferences is 2n/B, and the expected number of e-interfer-ences is 2e/B. If the set of base stations is partitioned into subset1 and subset 2 with B1 and B – B1 base stations respectively, theexpected number of n-interferences between a particular basestation in subset 2 with base stations in subset 1 is

(20) N = (2n / B)B1 / (B – 1) = 2nB1 / (B(B – 1)),

and the corresponding expected number of e-interferences is

(21) N = (2e / B)B1 / (B – 1) = 2eB1 / (B(B – 1)).

Assume now that all frequencies in subset 1 have been allocat-ed. The question is what is the expected number of different fre-quencies which are allocated to the base stations which interferewith a particular base station in subset 2. We make theassumption that these frequencies are randomly distributed.Then the expected number of different frequencies which areallocated to base stations in subset 1 which n-interfere with aparticular base station in subset 2 is

(22) F(1 – (1 – 1 / F)Nr) ≈ Nr = 2nrB1 / (B(B – 1)).

(The approximation is reasonable only if F is large compared toNr, but we use it anyway.)

Correspondingly, the expected number of different frequencieswhich are allocated to base stations in subset 1 which e-interferewith a particular base station in subset 2 is

(23) F(1 – (1 – 1 / F)Er) ≈ Er = 2erB1 / (B(B – 1)).

Thus, the number of possible allocations in subset 2 is reducedon the average by

(24) [2(3n + e)r / (B(B – 1))]B1(B – B1).

Then we have the following:

Number of constraints in LP no 1:

(25) B1F

Expected number of constraints in LP no 2:

(26) B2F – [2(3n + e)r / (B(B – 1))]B1B2


(27) B3F – [2(3n + e)r / (B(B – 1))](B1 + B2)B3


(28) B4F – [2(3n + e)r / (B(B – 1))](B1 + B2 + B3)B4

and so on.


These expressions are then set equal to each other.

We set

(29) k = 2(3n + e)r / (FB(B – 1)).

Then

(30) B2 ≈ B1 / (1 – kB1)

(31) B3 ≈ B1 / (1 – k(B1 + B2))

(32) B4 ≈ B1 / (1 – k(B1 + B2 + B3))

...

(33) Bs ≈ B1 / (1 – k(B1 + ... + Bs-1)).

Furthermore,

(34) B1 + ... + Bs = B.

A set of Bi-s which satisfy (30) to (34) may be found by trialand error. However, in order to obtain a set of Bi-s which issomewhat more robust to deviations from our assumptions weuse some rough approximations in order to obtain a frameworkwhere the subset sizes grow linearly.

From the above we see that

(35)

(assuming that k is small).

This gives

(36) Bi+1 – Bi ≈ kB2i.

We approximate the difference equation (36) by the differentialequation

(37) dB / di = kB2

with solution

(38)

Condition (34) gives

(39)

Finally, the Bi-s found by (38) are rounded in a way whichmakes (34) hold true.

C =1

k2

s +1( )2+

2B

ks s +1( )−

1

k s +1( )

Bi =1

1 / C − ki≈ C + kC

2i.

Bi+1 / Bi =1− k(B1 +L + Bi−1 )

1− k(B1 +L + Bi )≈ 1+ kBi

8 Concluding remarks andacknowledgement

At the time of writing the frequency allocation module is stillbeing experimented with and has most probably not found itsfinal form. It is expected that the module later will be incorpo-rated in a graphical user interface.

The author wishes to thank H. Moseby for valuable assistancein setting up and implementing the frequency allocation modulein MOBINETT.

Status

123

International research and standardization activities in telecommunication

Editor: Endre Skolt


In the status section of Telektronikk we have earlier presentedstandards work in ETSI and ITU, and research studies inEURESCOM, RACE and ACTS. In this issue of Telektronikkwe take a look at the European Co-operation in the Field ofScientific and Technical Research (COST) programme and pre-sent three projects which have had Telenor participation.

COST involves projects in a large number of areas from agricul-ture and biotechnology to telecommunications. All projectsunder COST are pre-competitive and the COST telecommunica-tions programme has since its introduction in 1971 operated as aEuropean forum for development of new technology, co-ordina-tion of basic research, and contributions to standardisationorganisations. Today, there are 25 countries participating inCOST projects, mostly EU member countries. COST is how-ever not a EU body. In the telecommunications programme thisyear, more than 2000 scientists have joined, and compared withother international research programmes, university scientistsrepresent a large part of the working force.

The most important study areas in COST include:

• Development and simulation of Networks, including Integrat-ed Satellite/Terrestrial Networks

• Multimedia and Image Communications

• Radio systems

• Optical technologies: Systems, components and reliability

• Telecommunications facilities for disabled and elderly people

• User requirements in Telecommunications

• Speech Technology

• Software Verifications and Validation.

Telenor R&D has actively participated in COST pro-jects addressing areas like speech technol-ogy, development of video codingtechnologies, telecommunicationsfor disabled and elderly peo-ple, fundamentals of

mobile and satellite communication and fixed services terres-trial communication. In the following we present results fromthree projects that recently have terminated.

In the first paper, Mr. Per Helmersen reviews what has beenachieved in the COST 219 project. COST 219 was set up in1986 and ended September this year. The main objectives of theproject have been to collect information about existing telecom-munications devices appropriate for disabled and elderly people,to stimulate activities in this field, and to survey the practicalneeds of disabled people and evaluate the future possibilities ofinformation technologies. Mr. Helmersen reports that telecomoperators are starting to view the increasing number of elderlyand disabled people as an important market. By meeting theneeds of customers with reduced hearing, vision and mobility,telecommunication service providers may gain a competitiveedge in the overall market.

In the second paper, Mr. Agne Nordbotten presents the achieve-ments in COST 235, a project studying radiowave propagationeffects on telecommunications system using frequencies in thehigher microwave and millimetre bands. The growth in radiocommunications has made it necessary to use frequency re-sources in these bands both for satellite and terrestrial commu-nication. More specifically, COST 235 has studied the propaga-tion effects caused by the atmosphere (attenuation, scattering,depolarisation) and ground terrain effects (multipath, shielding,attenuation from trees etc.). Telenor R&D has been an activeproject partner, and chaired one of the three working groups.

The last paper gives a comprehensive description of the resultsfrom COST 231, “Evolution of land mobile radio communica-tions”. The project has been organised into three workinggroups with focus on the radio system aspects and technologiesfor third generation standards, channel characterisation at fre-

quencies between 900 MHz and 3 GHz, and both sys-tem aspects and propagation at microwave

and millimetre wave frequencies. Theauthors are Mr. Per Hj. Lehne

and Mr. Rune H. Rækken.

125

IntroductionB Y E N D R E S K O L T


Figure 1 Telecommunication areas

Radio

comm

unication Tran

smis

sion

and

switc

hing

Health

care

infor

matic s

tanda

rd

Message handling and

electronic directory

Languages fortelecommunication

applications

SecuritySignal processing

Data netw

orks

Term

inal

equ

ipm

ent

and

user

asp

ects

TMN

Teletraffic and

dimensioning

Intelligentnetworks

Broadband

Service

definitions

In Europe, there are about 100 million elderly people and 50million who are disabled. In addition, there are those who sufferfrom dyslexia, learning difficulties or milder forms of intellect-ual impairment, as well as the 0.5 % of the population who aretemporarily disabled through illness or accidents.1 The preval-ence of most forms of impairment increases considerably withage and so the disabled population grows disproportionately aslife expectancy rises. Disabled and elderly people are a signifi-cant segment of the telecommunications market.

Access to telecommunications offers disabled users improvedindependence, mobility and quality of life. For many, it meansthe opportunity to work at their own pace and in their ownhomes. The telecommunications industry has made enormousprogress over recent years. Unfortunately, more advancedequipment often makes greater demands of the user and so, peo-ple who are all too familiar with the many barriers of daily lifecan actually be further isolated and disenfranchised by technol-ogy with the potential to remove many of these barriers. Any-one who cannot use telecommunications services will find ithard to gather information, make reservations, maintain socialcontacts or even call emergency services. As an increasing num-ber of services are being transformed into telecommunicationsor on-line services, the prospect of equal access for many ofthese user groups is threatened.

There are also intrinsic problems in the nature of new technol-ogy. As products are changing rapidly, shrinking in size andcramming more and more features into software, engineers havedifficulty in getting inside the chip or software to makemodifications while the product is being developed. A conse-quence of this is that existing devices for people who requirespecial solutions and interfaces are obsolete, non-competitiveand difficult to maintain. In addition to these technological con-straints, cost and lack of awareness have also been importantfactors in the poor provision made for disabled user groups.

In order to ensure that the needs of disabled people were prop-erly formulated and made available to the European telecommu-nications industry, regulatory authorities, research bodies, stan-dardization organizations and others, the COST 219 project wasset up. The project started in September 1986 and was termi-nated in its present form in September 1996. Two requests forextensions of the activity have been approved, making 219 oneof the longest running COST projects. During the 1987–94period the project received financial support from the EuropeanCommunity, DGXIII and DGV in order to make it possible forrepresentatives from small industry, universities as well as invit-ed experts on handicap problems to participate in the work ofCOST 219.

The COST 219 signatories2 have co-operated in order to pro-mote research into the field of telecommunications and tele-

informatics with the aim of proposing solutions to problemsrelated to the needs of disabled and elderly persons. The mainobjectives of the project have been to collect information aboutexisting telecommunications and teleinformatics devices andservices as well as ongoing research and development appropri-ate to disabled and elderly people, to stimulate activities in thisfield, to survey the practical needs of disabled people and toevaluate the future possibilities of information technology.Under the chairmanship of Jan Ekberg from the National R&DCentre for Welfare and Health in Finland, COST 219 has organ-ized 30 meetings and seminars focusing on a wide range of sub-jects such as text telephony, picture communication, elderly andtechnology, smart cards and terminals, speech technology, tele-phones and hearing aids, legislation and standardization,databases, etc. Twenty-nine publications (including 3 bookspublished by the Commission of the European Communities3)document 10 years of activity in the field.

The major activities of COST 219 have been carried out by var-ious working groups. The composition and focus of thesegroups have changed during the 10 years the program has exist-ed, reflecting not only social and technological change, but alsothe increasing maturity of COST 219. The following workinggroups were functioning in the final stages of the program:

WG1 Standardization and Legislation (Chairperson: C. Stephanidis, GR)

WG2 Videotelephony (Chairperson: Leonor Moniz Pereira, P)

WG3 Access to Communication Technology (Chairperson: Bob Allen, Irl)

WG4 Voice Communication (Chairperson: Patrick Roe, CH)

WG5 Technology Trends (Chairperson: Diamantino Freitas, P)

WG6 Text Communication (Chairperson: Kelvin Currie, UK)

WG7 Information Transfer (Chairperson: John Gill, UK)

Organizations such as ETSI – especially HF2 – have frequentlyused COST 219 as the basis for their own work in this area. Fur-thermore, the information obtained has been collated anddisseminated through publications and conferences acrossEurope and more recently in Eastern/Central Europe. Becauseof COST 219’s activities it was possible to include issues relat-ed to disabled and elderly persons in RACE and to define a newprogram, TIDE. The overall result is that a forum has beenestablished which is now internationally recognized as a refer-ence point on matters concerning disability and telecommunica-tions. One example of this recognition is the recent exposuresolutions for disabled customers received at Telecom 95. As adirect result of COST 219’s efforts an ITU stand at Telecom 95

126

COST 219: Future telecommunications and teleinformaticsfacilities for disabled people and elderlyB Y P E R H E L M E R S E N


1 The forgotten millions. Access to telecommunications forpeople with disabilities. DG XIII. Brussels, 1994.

2 The MOU has been signed by Austria, Belgium, Denmark,Finland, France, Germany, Greece, Hungary, Ireland, Italy,Malta, the Netherlands, Norway, Portugal, Spain, Sweden,Switzerland and the United Kingdom. Croatia and Sloveniahave started the procedures to sign. Active participation fromEC, (TIDE office, RACE office and Helios) should also bementioned, in addition to frequent collaboration with ETSI.

3 Von Tetzchner, S (ed.). 1991. Issues in telecommunicationand disability. Brussels, Office for Official Publications of theEuropean Commission. ISBN 92-826-3128-1-.

Roe, P R W (ed.). 1995. Telecommunications for all. Brussels,Office for Official Publications of the European Commission.

Olesen, K G. 1992. Survey of text telephones and relay ser-vices in Europe. Brussels, Commission of the EuropeanCommunities, Directorate-General, Information Technolo-gies and Industries, and Telecommunications.

was constructed as a smart home for elderly and disabled show-ing how telecommunications can be used to support independ-ent living. This ITU/COST 219 stand was visited by over10,000 attendees.

COST 219 was officially terminated in September 1996. A newCOST Action has recently been approved, however, ensuringthat information related to the needs of disabled and elderly per-sons will be made available also in the future. Signatories of theCOST Action 219bis MOU (1996 – 2001) will carry out the fol-lowing main tasks:

1) Analyze and propose new solutions intended (in order of pri-ority) to:

• make services generally accessible to all,

• make services adaptable when they cannot be made gener-ally accessible, or

• offer special solutions to meet explicit problems

2)Support co-operation between technical specialists oftelecommunications, teleinformatics, standardization, legisla-tion and specialists working for elderly and disabled persons

3)Evaluate new technical solutions for providing telecommuni-cation and teleinformatics services to elderly and disabledpersons

4)Actively disseminate information to relevant actors

5)Promote appropriate research activities.

Modern telecommunications have the potential to open up soci-ety to people with disabilities. The market is large and econom-ically attractive. It is obviously necessary and desirable thattelecommunications should develop in a way which improvesthe services available to all. There is sometimes a mistakenassumption that this will come about as a direct and inevitableresult of technical progress. The efforts of the COST 219 pro-gram have since 1986 demonstrated that an emphasis on stand-ardization, legislation and European-wide coordination is essen-tial. It has also become clear that a broad uptake of solutions isachievable only if equipment manufacturers and service provid-ers can be convinced that there is a market and that many of thespecial requirements of elderly and disabled users will also ben-efit normal untrained users and subscribers.

Within a liberalized telecommunications environment Telenorand other PNOs will find themselves being drawn towards theprovision of specialized services for the needs of persons withdisabilities and the large greying population. This will bethrough realizing the commercial opportunity that these custom-ers offer as well as in response to calls for regulatory action bygroups representing these customers’ interests. A third and per-haps equally important driving force will be the increasingrecognition of the importance of design for all policies.“Accessibility and Universal Design Principles” have recentlybeen approved and implemented by major US telecom operatorssuch as Pacific Bell and NYNEX. These companies have realiz-ed that by meeting the needs of customers with reduced hearing,vision, mobility or information processing abilities they are alsodeveloping products and services which benefit all of their cust-omers – not only those with specific and identifiable handicaps.In a market where ever-raging price wars make it increasinglydifficult for customers to base their decisions solely on cost,factors such as ease of use and the availability of flexible solu-

tions which can be adapted to meet specific needs and tastesmay be the only factors differentiating one provider from thenext. Carrying on in the tradition of COST 219, COST 219biswill continue to provide PNOs not only with insight into theneeds of their many disabled customers and information on howto meet them, but also give those cognizant of the commercialadvantages of universal design a clear market advantage and aflying start into the next century.


Introduction

The COST 235, which started in 1991, concluded its workthrough the arrangement of an International Symposium onBroadband and Millimetric Propagation Effects on TerrestrialRadio in October 1996. In that connection a report presentingthe objectives of the work, the main contributions and resultswith conclusions, discussions and also recommendations forfuture work has been issued by the Commission of the Europeancommunities [1].

This paper gives a short summary of the objectives and resultsfrom the project with some focus on the contributions fromTelenor R&D.

COST 235

The growth in radio communication as well as the technologicaldevelopment now makes it both possible and necessary to startusing the frequency resources of the higher microwave regionand in the millimetric region for both satellite and terrestrialcommunication. It was the realisation of the fact that thesefuture telecommunication systems would be strongly influencedby propagation effects from the atmosphere (attenuation,scattering, depolarisation) and ground terrain effects (multipath,shielding, attenuation from trees, etc.), as illustrated in Figure 1,which necessitated a thorough investigation into these phenom-ena in order to establish reliable and accurate prediction models,QoS and availability criteria for such systems. Since the projectstarted, the use of radio communication has increased evenfaster than expected. During the same period algorithms for effi-cient coding of moving video have been developed and thestandardisation of MPEG2 has resulted in possibilities fordevelopment of new services including interactivity creating aneed for broadband access for public services.

Systems based on the use of millimetric waves will normally below margin systems requiring very reliable prediction methodsand efficient fading countermeasures to insure cost effectiveoperation and efficient resource utilisation.

The applicational background for the results from COST 235thus turned out to be the very best at the end of the project andthe timing for the project very good. In Europe, several largerprojects based on the application of millimetric waves havebeen initiated under the Fifth Framework Program of the EU.

The work in COST 235 on these problem areas was organisedin three working groups on

• Frequency selective fade effects focusing on channel mod-elling, diversity improvement, dual polarisation phenomena,multipath occurrence and performance and availability crite-ria (clear air effects) (WG1)

• Flat fade effects and millimetric wave considerations address-ing the areas of gaseous absorption, the effects of hydro-meteors and the influence of atmospheric turbulence (WG2)

• Site shielding and interference reduction strategies like theeffects of urban environment on the radiowave propagation,terrain propagation models as well as antenna and systemsaspects (WG3).

The project was led by Martin M.P. Hall, Rutherford AppletonLaboratory. Terje Tjelta, Telenor R&D, has been the chairmanof Working Group 1 (WG1). Research groups from altogether14 nations has participated in the project.

The work in the project started with a total of 28 technical ques-tions which then were reduced to eleven areas for detailed con-sideration in the respective working groups.

Main results from the project

Under WG1 the work has focused on Quality of Service andavailability of systems in the microwave region. Improved pre-diction models have been developed and several contributionsmade to the standardisation process in ITU-R. Results frommeasurements on multipath occurrence in Norway have beencorrelated with meteorological data and a close relationship be-tween the refractivity gradient obtained from meteorologicaldata and multipath fading was found. It has been recommendedthat work in this area continues in another COST project.

A thorough investigation into the possibilities of different meth-ods for diversity improvement has been undertaken, and fromNorway it has been shown that there are great advantages in theform of equipment saving and space saving on the installationsite in using angle diversity since this requires only one antennaper installation as illustrated in Figure 2, which shows the basicprinciples for angle and space diversity, respectively.

During the project period new transmission systems based onSDH have been installed by different operators. For such sys-tems the error problems are different from previous systems, itis now a question of errored blocks rather than errored bits, andthese problems have turned out to be somewhat complicated inhandling. The relationship between block error and BER hasbeen analysed, and a method has been described that makes itpossible to convert old bit error measurements into block errorswhich can be evaluated according to G.826 parameters.

A model for dual polarisation of the radio channel has beendeveloped considering the separate contributions from bothantenna and propagation effects. Testing versus experimental

128

COST 235: Radiowave propagation effects on nextgeneration fixed-services telecommunication systemsB Y A G N E N O R D B O T T E N


fig.1

Figure 1 Atmospheric and terrain influences on millimetric wave propagation

data has beenperformed, andthe conclusionis that use ofdual polarisa-tion in terres-trialmicrowavesystems now isa mature tech-nology, thusincreasing thecommunicationcapacity of thatpart of thespectrum.

In the milli-metric waverange the pres-ence ofhydrometeorshas thestrongest effecton signal atten-uation. Themodels used so

far for prediction of the effects of precipitation have been testedand evaluated at frequencies below 40 GHz, and there hasactually been doubt about their validity in the higher frequencyrange. In this frequency range the contribution from scattering isimportant. Thus, the observed attenuation will not be determin-ed by the rain intensity only, but also depend on drop size distri-bution, resulting in different attenuation levels from event toevent and even within events depending on the variations indrop size distribution. For long term statistics these phenomenawill be more or less averaged out, but when such averages arebeing used, large deviations must be expected for shorter orlonger intervals depending on the details of such drop varia-tions. Figure 3 illustrates how the drop size distribution canchange during a rainfall. A knowledge of local meteorologymay be required for reliable dimensioning of such systems, anddevelopment of adaptive fading countermeasures is required.

The effects of precipitation on millimetric wave propagationhave been studied extensively in COST 235. A data base ofattenuation measurements, coupled with rain rate data, has beencollected from a number of links operated in Europe and a test-ing versus the different models proposed and in use has beenconducted. The deviations between some of the models isnoticeable and for Europe only 2–4 should be recommended. Inthis area it turns out that the models being used tend to under-estimate the fading problem over shorter distances.

At Telenor R&D the focus has been on obtaining an understand-ing of phenomena which are typical for the Norwegian climateand measurements have been running at Kjeller since 1993.Kjeller is well suited for such measurements. There are norm-ally several short and intense rainfalls with relatively largedrops and rapid variations in drop size distribution in the sum-mer season. During the period from September until Novemberthere will be both long-lasting rainfalls with smaller drops andmore intense rainfalls in between. At the end of the period theremay be some events with sleet/wet snow.

In spring there will be a long period with snow-covered terrainwhich represents the worst situation for reflection and multipathwhen the snow has recrystallised and gets wet.

It has been shown that the occurrence of sleet and hail hasstrong effects on the monthly distribution curves for the specificattenuation. Figure 4, which represents June 1993, illustratesthis based on measurements over a 630 m long path at Kjeller.The total amount of precipitation for this month was 27 mmwith one single hail event contributing 6.7 mm. The specificattenuation curves for 40 (solid lines) and 60 GHz (dashedlines), show that the hail event dominates totally for higherattenuation values.

Fading countermeasures are very important for the use of suchsystems.

Availability as a concept is much more complicated at milli-metric frequencies than at lower frequencies. In a sectored dis-tribution system (point-to-multipoint) the attenuation at a giventime will vary quite a lot with direction due to the inhomogenityof rain. Availability, however, is defined in a specified directionand will of course statistically be very much the same in differ-ent directions, but availability will not be the same for a giveninstant of time in different directions. If for instance AGC isused as a fading countermeasure there will be a difference insystems performance depending on the criteria used. Use of themost faded direction for control gives the highest availability aswell as strong signals in the unfaded directions which increasesthe interference problem in a cellular system. The differencebetween a point to point and a point-to-multipoint system withregard to fading countermeasures is obvious.

Another problem in this connection is that it would be of inter-est having some information on the dependency of the attenua-tion statistics during a day. It should not be at a minimum whenthe number of customers is the highest. Such information islacking.


Figure 2 Antenna installations for angleand space diversity

T

R1

R2

f

f

Space diversity

T

R1

R2

f

f

Angle diversity

20

18

16

14

12

10

8

6

4

210-5 10-4 10-3 10-2 10-1 100

40 GHz

60 GHz

Total month

Without hail event

Percentage of time

Spe

cific

atte

nuat

ion

in d

B/k

m

Figure 3 Variation of drop size during rainfall (as measuredwith distrometer at Kjeller)

Working Group 3 has analysed the effects of diffraction, scatter-ing and terrain in detail, with a special focus on urban areascharacterised by buildings, constructions and trees. The modelsdeveloped contribute significantly to the prediction of diffrac-tion, scatter and attenuation imposed by buildings and vegeta-tion as well as to the understanding of terrain propagationeffects and site shielding effects.

A prediction procedure aiming at estimating the effects ofscattering on a radio connection has been described.

Both attenuation and scattering from groups of trees have beenanalysed and a model for estimating the attenuation as a func-tion of vegetation depth has been recommended. A belt of vege-tation of 4–6 m can easily result in more than 20 dB of signalattenuation. Vegetation thus represents a major problem forpropagation of radio waves and in particular new systems forsatellite communication and point-to-multipoint access systemsand mobile communication in the millimetric wave area.

It has been shown that frequency reuse at millimetric frequen-cies will be reduced by the effect of scattering from buildings inurban areas. A method for calculating the effects of shielding ondigital communication systems in the frequency range from10 – 100 GHz has been implemented. The method can be usedto evaluate the performance of communication systems in anurban area.


Figure 4 Cumulative distribution of precipitation attenuation for June 1993with and without hail event representing 25 % of the total precipitation

Concluding remarks

The work of COST 235 represents a valuable contribution toimproved radio communication systems through improved pre-diction for the microwave frequency range to valuable contribu-tions for design guidance of new systems in the millimetre waverange which will now be gradually more used for communica-tion systems requiring high capacity channels. The informationcollected is of particular interest for different new ACTS pro-jects focusing on communication and distribution using milli-metric waves. For Telenor the participation in the ACTS projectAC215 CRABS (Cellular Access to Broadband Services) withthe main objective of developing and demonstrating a broad-band interactive point-to-multipoint distribution system, was aresult of contacts established in COST 235. Three of the ninepartners in CRABS participated in COST 235.

COST 235 has also represented an opportunity for writing sci-entific articles and presenting conference papers. One universitythesis was associated with the work at Telenor.

The work from COST 235 is partly continued in a new COST253 project on Satellite Systems from Ku-band and above, andanother project focusing on terrestrial systems is under con-sideration.

A well organised COST project represents an opportunity forinternational scientific work in co-operation within larger pro-ject areas. It represents possibilities for development and testingof new ideas which later evolves into systems work and applica-tions as well as the possibility for establishing valuable contactsin other organisations. However, nothing is free and possibilitiesare created through the activity of the individual partner.

Reference

1 COST 235. Radiowave propagation effects on next-genera-tion fixed-services terrestrial telecommunication systems.Commission of the European Communities, 1996. (Finalreport EUR 16991 EN.)

Introduction

The COST 231 project started on 6 April 1989, and was origin-ally scheduled to expire in April 1993. To align the time scaleof COST 231 with the time scales of interacting RACE projects,the COST action was then extended to 5 April 1996.

Chairman of the COST action was from the start Mr. PietroPorzio Gusto from CSELT, Italy. In 1992, Mr. Eraldo Damosso,also CSELT, took over the chairmanship.

The objectives were, among others, to identify the characterist-ics of third generation mobile radio and personal communica-tions systems currently under specification, and to providedesign methods and coverage models for their implementation.

This included studies on digital transmission techniques andradiowave propagation in the UHF band. The aim was to estab-lish prediction methods for attenuation and multipath effects inmacro, micro- and indoor cells by modelling and simulation ofthe radio transmission channel.

Also, long-term studies have been carried out on radio basedcommunications systems for broadband services (i.e. servicesrequiring a capacity greater than that of the basic ISDN ser-vices). This involved studies on a large spectrum of newtelecommunications technologies at millimetre wave andinfrared bands.

Thus, COST 231 has covered a variety of land mobile aspects,mainly focusing on personal communications systems providingvoice and data communications through small, cheap, handportable radio terminals.

This article highlights some of the work performed and present-ed within COST 231. The article is however not exhaustive, dueto the huge amount of work carried out within the COST action.

The main information source for this article is the COST 231Final Report [1].

Participation

At the end of the project, 20 signatory countries participated atthe Management Committee meetings: Austria, Belgium, CzechRepublic, Denmark, Finland, France, Germany, Greece, Ireland,Italy, The Netherlands, Norway, Poland, Portugal, Slovenia,Spain, Sweden, Switzerland, Turkey, United Kingdom. Usually,more than 80 people from about 70 participating organisationsattended the meetings.

Liaisons

During its period of existence the COST action had close rela-tions with most RACE projects related to mobile and personalcommunications, with other COST actions and several inter-national standardisation bodies (see Table 1). Additionally,there were liaisons with: UK Technical Party on Mobile Com-munications, UK LINK CDMA Project, and URSI (The Inter-national Union of Radio Science).

Organisation of the work

The work was split into three main areas and organised in corre-sponding working groups:

• WG1 – Radio System Aspects• WG2 – UHF Propagation• WG3 – Broadband Applications.

WG1 had its focus on the system aspects and technologies forthird generation standards. The objectives were to complementand evaluate the various RACE projects and to provide signifi-cant results to the standardisation bodies, especially on multipleaccess techniques (CDMA1, TDMA2, etc.) and new coding andmodulation techniques.

131

COST 231: Evolution of land mobile radio (including personal) communicationsB Y P E R H J A L M A R L E H N E A N D R U N E H A R A L D R Æ K K E N


RACE projects

CODIT R 2020 UMTS COde DIvision Testbed

MONET R 2066 MObile NETworks

MBS R 2067 Mobile Broadband System

MAVT R 2072 Mobile Audio Visual Terminal

ATDMA R 2084 Advanced TDMA Mobile Access

TSUNAMI R 2108 Technology in Smart Antennas for UNiversalAdvanced Mobile Infrastructure

COST Actions

227 Integrated Space/Terrestrial Mobile Networks

235 Radiowave propagation effects on next-generation fixed-services terrestrial telecommunications systems

244 Biomedical Effects of Electromagnetic Fields

245 Active Phased Arrays and Array Fed Antennas

ETSI TCs/STCs

SMG2 Radio aspects of GSM / DCS1800 / UMTS

SMG5 Universal Mobile Telecommunications System – UMTS

RES02 Land Mobile (except GSM, UMTS, DECT)

RES03 Cordless Communications, i.e. DECT, CT1/2

RES10 Radio LAN

ITU-R

TG 8/1 Recommendations on International Mobile Telecommuni-cations – IMT-2000 (formerly FPLMTS)

SG 3 Study group on Radiowave propagation

Table 1 COST 231 liaisons

1 Code Division Multiple Access.

2 Time Division Multiple Access.

WG2 studies were on the mobile channel characteristics at fre-quencies between 900 MHz and 3 GHz. Outdoor measurementsand modelling (i.e. macro- and micro-cells) were given specialattention. This included developing prediction methods wherethe influence of systems parameters like base station antennaheight and directivity and site location and environment, weretaken into account. The emphasis was gradually shifted towardsoutdoor short-range and indoor propagation and building pene-tration.

WG3 operated along a different axis, covering both systemaspects and propagation at microwave and mm-wave frequen-cies. The main focus has been on high data rates (above10 Mbit/s) short range communication in indoor environments.Both theoretical and experimental work has been carried out.Attention was also paid to system operation in outdoor environ-ments, and to infrared communications. Particular reference hasbeen given to ETSI’s High Performance Radio Local Area Net-works (HIPERLAN). Some attention has also been given tosafety aspects for infrared and millimetre wave radiation.

Main achievements

This chapter gives an overview of the main work and results ofthe research activities carried out within COST 231.

Radio channel characterisation

Knowledge about the radio channel is crucial in mobile commu-nications. A lot of effort has been put into understanding anddescribing the radio channel from different points of view.Attention has been increasingly focused on modelling the elec-trical field with respect to both the propagation delay and theincidence direction. Work has been put into defining the fielddirection-delays-spread function (FDDSF). The FDDSF charac-terises the disperse nature of the environment. Hence, bothspace- and time-dependence of the channel response can bedeveloped.

The “radio environment” is not a well defined parameter. Basic-ally, different propagation mechanisms will dominate in differ-ent environments. Therefore, categories of environments havebeen identified, each of them having characteristic propagationscenarios. Division and characterisation is given, both by celltype (i.e. macro-, micro- and pico cells) and area type (urban,suburban, rural, etc. for macro-cells). Still, environment vari-ability is present within a given category, influencing on theradiowave propagation properties. To deal with this, threesources of randomness have been identified:

• The arrangement, the mean height and the electrical proper-ties of buildings in an urban area

• The geometrical and electrical properties of the objects inter-acting with the electromagnetic field

• The receiver position and the factors leading to time varia-tions.

A common feature for all these is that they can only be describ-ed to a certain level of accuracy. Hence, they contribute to mak-ing the FDDSF and channel response (CR) into stochastic vari-ables.

The FDDSF is used to describe both the small scale / short termand large scale fluctuations of the channel.

A main goal for the work on channel characterisation and prop-agation measurements is to be able to model the channel forsimulation purposes. Therefore, a lot of effort has been put intothe channel measurement and modelling activity during thework of COST 231. Three different objectives have been aimedinto:

• Long term area simulations, to evaluate signalling perform-ance, coverage studies, cellular efficiency, and others.

• Narrowband system simulations, to consider transceiver per-formance. Examples in COST 231 are the evaluation of newmodulation techniques and optimised transceiver architec-tures.

• Wideband system simulations, which is the most interestingcase. This implies modelling the time disperse properties ofthe radio channel. The tapped delay channel has been fre-quently employed for this, with a large variety on complexitydepending on the use. A two-path Rayleigh (or Rice) modelhas been employed in evaluating different modulationschemes and for DECT system evaluation. Also a 6 tapsdelay line channel emulator has been used.

Channel sounding techniques

There are two classes of equipment for measuring propagationconditions of the mobile radio channel. Narrowband equipmenttransmits and receives CW (or a narrow frequency spectrum).This type of equipment is mainly used for path loss measure-ments. Wideband equipment – channel sounders – transmits afrequency spectrum above the coherence bandwidth of the radiochannel, and is used for channel characterisation.

Channel sounders are often categorised into three main classes:pulse, pseudo-random sequence, and frequency sweep (chirp)sounders. The two latter use pulse compression techniques.

In a pulse channel sounder, a short RF-pulse is transmitted, andthe received signal envelope is detected by the receiver. Onlyinformation about the received signal amplitude is obtained, it isthus not possible to get any information about the Doppler spec-tra. Unless very large transmitter power is used, the distancebetween transmitter and receiver is limited. In addition, thismethod is sensitive to interference from other services.

The signal used for sounding the channel by a pseudo-randomsequence sounder, is the carrier modulated with a pseudo-rand-om binary sequence. In the receiver a sliding correlator, a signalprocessor performing correlation or a channel matched filter, isused to estimate the channel impulse response. By using thistechnique it is possible to obtain a larger range between trans-mitter and receiver. If Doppler spectra are to be derived, fre-quency synchronisation between transmitter and receiver isneeded.

The frequency sweep technique is well known from radar theo-ry. Traditionally, a pulse compression filter in the receiver ismatched to the transmitted wave form. The resolution of themeasurement system is inversely proportional to the bandwidthof the frequency sweep, and the maximum measurable delay isproportional to the duration of the sweep. This measurement


technique also makes it possible to obtain large ranges betweentransmitter and receiver even when moderate output power isused. In addition, the method is resistant to interference fromother services, and gives good utilisation of the used measure-ment bandwidth.

Propagation data analysis

The main outcome of the wideband propagation measurementsis information about performance of given digital radio systemsin a given multipath environment. Some radio systems havevery simple air interfaces, and their quality of service are hencevery vulnerable to interference and multipath propagation.Other radio systems have built in measures to compensate forthe influence of multipath propagation, or even utilise theenergy received from the direct (if any) as well as all the reflect-ed signal components (multipath diversity).

When designing a radio system, or when choosing amongdifferent radio systems to be used in a given environment, it istherefore very important to have some measure of the influenceof the propagation conditions on the radio system performance.

Impulse responses measured by a channel sounder can be fur-ther processed to give parameters indicating performance of dif-ferent radio systems exposed to various propagation conditions.Important statistical parameters describing the channel impulseresponses are:

• Mean Delay (MD): The first order moment of the IR; thepower-weighted average of excess delays.

• Delay Spread (DS): The second order moment of the IR; thepower-weighted standard deviation of the excess delays.

• Fixed Delay Window (FDW): The length of the middle por-tion of the IR containing a certain percentage of the totalenergy of the IR.

• Sliding Delay Window (SDW): The length of the shortest por-tion of the IR containing a certain percentage of the totalenergy of the IR.

• Delay Interval (DI): The interval between the first time thepower of the IR exceeds a given threshold and the last time itfalls below the threshold.

• Q16 ratio: The ratio of the power inside to the power outsidea window of duration 16 µs. For each IR the window is slid tofind the position with highest power inside the window.

These parameters can be used to evaluate how well a digitalradio system will work in a given multipath environment, andhow damaging multipath propagation is for the communication.For a given modulation method it is possible to calculate the biterror rate as a function of the delay spread parameter, assumingthat a channel equaliser is not present in the receiver.

If, however, the receiver is equipped with a channel equaliser,the relevant parameter to predict the radio system performancehas to take into consideration the ratio of the power inside to thepower outside the equaliser window. The signal componentsreceived within the equaliser window can be utilised and is thusa useful signal, whereas the signal components falling outsidethe equaliser window will contribute to the overall co-channelinterference. Knowing the length of the equaliser window and

the necessary carrier to co-channel interference rate for a givenradio system to operate properly, it is thus possible to use the Q-parameter to predict performance of a radio system employing achannel equaliser.

Examples of how to predict radio system performance using thementioned parameters is given in the chapter “GSM and DECTsystems”.

Propagation work reported from related projects

A main goal for COST 231 has been the contribution to thedevelopment of UMTS. This required a more detailed descrip-tion of the wideband channel than the one suggested by COST207 for GSM. Thus, there have been close connections with theRACE projects ATDMA and CODIT for this purpose. Threemodels have therefore been presented and studied in COST 231:

• The COST 207 approach. These are 6 and 12 taps modelswith the following basic elements: Tap delay values followdifferent profiles, tap mean power and Rayleigh (or Rice, forthe first ray) distribution, four different Doppler spectrumtypes. These models have been subject to corrections and alsonew sets of parameters have been produced covering othertypes of environments than original.

• ATDMA simulator. Two complementary approaches werefollowed: Stored Measured Channels and Synthetic Channels.A large number of measurements have been performed todefine a typical channel and some atypical channels.

• CODIT simulator. These activities have been mostly aimed atthe development of a synthetic model for WSSUS3, but thenextended to non-WSSUS scenarios by using modulating func-tions in the filter coefficients.

Both projects ATDMA and CODIT have developed hardwarechannel simulators for use in their real time testbeds.

Antennas and antenna diversity

The antennas make up the connection between the communica-tion system and the radio environment. The primary task of theantenna is the connection between transmitter and receiver, butequally important is the possibility to suppress or compensatefor transmission media impairments. Three major areas havebeen considered in the latter case:

• Antenna gain and efficiency to suppress noise

• Multi port antenna diversity to suppress temporal fading andinterference

• Adaptive antenna directivity for spatially selective inter-ference suppression.

Antenna designs

Several antenna designs have been presented. Extreme wide-band antennas have been designed giving a bandwidth of up to15 GHz. For wireless LANs (WLANs), patch antennas printedon top of PCMCIA cards have been designed and tested in the


3 Wide-Sense-Stationary Uncorrelated-Scattering (channel).

5 GHz band (HIPERLAN). Conical horn antennas give nearuniform cell coverage. For mm-wave applications, dielectricantennas are suitable because they are easy to produce. Theseare dielectric leaky wave antennas which can be designed toradiate almost to broadside, i.e. orthogonal to the antenna axis.They have a very strongly frequency-dependent beam-pointingangle, a feature that can be used in a wireless LAN based onFDMA/SCPC4 and using dynamic channel allocation (DCA).

Antennas for portables have so far been mostly external wireantennas. Internal antennas for hand sets are of increasingimportance and interest, partly because of their advantages indesign. The radiation coupled dual-L antenna is a new antennadesigned e.g. for cordless telephones. Work has also been per-formed on antenna configurations for reducing body loss andabsorption of radio frequent energy in the user’s body.

Adaptive antennas

Through the close co-operation with the RACE TSUNAMI pro-ject, a lot of work has been performed on antennas for base sta-tions (BS). Keywords here are:

• Angular relations between BS antennas and radio environments• Environmental influence on perceived antenna pattern• Adaptive base station antenna arrays.

The latter case is particularly interesting. Adaptive, or “smart”,antennas denotes a whole concept of technologies, ranging from“simple” steerable directive antennas to suppress interferenceand increase range, to spatial separation of users (full SpaceDivision Multiple Access – SDMA) to increase the systemcapacity.

Antenna diversity

Another important antenna aspect is diversity. This is used atthe base station in all current mobile systems, e.g. NMT andGSM. In COST 231 a whole range of diversity techniques havebeen studied. Diversity separation can be utilised in severalradio channel environments. Possible diversity schemes are:space diversity, antenna pattern diversity, antenna orientationangle diversity, polarisation diversity and time/frequency diver-sity.

Different combining techniques falling into two major groupshave been studied: selection and summation. The characteristicfeature of selection diversity is that only one branch is active ata time. The schemes in the second group apply weights to allthe branches and the output is the weighted sum. The differentschemes have different complexity and gain, with switchingdiversity offering the lowest complexity but often only marginalimprovement. On the other end of the scale, summation usingequal-gain combining (EGC) or maximum ratio combining(MRC) gives a theoretical improvement in signal to noise ratioof 3 dB.

Diversity can also be divided between macro and micro diver-sity. Macro diversity means combining the signals received at

two or more base stations. This will reduce the influence oflong-term fading caused by shadowing objects. In this case thespatial separation corresponds to the cell size.

In theory it is possible to reduce the long-term fading margin by10 dB with selection combining of two uncorrelated branches.As an example, the fading margin for indoor micro-cells isreduced by 7.5 dB with a spatial separation (cell size) of 9–17 λ.The diversity gain is found to be 6 dB in the same case.

Micro diversity is employed by using two (or more) antennascollocated at the same mobile terminal or base station. Thistechnique reduces the influence of the short-term fading. Microdiversity shows a significant gain when two antennas are used,but it is difficult to obtain further improvement using three ormore antennas. The estimated diversity gain based on measure-ments done on 1700 MHz in different propagation environmentswith an antenna separation of 2.5 λ is 6.5 dB for line of sight(LOS) and 8.6 dB in non line of sight (NLOS) cases.

Antenna diversity can also be employed to combat inter symbolinterference in time dispersive channels. The correlation of thedelay spread values between two separate branches was estimat-ed at 0.4 in an NLOS micro-cell environment by measurements.In this case, the delay spread was reduced by 20 % by selectionof the branch having the smallest instantaneous delay spread.

Propagation prediction models

Predictions of propagation characteristics are necessary forproper coverage planning, the determination of multipath effectsand interference also need to be taken into account when plan-ning how to deploy a mobile network. In COST 231, extensivework has been performed to develop good models.

Four basic mechanisms describe the phenomena which influ-ence radio wave propagation: Reflection, penetration, diffrac-tion and scattering. For all practical predictions, these mecha-nisms must be described by approximations.

A three stage modelling process has been the basis for the prop-agation modelling work within COST 231:

1. The real (analogue) terrain has to be transformed to digitalterrain data.

2. Defining mathematical approximations for the physical prop-agation mechanisms.

3. Developing both deterministic and empirical propagationmodels.

A large amount of work has been carried out during the COSTaction, and only some highlights will be presented here.

Improved propagation models for macro-cells have been devel-oped. The COST 231 – Hata Model is an extension to the wellknown Okumura-Hata model. This is an empirical path-lossmodel originally valid for frequencies up to only 1000 MHz. InCOST 231, Okumura’s model has been extended to higher fre-quency bands. The COST-Hata Model is valid in the frequencyband 1500–2000 MHz, thus covering DECT, DCS1800 andUMTS bands. The validity of the model is restricted to largeand small macro-cells, with base station antenna heights aboverooftop levels adjacent to the base station.


4 Frequency Division Multiple Access / Single Channel PerCarrier.

Another important model has been developed, also as a modifi-cation and extension to earlier work. The COST 231 – Walfisch-Ikegami Model (COST-WI) is a statistically based path lossmodel for urban macro-cells. No topographical database is used,only characteristic values for buildings and streets are insertedto predict the coverage. The model distinguishes between LOSand NLOS cases. In the LOS case (e.g. for street canyons), freespace loss is assumed. Diffraction losses over rooftops are usedto model the NLOS case. This model has been accepted by ITU-R and is included in Report 567-4. The mean error using thismodel is ± 3 dB when the base station height is just aboverooftop level. Performance is poor, however, when base stationantennas are below the roof. This model is also restricted to fre-quencies below 2000 MHz.

None of these models consider multipath propagation.

Also small and micro-cell models have been developed, modelswhich are very useful for e.g. DECT-planning purposes. Thisincludes both 2- and 3-dimensional models based on ray-tracingtechniques. An example is the Ericsson micro-cell model. Thisis based on a mathematical method for path-loss predictions. Itis recursive, very simple and computation-time efficient. Thepath loss is determined along paths which follow the differentstreets.

Additionally, separate studies have been done on building pene-tration, indoor propagation models and models for tunnels, cor-ridors and other special environments, e.g. railway environ-ments and high-speed trains. More details are given in [1].

GSM and DECT systems

Work has also been carried out in COST 231 with relation tosecond generation personal communication systems, specific-ally GSM and DECT. The performance of basic versions ofsuch systems is relatively well known, and the focus of the workwas therefore on advanced features and possible limitations.These topics are of importance not only in relation to the fullexploitation of the potential of the systems, but also in terms ofpossible evolutionary transitions towards the third generationvia progressive enhancements of the radio and network per-formance.

Advanced techniques for GSM and DCS 1800

It is assumed that GSM and DCS 1800 work properly if theusable energy inside the equaliser window is 9 dB higher thanthe energy outside the equaliser window, i.e. carrier to co-chan-nel interference ratio is higher than 9 dB. Hence, if the Q16-parameter defined earlier is higher than 9 dB, GSM or DCS1800 will not be limited by multipath propagation.

Among the techniques that can be employed to GSM to increaseboth quality of service and increased capacity is the introductionof random frequency hopping. Quality is increased compared tothe non hopping case due to improved burst decorrelation forslow moving users. Capacity is increased by up to 70 % due tointerference averaging in the network, giving possibilities oftighter reuse patterns compared to the non hopping case.

Also the gain of using different antenna diversity schemes forGSM has been simulated. Typical values of diversity gain are

5 dB (ranging from 0 to 7.5 dB). Using ideal adaptive antennaarray and hence reducing interference and hence blocking cantheoretically increase GSM capacity by up to 400 % in a givenscenario. More detailed results are given in [1].

Extending the usage of DECT

Also DECT (Digital Enhanced Cordless Telecommunications)and its performance have been studied within COST 231. DECTis intended to provide cheap equipment and offer high capacitycompared to GSM. In addition, there is no need for frequencyplanning in DECT due to dynamic channel allocation.

DECT was originally designed to be a residential and businesscordless system, but lately, the use has also been extended topublic DECT and WLL5 applications. Extending the use ofDECT to outdoor scenarios creates challenges with regard tosystem performance due to the outdoor propagation conditions.In contrast to GSM there are no remedies in DECT to compen-sate for the influence of the radio channel, i.e. no channel cod-ing, interleaving or channel equalisation are present. Hence,DECT is vulnerable to interference and multipath propagation.

The delay spread parameter defined earlier in this article gives agood measure of DECT performance limitations due to multi-path propagation. There is a rule of thumb saying that in a sys-tem without channel equaliser, using GMSK6-modulation, therewill be an irreducible bit error rate (BER) of approximately 10-3

if the delay spread parameter exceeds one tenth of the symbolinterval. A BER of 10-3 is defined as performance limit for thespeech service in DECT. This gives a maximum tolerable delayspread in the order of 100 ns for a DECT implementation notemploying channel equalisation or advanced antenna diversitytechniques. Hence, DECT performance will in many scenariosbe limited by multipath propagation.

DECT tolerance to multipath propagation can, however, beimproved by use of adaptive sampling instants (sampling atmaximum opening of eye pattern) or use of advanced antennadiversity schemes. Error rate driven diversity schemes generallygive better performance than received signal strength drivendiversity schemes.

One advantage of the Time Division Duplex scheme used byDECT is that the uplink (hand portable to radio fixed part) anddownlink (radio fixed part to hand portable) are reciprocal whenmobiles are not moving. Hence, antenna diversity employed inthe radio fixed part will improve both uplink and downlink per-formance7.

In future PCS8 applications of DECT, both time dispersion andportable movement may be significant, and simple switch diver-sity will not provide sufficient quality. Even advanced diversity


5 Wireless Local Loop.

6 Gaussian Minimum Shift Keying.

7 This will not work for moving handsets, because the radiochannel will then change between uplink and downlink trans-mission.

8 Personal Communications System.

algorithms have limitations since they only deliver uplink gainor both. Equalisation can therefore be considered as a possibil-ity. Simulations have shown that a simple 2-state Viterbi equal-iser can extend the delay spread range (in high signal to noiseratios) to at least 600 ns.

Also experiences from a DECT field trial in a multipath en-vironment have been reported. The DECT trial system consistsof 240 handsets, 160 radio base stations and a central controlfixed part, and provides coverage and seamless handover out-doors and partly indoors in an area of about 2.5 square kilo-metres.

Keeping in mind the potential of DECT in an outdoor environ-ment both as a technology providing local mobility and as afixed radio access solution, one of the purposes was to under-stand the strengths and weaknesses and improvements neededon current versions of the DECT system in order to operate inboth indoor and outdoor environments.

Experience from the trial has given valuable information aboutthe expected cell range in a variety of different indoor environ-ments ranging from more than 50 metres in open hall areas toless than 15 metres in heavily reinforced areas. As a result ofthe limited cell radius careful site planning can reduce the num-ber of indoor base stations by more than 30 %.

In outdoor areas, it has been found that the speech quality isgenerally good as long as there is a free line of sight path be-tween the base station and the terminal, and no major reflectingobjects are in close vicinity. It has also been found that thespeech quality is variable in open squares (typically surroundedby reflecting buildings) even when there is a line of sight pathbetween the BS and the terminal and high average received sig-nal strength levels are measured. The link quality is also de-pendent on whether the user is moving or standing still, and isalso affected by the user’s orientation and positioning of thehandset relative to the base station.

The experience in the trial shows that the DECT technology is astrong candidate for providing speech services and mobility inindoor domestic/business/industrial environments. However, forproviding outdoor local mobility, the technology is relativelyimmature and too sensitive to radio propagation conditions.DECT performance would improve significantly if advanceddiversity techniques or a channel equaliser are introduced tocope with multipath propagation. This should be kept in mindwhen planning outdoor DECT implementations for public use.

Advanced radio interface techniques and

performance

General radio system aspects, with the focus on transmissiontechniques and their performance have been studied. Severaltechnical documents and reports dealing with analysis of radiointerface techniques for third generation mobile radio systemhave been published in the open literature. Both linear and non-linear modulation techniques are analysed and optimised forvarious radio propagation conditions.

The capacity of cellular TDMA-systems can be greatly increas-ed if enhanced techniques like adaptive modulation, adaptivecoding or dynamic channel allocation (DCA) are applied.

CDMA has been studied and performance results of DS9-CDMA and FH10-CDMA and hybrid systems presented.

If the advantages of TDMA and CDMA are combined, a furtherenhancement of the system capacity can be obtained. CDMAcan be mixed with OFDM11 to provide the system with a multi-ple access capability.

Potential radio interface subsystems

Time Division Multiple Access – TDMA

The most important single project that has provided results toCOST 231 is the RACE ATDMA project. The goal was todesign a radio interface based on the TDMA principles, fulfill-ing the following requirements of UMTS/IMT-2000 comparedto second generation systems like GSM:

• A wider range of services in a wide range of environments• Better quality and reliability• Higher capacity• Easier network planning and deployment.

A minimum number of radio bearers were selected to supportall services:

• Voice (high and normal quality)• Data service with low delay (9.6 kb/s - 2 Mb/s at < 30 ms)• Data services with long delay (9.6 kb/s - 2 Mb/s at < 300 ms)• Data services with unconstrained delay.

The concept of an adaptive radio interface was also important.The aim is that the radio interface should adapt to cell types andassociated propagation conditions, interference and sourceactivity factor. The system has been evaluated both by analysisand by the project simulation testbed, and the conclusion fromthe RACE ATDMA project was that it fully meets the require-ments, making it a suitable candidate for the UMTS air inter-face.

Code Division Multiple Access – CDMA

CDMA techniques were also studied in several projects withliaisons to COST 231. The UK DTI/EPSRC LINK PersonalCommunications Programme, project PC019 has carried out arigorous evaluation of CDMA with the following key object-ives:

• Study of UMTS requirements – teleservices and bearers

• Comparative study of FH- and DS-CDMA for UMTS serviceprovision

• Design and development of a DS-CDMA field trial system

• Demonstration of a working DS-CDMA radio link, with asubset of the proposed UMTS services.

At the end of the COST action, the project conclusions werethat a DS-CDMA radio link based on a single mobile, base sta-


9 Direct Sequence CDMA.

10 Frequency Hopping CDMA.

11 Orthogonal Frequency Division Multiplex.

tion and mobility manager for a sub-set of UMTS teleservicesare viable. Two common channels and three dedicated channelsper user are supported on the downlink, while the uplink sup-ports a common random access channel and two dedicatedchannels per user. The channels are code separated with a com-mon chip rate of 8.192 Mchip/s.

Studies have also been performed on a hybrid access systemcalled Code Time Division Multiple Access – CTDMA, whichincorporates both CDMA and TDMA aspects. Finally, studieshave been done especially on Joint Detection (JD) CDMA.

Broadband systems

In COST 231, broadband systems have been addressed as a sep-arate issue by WG3. The emphasis has been put on wirelesscommunication to mobile or portable equipment offering veryhigh data rates. “High” data rates are for this purpose defined tobe greater than 2 Mb/s to ensure no overlap with UMTS. Exam-ples of such systems are Wireless Local Area Networks –WLANs. These systems will be the main focus for the proposedfollow-on action (see below).

Both infrared (IR) and millimetre wave (mmw) technologieswere initially considered. IR were early eliminated due togrounds of eye safety, thus leaving the focus on mmw basedsystems, mainly operating at 40 and 60 GHz.

Millimetre wave propagation

A lot of work has been done on mmw propagation issues. In theRACE MBS project, studies have been performed on materialcharacterisation at 60 GHz, concluding that walls made of allmaterials except glass and plasterboard provide sufficient isola-tion to enable frequency re-use in neighbouring rooms. Propa-gation has been studied both by measurements and modellingfor indoor and outdoor scenarios. Both narrow- and widebandmeasurements have been conducted. A statistical model for thepower delay profile of the indoor mmw channel has been deriv-ed from the measurements. In the outdoor environment, theoxygen absorption is expected to limit cell radii to less than1 km.

Wireless LAN

Not only mmw based technology has been the issue. Also trans-mission techniques for HIPERLAN have been studied. DSSS12

and RAKE receivers, multicarrier or OFDM modulation, an-tenna diversity and adaptive equalisation are all techniques thatcan be used to overcome the limitations set by the widebandmultipath disperse channel.

Wireless ATM

Finally, MAC13 protocols for ATM access to the public B-ISDN have been studied. Other topics are protocols for randomaccess to a wired backbone network and ad-hoc networks. Sev-

eral anti-collision protocols are described, e.g. request/permitmechanisms, busy/free declaration by the use of inhibit bits onthe downlink. For ad-hoc networks with no central BS to con-trol the access, distributed protocols must be used either forsensing the channel at the transmitter or using handshake proce-dures.

The follow-up: COST action 259

The MoU for COST action 259 – Wireless Flexible Personalis-ed Communications – was prepared in April 1996 and approvedby the COST Technical Committee Telecommunications TCT[2]. The final approval has taken place at the October 1996meeting of the Committee of Senior Officials (CSO), and theMoU is now open for signing. A probable kick-off will be inJanuary–February 1997.

The main objective for the new action is to

“... increase the knowledge of radio system aspects for flex-ible personalised communications, capable to deliver differ-ent services exploiting different bandwidths, and to developnew modelling techniques and related planning tools, inorder to guarantee the continuity (and quality) of service,delivered by networks of widely different capabilities andstructures, across a number of environments.”

The title of the action tries to summarise the objectives.

The cost of the action is estimated at 17.7 million ECU.

The background for the new COST action is the medium/longterm scenario envisaged for telecommunications by the Euro-pean Commission for the time period 2000 – 2005. This periodwill be characterised by the emergence of PCS with a full inte-gration of user, terminal and service mobility, the IntegratedBroadband Communications (IBC) and network intelligence.The COST action is intended to play the same role as COST207 and COST 231 did in the past on GSM, DCS1800 andUMTS, but now on the progressive deployment of UMTS andMBS.

References

1 COST 231 Final Report. Digital Mobile Radio : COST 231view on the evolution towards 3rd generation systems.Berlin, Springer. To be published in November 1996.

2 Memorandum of Understanding for COST Action 259.Wireless flexible personalised communications. April, 1996.


12 Direct Sequence Spread Spectrum.

13 Medium Access Control.


Kaleidoscope

139


The teenager clicks with his PC mouse on the Internet icon,surfing for information somewhere in the world and has it print-ed out with illustrations. Or he collects his incoming e-mailfrom Internet friends in Australia and other continents. Or chatswith a friend in Africa. All he pays is the local telephone tariff!He is, maybe, somewhat annoyed that it takes too many secondsto get the information on his monitor, or that the sound qualityis below that of a CD. But never mind, he knows that this willbe better in a couple of months or perhaps a year.

Radio link systems (radio links), satellites, computers,information technology: Today, all this is taken for granted. Butthat was not the case in the IT Stone Age. And when did the ITStone Age end, – or are we still at the trailing end?

Until just some 40 years ago no telephone lines existed acrossthe Atlantic Ocean. Conversation between Europe and Americawas only possible on short wave radio when the ionosphericconditions were favourable. Then, the first transatlantic tele-phone cable was put in operation. Almost unbelievable: 24telephone channels between the two continents could be used!

In Norway at that time you could get through a telephone callbetween e.g. the two cities Oslo and Trondheim, provided thata You were lucky enough to have a telephone – the waiting list

was several years

b You had lent the telephone company NOK 3,000

c You could afford to pay express fee for long distance calls

d You had the time to wait for a couple of hours

e The transmission quality was sufficiently high to make con-versation possible.

No wonder that today’s teenager lacks the imagination toappreciate fully the enormous development that has taken placeduring just a few decades!

Radio links have played an important part in making Norway aworld pioneer in telecommunications. Below, we present thedevelopment which has taken place in the transmission networkof Telenor and its predecessors.

During World War II and the years to follow, telephoneconnections were a luxury in Norway, available only to a fewpeople. Telegrafverket, which had the exclusive monopoly onelectrical communications by wire and radio, was indeed not apopular state owned agency in Norway. This was, however, a littlebit unfair, as the parliament grants for telecommunications everyyear were less than needed to cope with the increasing demands.

In 1969 Telegrafverket was reorganized and was renamed Tele-verket. The next major reorganization took place on 1 November1994, when Televerket became a state owned limited companynamed Telenor. Today, the monopoly is almost dissolved, andTelenor has for several years been one of the world’s leadingtelecommunication companies. This position has been gained,partly due to the development of radio links carried out by Tele-verket/Telenor, The Norwegian Joint Signals Administration andthe Norwegian industry during the last 50 years.

The rest of this paper mainly concentrates on the activitieswhich have taken place within Telegrafverket/ Televerket and afew of the key persons involved. But first a brief look in the rearview mirror.

Prelude: A brief history of radio

First came Faraday, Maxwell and Hertz

The English scientist Michael Faraday (1791 – 1867) is perhapsone of the greatest experimental scientists who has ever lived.In 1831 he discovered the electromagnetic induction and gavean explanation of the electromagnetic fields.

The discovery of Faraday laid the foundation for the Scottishscientist and mathematician James Clerk Maxwell (1831 –1879) when Maxwell presented his theory on electromagneticwaves in 1873. His theory is defined by four differential equa-tions, which have since been a basis in all education in radiocommunication theory.

In 1887 the German professor Rudolf Hertz (1857 – 1894) per-formed the first practical experiment which proved thatMaxwell’s theory was correct. He used a 3 metre long copperwire with a 30 centimetre spherical ball at each end. In the mid-dle of the wire was an inductor fed by an interrupted DC curr-ent. Sparks were generated between the two ends, and electro-magnetic waves were emitted. From another copper wire whichran parallel to the first one he was able to extract small sparks.

In 1888 he performed an experiment with a set-up rather similarto a radio link system. He used parabolic disks as directionalantennas and transmitted radio waves in the 10 centimetre bandacross his laboratory.

Then followed Marconi and the spark transmitter

Based on Hertz’ first practical experiment the Italian GuglielmoMarconi (1874 – 1937) developed the spark transmitter anddemonstrated in 1895 that radio signals could be transmittedover several kilometres. He was granted a patent in England in1896, one hundred years ago.

The spark gap transmitter became the dominant method forradio transmitters until World War II. Even after the war it wasextensively used in emergency transmitters in ships.

In 1901 Marconi demonstrated the first radio telegraph com-munication across the Atlantic Ocean. Only five years later thefirst radio telegraph connection was established by Telegraf-verket between Sørvågen on the mainland and the islandsVærøy/Røst in Lofoten in the north of Norway. In May 1996 amuseum was inaugurated at Sørvågen to celebrate the 90 yearsanniversary of radio communication in Norway.

At the receiver end the technological breakthrough came in 1907when the American Lee de Forest developed the triode tube.

1934: Radio link system between Dover andCalais

The development within the radio field followed differentdirections. The first commercial radio link system was put intooperation between Dover, England, and Calais, France, in 1934.The radio frequency used was 1700 MHz (UHF), and the trans-mitter output power was 1 watt. 3 metre parabolic antennaswere used. The system was amplitude modulated and couldtransmit one telephone channel and one telegraph channel.

141

50 years of radio links in Norway The transition from the telecommunication Stone Age

B Y K N U T E N D R E S E N


The first multichannel radio link system was established byClavier between the Channel Islands and the English coast in1935. Due to the great depression in the 1930s, the experimentswere, however, not followed up. The principles, which werebased on the use of very short radio waves in the centimetreband, were further developed by the Armed Forces shortlybefore and during World War II. The development during thewar was mainly for military purposes, in particular radar sys-tems.

The first multichannel radio link system in the VHF frequencyband was put in operation in 1936 by the British Post Office(GPO) between Scotland and Northern Ireland. The capacitywas nine telephone channels, and a frequency of 65 MHz wasused with amplitude modulation. This modulation method, how-ever, gave rise to insurmountable problems in multichannel sys-tems due to intermodulation. As a consequence, both in Franceand in USA pulse modulation was tried and a few systems weredeveloped and produced. From 1939 onwards, frequencymodulation was used, first in Germany by Telefunken and laterin other countries.

Technological development during and afterWorld War II

The Englishman Ranwell developed the magnetron tube for theGHz band (centimetre waves) in 1939. It was to be used for radarsystems. The output peak pulse power was several thousandwatts. During the war both the magnetron and the klystron werefurther developed by British and American scientific institutes.

After the war the development of radio link systems progressedvery rapidly, and USA became the leader in this field. Publictraffic demand increased and the need for larger capacity grewrapidly. The need for world wide telephone connections alsoincreased and also the quality had to be improved. The require-ment was to obtain better linearity and lower noise in the sys-tems. Therefore, new technology on higher radio frequencieshad to be developed.

1947: AT&T with broadband radio link systemsfor TV and radio transmission

In 1947 the American Telephone and Telegraph Corporation(AT&T) put into operation a radio link system between NewYork and Boston. The capacity was 100 telephone channels inthe 4 GHz frequency band. In 1950 AT&T had in operation12,000 km of its TD-2 system for TV transmission.

In 1952 a radio link system between New York and San Fran-cisco comprising more than 100 stations was put into operation.

In Europe, the world’s first radio link system using travellingwave tubes (TWT) was put in operation in 1951. The link wasestablished between Manchester and Scotland for TV transmiss-ion and was operating in the 4 GHz band. In the 1950s severalradio link systems were implemented in Europe, both nationallyand internationally.

In 1952 more than 25 million circuit kilometres transmissioncapacity were in operation in USA, Europe and other countriesin the world.

International co-operation was promoted through the Interna-tional Telecommunication Union (ITU, founded in 1865) and itsConsultative Committees (CCITT and CCIR). CCITT (nowITU-T), the agency responsible for technical studies, operationand tariff questions, stated in its recommendation No. 40 in1951 that cables and radio link systems should be considered asequal transmission media for telephony. Radio link systemsshould as far as possible follow the recommendations issued forcable systems concerning quality and other relevant parameters.

CCIR (now ITU-R), responsible for technical studies concern-ing radio communication, issued its first recommendation forradio link systems in 1953 (CCIR Rec. No. 128).

As from the mid 1950s the role of the radio link systemsbecame increasingly greater. Already in 1972 more than twothirds of the long distance traffic and nearly 100 % of the TVand radio programs were transmitted by radio link systems inNorway.

1926: First public fixed radio telephoneconnection in Norway

In Norway the use of fixed telephone connections has a longtradition. The first fixed radio telephone connection was estab-lished between Kristiansund and Grip in 1926. The equipmentwas very simple. It operated on medium waves and usedamplitude modulation. During a ten year period approximately60 of these connections were established to isolated settlementson islands along the Norwegian coast. They were in operationright up to the 1960s.


Torbjørn M. Forberg

employed as a technician at Fornebo and later he became a tele-graph operator. In 1925 he won a competition in the use of aKleinschmidt perforator, and in 1927 he won a competition inoperating a Corona Morse telegraph machine.

Forberg now took his college education. He then worked at theOslo broadcasting transmitter at Ekeberg near Oslo, and onesummer on Spitzbergen. After that he attended the NorwegianUniversity of Technology in Trondheim, where he graduated asan M.Sc. in 1935, more than ten years older than his fellowstudents.

On his return to Telegrafverket, no suitable job for his qualifi-cations was available, so he worked for 4 to 5 years in thegoverning body Telegrafstyret as an engineer with the title ofoperator. In 1937 he moved over to Tryvasshøgda Radio inOslo.

The economy of Telegrafverket at the time was extremely bad.The Radio Office had an annual budget of merely NOK 50,000!This left no room for great investments.

Primitive medium wave radio stations forconnections to the islands

Together with other engineers Forberg installed a number ofsmall radio stations in the medium wave frequency band, mostlyto isolated islands in the north of Norway. In 1942 there weremore than 60 of these radio connections in operation.

During World War II and the following years he continued towork with the radio connections. He had a very good co-operation with Aldor Ingebriktsen, a fisherman, member of theNorwegian Parliament and head of the Parliament’sCommunications Committee, and later president of Lagtinget,the Law House of the Parliament. Ingebriktsen was a very eagerspokesman and promoter for the island connections, being arepresentative for northern Norway. “If we don’t manage toimplement these radio connections, I dare not return home,” hesaid, and received full response from Forberg.

In 1945 Forberg was promoted divisional engineer A in Tele-grafstyret and advanced later to higher positions.

The first public Norwegian radio links

Just after the war the term “radiolinje” entered the communi-cations debate in Norway. The word “linje” (line) was widelydiscussed. “Linje” was used for physical wire connections in thenetwork. France used the word “hertzien cables”, and USA usedthe term “radio links”; should we in Norway use the word“ledd” (link), “kjede” (chain) or something else to describe aradio relay system?

Well, in spite of terminology problems; the first Norwegianradio link, with a capacity of one telephone channel and onetelegraphic channel, was tried out between Haugesund and theisland Utsira in 1945 by regional engineer Arne Bjørntvedt,Rogaland telephone district, and communication officer Lehnefrom the Army. They had taken over German military equip-ment left over in Norway when the war ended. The radio linkequipment was produced by Telefunken and operated in theUHF frequency band (540 MHz).


A prelude to the first civilian radio linksystems in Norway

A pioneer from the high Arctic North

Torbjørn M. Forberg is indisputably “the radio link pioneer” inTelegrafverket. He was born in 1901, in the same year thatMarconi demonstrated the first radio telegraph transmissionacross the Atlantic Ocean. He grew up in the small communityKvitnes at the Tana fjord in northern Norway. The Post Officebelieved that the place was a community belonging to the min-ority Same group (Lapps) and renamed the postal addressMuoregággu. “Today, I am slightly in doubt if I was reallyborn,” says the now 95 years old veteran.

The initial developments took place under very poor economicand technological conditions. “We planned and built a tele-phone network for eternity, but the technological developmentwas so rapid that in a few years the systems were outdated.When I started in Telegrafverket, the radio tube was not yetinvented. When I ended my career the tubes were outdated,” hesays, referring to transmitter tubes for high frequencies.

The possibility of getting further education after the primaryschool in Kvitnes did not exist, but Torbjørn’s father saw anadvertisement for a secondary school in Nordfjordeid in westernNorway. “You may apply for this school, on the condition thatyou complete it in one year,” said his father. Torbjørn went toNordfjordeid where he met young people from all over Norway.But this was a private school, and after a year Torbjørn had towalk across the mountains to an authorized school in Volda,where he passed the examination with the most brilliant resultsever in his career.

Training as a telegraph operator at StavangerRadio

Shortly before World War I, in 1914, Telegrafverket had order-ed equipment for Stavanger Radio from Marconi and startedtests in 1918. But Telegrafverket lacked telegraph operators andhad to start a training course. Forberg attended the first courseand in 1921 he could start his career at Stavanger Radio, astation which somebody has characterised as “built on shearoptimism and ignorance”.

The station was located very close to the North Sea to minimizethe distance to America. The antenna was almost 3 kilometreslong and was mounted on 100 metre high masts. The receiverpicked up a lot of cosmic noise but very little of the wantedsignals. On the other hand, the spark transmitter in Stavangerspread noise to the whole world!

Crystal detector receivers were used, and the informationreceived was engraved on phonograph rolls. Even if somedevelopment had taken place during the war, the project wasrather hopeless. In order to avoid a complete failure, Marconisent over to Stavanger a competent engineer who worked dayand night. He built a directional frame antenna and succeeded inimproving the receiving signals.

In 1925, Stavanger Radio was put out of operation and theactivity was transferred to Fornebo near Oslo. Forberg was

In 1946, Telegrafverket put in commercial operation the firstpublic radio link system between Sørvågen and the islandsVærøy/Røst by using the same Telefunken equipment. Later,similar systems were put in operation between Sørvågen –Bodø, Bodø – Svolvær and Gjøvik – Ringsaker.

The Norwegian Defense Research Establishment(FFI) develops and demonstrates a NorwegianSHF radio link system Bergen – Haugesund1

By the end of the 1940s, FFI, in co-operation with ChristianMichelsens Institute in Bergen, developed a radio link system inthe 3.3 GHz band. It was put on trial between Bergen and Hauge-sund in 1951. The development work was done by Norwegianengineers who had worked at British research institutes during thewar. The engineers Bjørntvedt and Gustavsen from Telegraf-verket participated very actively in the measurement on the sys-tem at the stations Rundemannen and Stord. The trials went well,and a connection of one telephone channel was put in operation.

The system was later extended to Haugesund Telegraph Stationusing the mountain Steinfjell as a repeater station. In Bergen thesystem was extended down to Bergen Telegraph Station byusing a passive reflector on the mountain Sandviksfjellet. Thecapacity was now extended to 8 telephone channels by using amultiplex system (type MEK 8) which was provided by Tele-grafverket. The multiplex system had been taken over from theGerman military forces and was designed for use with copperwire. Also British military equipment was used. FFI producedthe equipment and installed and commissioned it, and also train-ed the maintenance crew. With the capacity of these 8 telephonechannels in addition to the 2 existing wire connections, thecapacity between Bergen and Haugesund was increased by afactor of 5!

A/S NERA Bergen was established, and the FFIsystem was extended to Stavanger

FFI further improved the system. In order to extend the systemto Stavanger and also establish a radio link system betweenBergen and Oslo, FFI signed a contract with A/S NERA, Oslo.In 1951 NERA established a separate division in Bergen, NERABergen, to manage the contract.

The contract committed NERA to deliver a radio link systemwith a capacity of 12 telephone channels in the 3 GHz band. Inthe new NERA equipment, the klystron CV67 was used. Due toinstability problems with the klystron the system to Stavangerwas delayed until 1955 before it was in operation. Because ofthe problems caused by the imported CV67 klystron NERAdecided to develop their own klystron in co-operation with FFIand NTH. The new klystron was produced by NERA.

The establishment of the NERA division in Bergen and theimplementation of the radio link between Bergen and Oslo gavethe kick-off for the building of a country wide radio link systemin Norway. The Norwegian Joint Signals Administration, FFSB

(Forsvarets Fellessamband) was established in 1953 to co-ordinate the implementation of the network. Another separateradio link network was established by the Norwegian Air Force(Luftforsvaret) in order to solve the communication problemsbetween the military airports in Norway. They used equipmentfrom the American company Philco in the 7 GHz band, but thesystem was not very efficient and was operated for only a fewyears. The military radio link systems are described elsewhere.2

Good co-operation between Telegrafverket andthe Army on the operating level, but it was windyat the top

Between the operative staffs in FFI, FFSB, and Telegrafverketthe co-operation was always good. Delegates from both partiesparticipated in international fora. They were not always inagreement, but: “When one is attending a meeting and has toexpress oneself in public it may happen that one uses expres-sions that are a bit provocative,” says Torbjørn Forberg with asmile.

Also at the top levels in Telegrafverket, FFI, FFSB, and theindustry the co-operation at the beginning was good. As anexample, one may mention that the annual convention “Studie-møtet i radioteknikk og elektroakustikk” (Study group in radiotechnology and electro-acoustics) was started in 1948 on the ini-tiative of Helmer Dahl, chief of research, FFI’s radar division;Sverre Rynning Tønnesen, managing director of Telegrafverket;A. David Andersen, manager and chief engineer, DavidAndersen Radio; and Fr. Brodtkorp, R&D engineer, TandbergRadiofabrikk. The first convention was held in Åsgårdstrand,with radio link systems as the opening topic. Also in almost allof the following annual conventions – 49 to date – radio linkshas been a central theme.

In the first conventions the discussions between the representa-tives of Telegrafverket, FFI and FFSB were, however, veryaggressive. The discussions also went on within the organisa-tions. On the surface the issues were the use of radio linksversus cable systems, the choice of frequency bands to be used,reliability of the systems, and the economy. There were, how-ever, deep undertones. The level of the discussions might indi-cate that not all of the participants would be suitable for jobs inthe diplomatic corps!

In retrospective one may say that the debates were in fact veryuseful. No problems were hidden under the carpet and withstubbornness the participants stressed the technology to itsutmost to prove their points. Without this endeavour from allparties involved, Norway would hardly been able to establish itsstrong international position in the world market today.

A study trip to USA in 1950 gave Telegrafverketthe start-up kick

In 1950, an event took place that really initiated the start up foran independent development of the radio link systems insideTelegrafverket. Through financial resources made available by


1 In general, Norwegian abbreviations will frequently be usedin the text, e.g.:FFI: The Norwegian Defense Research EstablishmentFFSB: The Norwegian Joint Signals Administration

2 E.g. Knut Endresen: Fra topp til topp. Kampen om radiolinjene(From peak to peak. The fight for the radio links). INTRA sivil-ingeniør Knut Endresen, Oslo. ISBN 82-993291-0-8.

the Norwegian Broadcasting Company (NRK), which beganthinking of implementing TV in Norway, Torbjørn Forberg gota 3 months grant from Standard Telefon og Kabelfabrik to studyradio link systems in the USA. He visited several radio link sta-tions, met the key people and acquired a good knowledge ofmaintenance experience, traffic prognoses and future plans forimplementing radio link systems. Of special interest was thebroadband systems TD-2 which had been put in operation be-tween New York and San Fransisco.

The reports from this system and others indicated that broad-band radio link systems were a very promising method forestablishing a communication network in mountainous Norway,and that less manpower and equipment resources were neededthan one had previously believed. The systems could be imple-mented quickly and economically. Future expansion of the sys-tems would require small additional costs, and the reliability ofthe systems was very satisfactory.

Great scepticism, because “radio is always radio ...”

“On my return to Telegrafverket I became the real EXPERT onradio link systems,” says Forberg. He was giving several lec-tures in Telegrafverket on the promising new technology andthe stability of the systems: “But I had to be very careful, for thescepticism was strong and many were very doubtful about thisnew technology. They said: ‘Radio is always radio and weknow its performance ...’ The American systems operated on 3– 4 GHz . When I once said that I thought that radio link sys-tems could obtain the same reliability as fixed copper wire andcable systems one of the listeners to the lecture said: ‘Now weshall hear this, too ...!’ ” But, as mentioned above, ITU (TheInternational Telecommunication Union), had already in 1951said in their recommendation (CCITT rec. No. 40) that cablesystems and radio link systems should be equal as transmissionmedia.

Budget proposal for a radio link system in 1951was turned down in favour of a coaxial cablesystem

In January 1951 a budget for an extension of the existing mili-tary system Bergen – Haugesund – Stavanger (which was nowalso used by Telegrafverket) across the mountains to Oslo wasproposed. The system was planned to start from Jåttanuten closeto Stavanger, via Skaulen (1575 metres above sea level),Gaustatoppen (1883 metres above sea level), and to Tryvass-høgda close to Oslo, a distance of 331 kilometres. It was pro-posed to use a 36 channel frequency modulated system fromMarconi, England, on this route. On the route between Bergenand Stavanger the proposal was to change the existing FFI sys-tem with a system from Storno, Denmark. The latter system waspulse modulated with 24 channels operating in the VHF band.

The proposal for the new Stavanger – Oslo link was, however,rejected by Telegrafverket. Later on, in 1953, a Marconi systemwas installed between Bergen and Stavanger. The system waslater moved to Steinkjer – Namsos in 1954.

The official reason for Telegrafverket for not implementing theproposed link was lack of investment capital. With the limitedfinancial resources Telegrafverket preferred to extend the co-axial cable system with a capacity of 600 channels between

Oslo and Gjøvik, with a cable across Hardangervidda to Bergen.Installation of the cable between Oslo and Gjøvik had started in1946. The total system Oslo – Gjøvik – Bergen was completedin 1956, after ten years installation time.

Positive field tests in Trøndelag

In co-operation between Telegrafverket and FFI an experiment-al system was put up between Gråkallen close to Trondheim andOffenåsen, Steinkjer in the summer of 1951.3 Wave propagationmeasurements were carried out over the 89 km hop during oneyear . The frequencies used were in the VHF (160 MHz), UHF(540 MHz) and SHF (3.3 GHz) bands.


3 Bjarne Hisdal: Feltstyrkemålinger Trondheim – Steinkjer(Field strength measurements Trondheim – Steinkjer). Tekn-isk Ukeblad, No. 18, 1953, p. 363.

Field strength measurement in the VHF, UHF and SHF bands at Gråkallen inTrondheim of signals transmitted from Offenåsen, 1952


Left: This Marconi system with 48 channels in the VHF band wasused e.g. on the Steinkjer – Namsos connection, 1953.

Right: The Philco SHF link with 24 channels in the 5 – 6 GHzband with time division pulse amplitude modulation was usedby the Norwegian Air Force for NATO connections to the air-

ports Bardufoss – Lista – Sola during and after the Korean war1950 – 1953

Power panels for the Marconi 48 channels VHF link are carriedup to the site Offenåsen on the radio link between Steinkjer and

Namsos in 1954

A VHF yagi antenna being mounted at the Falkhettasite near Rørvik on the Trondheim – Bodø link 1956

The conclusion was very clear: “The reliability is primarilydependent on the power supply system and the quality of theradio equipment and to a very little extent on the atmosphericconditions if the planning is performed well. The propagationconditions in Norway are very favourable. Therefore thereliability of the system should not be worse that in othercountries,” writes Forberg in “Verk og Virke”.

The first systems from Marconi: Trondheim –Steinkjer – Namsos on VHF

In 1952 Telegrafverket ordered radio link equipment in theVHF band with a capacity of 48 channels from Marconi for theroute Trondheim – Steinkjer. The system was opened for trafficin 1953 and a year later was extended to Namsos with equip-ment removed from the Bergen – Haugesund link.

In 1954 FFI opened the military radio link system betweenBergen and Oslo. It had a capacity of 22 telephone channels and36 telegraph channels and was operating in the 3.3 GHz band.At first, Telegrafverket was not interested in this link, but laterleased several channels on it.

During the following years Telegrafverket installed severalradio link systems in the VHF band with equipment fromMarconi. The first system with a capacity of 36 channels wasput in operation between Bodø and Svolvær in 1955.

Low capacity radio links to islands

During the period 1954 to 1957 several pieces of single channelsystems in the VHF band was ordered from Norsk Marconi andStandard Telefon og Kabelfabrik. The equipment, which wasproduced in Norway, was inexpensive and rather simple. It wasmainly used to replace the old medium wave radio equipment inTrøndelag and northern Norway.

Low capacity systems were also bought from the English com-pany Pye. They were used in the local networks and had amaximum capacity of 6 to 8 channels.

In 1959, some 2,000 circuit kilometres of island connectionswere in operation.

Gjøvik – Trondheim: Coaxial cable lost to radiolink system

It was originally planned to connect Oslo with Bergen and Trond-heim with the extensions of the coaxial cable system first estab-lished between Oslo and Gjøvik. As mentioned above, the systembetween Gjøvik and Bergen was completed in 1956. The plansfor extending the system from Gjøvik to Trondheim were, how-ever, dropped, and it was instead decided to build a radio link.

The radio link was implemented during the period 1955 – 1956,using as sites the link stations built by the Norwegian Air Forcefor the Philco link. The Philco link had been established duringthe Korean war 1950 – 1953. The new radio link system had 48telephone channels. It was produced by Marconi and operated inthe VHF band.

Severe icing problems on the mountain peaks

The link between Gjøvik and Gråkallen near Trondheim hadmainly rhombic VHF antennas. Some yagi were also used.Severe icing occurred, in particular on the rhombic antennas. Tosolve the problem ordinary 12 volts transformers used for melt-ing frozen water pipes were mounted as de-icing equipment onthe antenna elements. “The amount of power needed for meltingthe ice was rather high!” says Asbjørn Nilsen. For the yagiantennas plastic cylinders were mounted around the antenna ele-ments in order to minimize the icing problems.

All the way to Kirkenes in 1960

In 1957, the radio link proceeded northwards. The gap betweenNamsos and Bodø was closed by a VHF radio link. BetweenBodø and Svolvær a system was already in existence since 1955.

In 1959 a radio link was built between Bodø and Narvik, operat-ing in the 400 MHz band with a capacity of 60 channels. Thesystem was produced by Elektrisk Bureau, Norway, and wassimilar to a system put up between Bergen and Florø on thewest coast in 1957.

In 1959 – 1960 Telegrafverket implemented the so-called“Arctic coast link” with a capacity of 48 channels along thecoast of Finnmark between Hammerfest and Vardø. At the sametime several single channel radio link systems were imple-mented on Finnmarksvidda to give better connections to theLapp (Same) areas.


70

60

50

40

30

20

10

00 20 40 60 80 100

Dipole voltage, µV

Antenna height aboveground, metres

437 MHz

To obtain a free Fresnel zone one had to use high masts, as evidenced by fieldstrength measurements by Elektrisk Bureau of signals received in Arendal from

the Risør transmitter as a function of the antenna elevation, 1958

Between Tromsø and Hammerfest a system operating in the2 GHz band with a capacity of 120 channels was put inoperation in 1963. It was produced by Elektrisk Bureau and waslater expanded to 300 channels.

Southbound with equipment from ElektriskBureau on UHF

From Oslo and southwards the radio links were operating onUHF. Elektrisk Bureau was the supplier with equipment in the400 MHz band and a capacity of 60 channels. In 1955 the sys-tem between Oslo and Arendal was put in operation.

In 1956 the system was extended to Kristiansand to giveconnection to Jutland, Denmark.

In 1955 a system between Oslo and Fredrikstad was opened,and in 1957 a similar system was, as mentioned above, estab-lished between Bergen and Florø.

Well prepared for the expected national plan foran FM and TV network

During the years 1950 to 1958 long term planning was made tomeet future requirements. More than a hundred station sites hadbeen selected all over the country, many of them more than1,000 metres above sea level. Altogether more than 200 siteshad been surveyed.

The surveys were very time consuming since one was depend-ent on optical sight between the sites. It was hard physical work

1914 to seek the loneliness on Tronfjell which he considered aholy mountain. He gave lectures on Eastern philosophy at OsloUniversity, and he had plans to build a World Peace Universityon the mountain.

Sites of great beauty

The sites selected had often a very beautiful scenery. In an art-icle in “Verk og Virke” in 1958 , Harald Hauge Heskestad be-comes almost lyrical in his description of the site Hestmannen(The Horseman) on the route Sandnessjøen – Bodø:

“The radio link station Hestmannen is situated like a SoriaMoria castle on the shoulder of the rider. From the westernside of the building one can see how the ocean waves arebreaking along the shores more than 500 metres below. And


4 S.D. Sissons: I storm på Tronfjell (In gale force on Tronfjell).Verk & Virke, No. 1, 1959, p. 19.

and required good physical fitness, but according to Forberg: Itwas almost immoral to receive monetary compensation forwalking in the mountains, as mountain tourists had to pay forthe same experience!

In order to select the proper route for the radio links it wasnecessary to estimate the future traffic demand and its routing,the length of the hops, the Fresnel zone clearance, interferenceconditions, the climatic conditions at the site, physical accessfor maintenance, the possibility to have electrical power to theplanned station, and so forth.

The highest site between Oslo and Trondheim was Tronfjell,1,666 metres above sea level. The existing access road ended atan altitude of 850 metres. From that point there was only a foot-path further to the top where the measuring equipment had to beplaced. The mast used for the measurements and a prefabricatedcabin were brought to the top by a helicopter and were firmlyguyed. The mast was telescopic for adjusting the height of theparabolic antenna used for the propagation measurements.

The icing problems on the site were severe, and one stormynight the mast was blown down. It was then re-erected andsecured by better guys.

One day in December 1958, the Marconi engineer S.D. Sissons,4

together with Hans Walland from Telegrafverket and twotechnicians, were on their way up to the top carrying equipmentwhen they were suddenly overtaken by a heavy snow storm.The visibility was down to 5 metres and they lost their sense ofdirection in spite of their compass, and they planned to dig acave in the snow for the night. Fortunately, Kåre Johansen, whohad previously arrived on the top site to install equipment,guessed what was happening and he started banging on somemetal plates which he found at the site. By following the soundthey managed to reach the top after a very tough and exhaustingwalk.

Another technician who had started towards the top a few hoursearlier had, however, not arrived. After 6 hours of tough wadingin deep snow he managed to return to the valley again, and arescue expedition that was set up could be cancelled.

When a site had finally been selected, the installation peoplestarted the planning of the infrastructure. They had about twoyears to plan how to transport more than a thousand tons ofconstruction material to isolated mountain peaks all overNorway. Very often access roads or funiculars had to beconstructed, and in some cases helicopters were used. Thenfollowed the planning of the maintenance scheme. Therefore,one was well prepared when the plans for a nationwide radiolink network to serve FM broadcasting and TV transmitterswere proposed in 1959.

The Tronfjell site attracted many visitors. One of them was Mr.Beer, a disciple of Sri Ananda, who settled on Tronfjell wherehe is now buried. This Indian philosopher came to Norway in

The site on Aaleberget in Fredrikstad 1958. The height of themast was 50 metres


Parabolic antennas and V-shaped reflectors mounted in mastsat the Tryvann site in Oslo (Elektrisk Bureau, 1958)

The building on the Hestmannen radio link site islocated 500 metres above sea level near the Arctic

circle. The antennas are located at 568 metres abovesea level. (1957)

the eagles are flying around the hat of the rider. The yagiantennas in the mast seem to attract the eagles, because oneday when we arrived at the station we heard a lot of noisefrom the antennas. When we came closer, we saw the eaglestaking off.

From the top of Hestmannen you can see towards the southDønnamannen close to Sandnessjøen, and towards the westone can see Trænastaven which is the stick belonging toDønnamannen. The Arctic Circle crosses through the bay atthe northern end of Hestmannen, and further to the north isRødøyløva. In the direction of Telesund towards north eastone can see the glacier Svartisen. Hidden behind Svartisen isGlomfjord. At the inlet to the fjord is Ørnes. From there theroad goes to the next radio link station, Kunna. The roadturns in a valley which opens towards the ocean. The housesare secured to the ground by using guys in order to withstandthe storms from the north when the wind is compressed be-tween Skrovfjellet and Skjeggen.”

Radio link systems: At first a small group within the Radio office

At the beginning the radio link activities were taken care of by asmall group within the Radio office (Radioanleggskontoret) ofthe board of Telegrafverket (Telegraf-styret). Head of the Radio office wassenior engineer Paul Falnes.

In 1955 only four people were work-ing in the Radio link group, headed byForberg: Hans Fremming, Alf KetilHageler, Sverre Knudsen, and AsbjørnNilsen.

Asbjørn Nilsen, born in 1926, had firstattended a signalling course in theBritish Navy in Scotland and there-after the radio school at the MaritimeCollege in Oslo. After four years atsea as a telegraph operator he took hisengineering examination at Stock-holm’s Tekniska Institut in Swedenand was employed by the Radio officein 1955.

Alf Ketil Hageler, born in 1925, wasemployed by the Radio office in 1953.He was an electronics engineer fromThe Norwegian University of Techno-logy, NTH, and had taken his diplomaon radio link systems in 1953. Duringhis thesis work he participated activelyin the propagation tests betweenGråkallen and Offenåsen mentionedabove.

Hans Fremming, born in 1926, wasfirst employed as a telegraph operatorat Gardermoen airport. He then tookan M.Sc. degree at an American uni-

Harald Hauge Heskestad

Asbjørn Nilsen

versity. He came to the Radio link group in 1955 and startedworking with the islands connections.

Harald Hauge Heskestad, born in 1925, started in Telegraf-verket in 1944. After having completed Telegrafverket’sinternal training course (Lavere kurs), he was ordered to startworking as a telegraph operator at Arendal telegraph station.But when the war ended in Norway in May 1945 he was tem-porarily ordered to guard the German radio station at Lista air-port. The German occupants had established radio connectionsbetween the airports and their central command during the war.Heskestad took Telegrafverket’s higher course in 1949 – 1950.After serving for two years in Lofoten he went to Switzerlandand got his M.Sc. degree at Eidgenössische Technische Hoc-hschule in Zürich. He started work in the Radio link group in1957.

1959: The Radio link group is established as aseparate office

In 1959 the Radio link office was established as a separate off-ice inside the Radio technical division, headed by senior en-gineer Forberg. The number of employees, now nine, soonincreased to more than 30 persons. As the work load increased,special groups were established:

• Planning (headed by Fremming)• Implementation (headed by Ketil Hageler)• Maintenance (headed by Asbjørn Nilsen).

A maintenance centre was also established. This group wasplanned by Lars Håland and Ole Johan Haga and was headed byRolf Sæhle.

Frequency problems and TV transmission pavedthe way for SHF

Use of the VHF band started to create problems for the Marconiequipment. The expansion of FM broadcasting and TV trans-mitters in the VHF band created interference problems for theradio link systems.

In the early 1950s the discussion on future TV transmissions inNorway began. TV would require substantially more bandwidththan could be realised in the VHF band. Therefore, the use ofSHF frequency band had to be used. The radio links using theVHF band was therefore steadily relocated further to the northof Norway as Telegrafverket started to implement SHF systemsin the southern parts of the country.

TV opens a new era for the radio links

On 25 June 1957 the Parliament (Stortinget), made the formaldecision that a country wide TV network was to be implementedin Norway. The decision marked the start of a new era for theradio link system development in the transmission network.

Until then, the radio link budgets had been very meagre. Now,one suddenly got sufficient government resources for planning amodern radio link network, and the former intense discussionabout the use of frequency bands could be ended. The new radiolink systems were planned to operate in the SHF band with acapacity of at least 600 telephone channels. As far as possible thenetwork should be implemented by using common infrastructuretogether with the Norwegian Broadcasting Company, NRK. Theexpenses would then be less for both parties. One would havepossibilities for common re-routing and more economical maint-enance and supervision of the systems. A co-operation withFFSB in order to use their infrastructure was also proposed, butthis was turned down for military security reasons.

In 1960 – 1961, Telegrafverket had some 115,000 circuit kilo-metres of multichannel radio link systems in use in the network.This capacity amounted to 17.8 % of the total capacity of thenetwork and had passed the capacity of the coaxial network(17.5 %). Open wire lines and high frequency systems oncopper cable were still the dominating transmission media inTelegrafverket with 64.7 % of the capacity.


Long term plan 1959 – 1965 for covering the entire country with broadbandradio links for telephony and TV

Country wide plan for broadband radio linksystems

As mentioned above the Radio link office (Radiolinjekontoret)was established in 1959 and the first long term plan (1959 –1965) for implementation of a radio link network for telephonyand TV transmission covering the entire country was presented.

The backbone system should be:

• A two way radio link system for up to 960 telephone channels

• A one way radio link system for TV transmission from Oslo

• A two way back up radio link system for up to 960 telephonechannels, which could alternatively be used for transmittingTV, and could automatically switch between the telephoneand TV traffic.

• A two way narrow band auxiliary radio link system for con-trol and supervisory signals, service channels and soundchannels for radio programmes.

The plan was for four main routes:

• From Tryvasshøgda in Oslo towards the north passing theeastern part of the lake Mjøsa, along Østerdalen to Trondheimand further north to Hammerfest and Kirkenes.

• From Tryvasshøgda to Jonsknuten close to Kongsberg, alongthe south coast and further on to Bergen, Møre on the westcoast, and then north to Trondheim.

• From Tryvasshøgda towards the west coast to Bergen acrossthe mountain range Langfjellene.

• From Trondheim to Bodø – Harstad – Tromsø – Vadsø.

The terminal stations as well as the repeater stations should havethe possibility of inserting and extracting TV signals along the way.

1959: The radio link system Oslo – Karlstad linksNorway with Eurovision and Nordvisjon

The first broadband radio link system for television transmissionin Norway was opened 1 October 1959 between Oslo and Karl-stad, Sweden, as a connection to the Eurovision and Nordvisjonnetworks. On 10 December 1959 the Nobel Peace Price cere-mony was televised to the Eurovision network for the first time.

The Norwegian part of this radio link system was financed byTelegrafverket alone. The equipment was delivered from Marconiand had a capacity of 600 telephone channels. In addition, an aux-iliary link with a capacity of 60 channels was installed.

Simple equipment – but hard to maintain

The terminals were installed in a tower near Karlstad and atTryvasshøgda in Oslo with repeaters at Sunne in Sweden andRundelen in Norway. The first transmission was from a meetingin the Nordic Council (Nordisk Råd). The system operated inthe frequency band 3.8 – 4.2 GHz. The link was very vulnerablein use as it was not a heterodyne system with an intermediatefrequency. A shift oscillator of 213 MHz simply changed thefrequency at each hop. Since no IF frequency existed in the re-peaters it was impossible to make an IF loop on the systemwhen faults occurred, and fault location was very difficult.


Measurement mast at Tronfjell in winter 1958 – 1959

Field strength measurements at Tronfjell, 1666 metres above sea level.Winter 1958 – 1959

Even if the technical solution and the equipment were rathersimple, the maintenance problems were great. Therefore, noadditional equipment of this type was ordered. The system wasin 1966 replaced by a system produced by the Italian companyMagneti-Marelli, which later became GTE, Italy.

Safer with a solid partner as a supplier

The dominating suppliers of VHF and UHF radio link systemsto Telegrafverket in the 1950s were Marconi, England, andElektrisk Bureau, Norway. Both companies possessed a hightechnical know-how in the radio field, and it was consideredimportant that the responsibility for technical weaknesses couldbe pinpointed to the suppliers. And as the management of Tele-grafstyret in the early 1950s were very sceptical to radio linksystems, the engineers found it safer not to be guinea pigs, try-ing smaller producers.

1960: Norwegian TV officially opened, with aNERA radio link Oslo – Bergen

NERA Bergen won its first contract with Telegrafverket for theOslo – Bergen radio link in 1960. The system had a capacity of300 channels and was installed between Tryvasshøgda, Oslo,and the TV transmitter station Ulriken in Bergen with repeaterstations at Jonsknuten, Gaustadtoppen, Snønut, Lysenut and

Fanafjellet. It was put in operation on 20 August 1960 when theTV was officially opened in Norway by King Olav. From therepeater station Lysenut, a temporary branch link was installedto feed the Bokn TV transmitter, giving TV coverage to thenorthern part of Rogaland.

STK won the contract for the Oslo – Trondheimsystem

The next broadband radio link system for TV transmission wasinstalled between Oslo and Trondheim. In 1961 the system wascommissioned to feed the TV transmitter at Melhus and was oneyear later expanded to also include telephone traffic to Trond-heim. Standard Telefon og Kabelfabrik won the tender withequipment from STC, England, in competition with NERABergen. The capacity was 960 telephone channels.

A decisive argument for choosing the STK offer was that theirequipment had higher output power. Therefore, one could usesmaller parabolic antennas than by using the NERA equipmentand thereby reduce the icing problems. NERA was at that timeusing SHF triodes with only 1 watt output power. STC – likeMarconi and Siemens – used travelling wave tubes with muchhigher output power. Later, NERA also changed to travellingwave tubes, giving an output of 10 watts.

The STK system was no big success. It had a complicated tubesystem, designed for manned stations. There were a lot of sys-tem failures giving much maintenance work. Several modifica-tions of the equipment were done, but the maintenance costswere still high and the system was taken down after ten years ofoperation. This was not to the dissatisfaction of the coaxialcable fans!

A man and his dog ...

Countries like Denmark and England had relatively easy accessto the stations and could therefore have maintenance personnel atthe stations. With a few exceptions Telegrafverket’s stations wereunmanned. To the surprise of the sceptics this resulted in a highsystem reliability. The reason was simple: Visiting the stationshigh up in the mountains was so physically demanding and tookso much time that the technicians preferred to do a thoroughlygood maintenance job when they had reached the top, and theymade sure that everything was in order before they left.

“At the manned stations or at the stations it was easy to reach itwas more tempting to postpone the work until the next day oruntil the next visit,” says Asbjørn Nilsen, who quotes a Danishcolleague: “The staffing at the stations should consist of a manand a dog. The task of the man is to feed the dog, and the dogshould prevent the man from touching the equipment.”

It was not without danger to do the work at the mountain peaksin Norway. Harald Heskestad reports from a visit together withtwo technicians to Hestmannen just before Christmas in 1957:The station is situated on a steep mountain 500 metres abovesea level. A heavy snowstorm started and it was impossible togo outside the station. They were isolated for three days and hadfood for only one day. Luckily, they had brought with themthree bottles of the special Norwegian Christmas beer. Byrationing the consumption to one bottle a day, the spirit waskept high. After three days they managed to climb down the


The Trolltind site 924 metres above sea level on the Tromsø –Hammerfest connection, 1960. The cabin with the equipmentwas burnt down in 1967, making evident the need for alterna-tive transmission paths and for transportable mobile radio link

equipment for restoration

Hard to distinguish the details? Provisional site with 300 chan-nels VHF link at Trolltind 924 metres above sea level on the

Tromsø – Hammerfest link, 1963. The wind is pushed low by themountain formation and a 2 – 3 metres deep heap of snow is formed

mountain side by using a rope in the deep snow. They lookedwith envy to an eagle who flew elegantly from the yagi antenna.After this experience, emergency food was placed at everymountain station.

The implementation of the broadband radio link systems duringthe decade 1960 – 1970 was quite extensive. Ring structureswere built in southern Norway; Oslo – Trondheim and Oslo –Kristiansand – Stavanger – Bergen, and later on, Bergen – Åle-sund – Trondheim. In parallel, the system Trondheim – Bodø –Tromsø – Hammerfest – Vadsø was implemented.

The first systems were implemented by using equipment fromMagneti-Marelli, Italy. Later, NERA Bergen had completed thedevelopment of their equipment NERA 960-4G (NL50) and thissystem was operating satisfactorily. Therefore, the rest of thebroadband system was mostly implemented with this type ofequipment.

The radio links were built as a 2+1 system with a capacity for960 channels:

• One channel transmitted telephone traffic• One channel transmitted TV signals• One channel was used as a common reserve.

An additional auxiliary radio link had a capacity of 300 chann-els and was produced and delivered by Elektrisk Bureau.

From Trondheim towards northern Norway NERA Bergen andEB equipment were used up to Tromsø. From Tromsø toHammerfest equipment from Marconi was at first used, but waslater, in 1969, replaced by new 1800 channel equipment in thelower 6 GHz band produced by NERA Bergen.

The country wide plan for a broadband radio link system for TVand telephone traffic was realised as shown in Table 1.

In addition to the backbone systems, 19 broadband radio linksystems were implemented as spur links to the TV main trans-mitters that were located outside the backbone system. Inaddition, 37 radio link systems were in use with a capacity of120 to 300 channels, 41 radio links with a capacity of 24 to 60channels, and 126 systems with a capacity of 1 to 12 channels.

The first breakthrough for delivery from NERAto Telegrafverket was in 1960

As mentioned above, the first radio link system – with acapacity for 300 telephone channels – for TV transmission fromOslo to Bergen was ordered from NERA Bergen in 1960. Butthe real breakthrough for deliveries from NERA came in 1964when Telegrafverket ordered equipment for the backbone sys-tem. Then the capacity and the quality of the NERA systemswere acceptable to Telegrafverket.

The following systems were delivered by NERA:

• Bergen – Ålesund

• Ålesund – Trondheim

• Trondheim – Mosjøen – Bodø

• Bodø – Harstad – Tromsø

• Oslo – Bergen

• Oslo – Gothenburg

Elektrisk Bureau delivered 300 channelsauxiliary systems

Elektrisk Bureau A/S delivered radio link systems in the 2 GHzband with a capacity of at first 120 channels and later modified to300 channels. The systems were used as auxiliary systems on theroutes Kristiansand – Stavanger – Bergen – Ålesund – Trondheim– Mosjøen – Bodø – Harstad – Tromsø – Hammerfest.

During the period 1962 to 1970 more than 80 % of the radiolinks delivered were produced by the Norwegian manufacturersNERA Bergen and Elektrisk Bureau.

Transistorized equipment from “the small birdon the eagle’s back”

The world’s first solid state radio link equipment was producedby the American company RCA. One of these systems (CW60)


The snow was often 2 – 3 metres deep at the Kistefjell site inTroms 1004 metres above sea level. Therefore one had two

entrances: One summer entrance through the door at groundlevel, and another winter entrance on top of the building

Major reorganisation

During this period a major reorganisation of Telegrafverket tookplace. The managing board Telegrafstyret became Teledirektor-atet in 1969, and Telegrafverket was renamed Televerket. In1970 Forberg retired as leader of the Radio link office and wassucceeded by Hans Fremming, who unfortunately died the daybefore he should take his position. Ole Johan Haga was thenappointed new leader of the Radio link office.

Ole Johan Haga first attended the technical college of theNorwegian Air Force as a radio link specialist. He then gra-duated at the University of St. Andrews, Scotland, in 1960 andwas employed at the Radio link office in the same year.

During his first years as an engineer he was working with lowcapacity systems – often single channel radio link systems forconnections to islands along the coast of northern Norway.Later on, he was working more with large capacity systems andwas responsible for implementing the broadband system Kristi-ansand – Stavanger – Bergen and further from Trondheimtowards the north of Norway as the TV radio link systems wereimplemented. Haga headed the Radio link office until the nextmajor reorganisation of Televerket in 1972 when he moved to


was delivered to Telegrafverket by their Norwegian representa-tive NERA. This system had a capacity of 300 channels and wasinstalled between Ålesund and Molde in 1965 – 1966. This wasthe first transistorized equipment in operation in WesternEurope at that time. Even the output power tube was solid state.The equipment was easy to maintain and the maintenance costwas very low.

The good performance of the systeminspired NERA Bergen to start theirown development of solid state equip-ment – at first with 300 channelcapacity and later with 1800 channels.

Bernt Ingvaldsen, who was president ofthe Parliament and chairman of theboard in the parent company of NERABergen, said: “A small bird flies higherby starting from the eagle’s back.”

The 1800 channel equipment was firstput into operation between Tromsøand Hammerfest in 1969. Thenfollowed Oslo – Bergen in 1970, andOslo – Gothenburg in 1972.

Main supplier km Year No. of Auxiliary TV channels channels channels

Oslo – Karlstad Marconi 170 1959 (1800)* + 600 60 Two way

Oslo – Bergen NERA, Bergen 360 1960 300 1 One way

Oslo – Trondheim STK 397 1963 (1800)* + 960 120 One way

Oslo – Kristiansand Magneti-Marelli 265 1963 (1800)* + 960 960 One way

Kristiansand – Stavanger Magneti-Marelli 179 1963 960 300** One way

Stavanger – Bergen Magneti-Marelli 164 1963 960 300** One way

Bergen – Ålesund NERA Bergen 268 1965 960 300** One way

Ålesund – Trondheim NERA Bergen 252 1966 960 300** One way

Trondheim – Mosjøen NERA Bergen 327 1965 960 300** One way

Mosjøen – Bodø NERA Bergen 204 1965 960 300** One way

Bodø – Harstad NERA Bergen 204 1966 960 300** One way

Harstad – Tromsø NERA Bergen 134 1966 960 300** One way

Tromsø – Hammerfest NERA Bergen 244 1969 1800 300** Two way

Hammerfest – Vadsø Siemens 323 1967 1800 12 Two way

Oslo – Bergen, new route NERA Bergen 452 1970 1800 300 Two way

Oslo – Gothenburg NERA Bergen 215 1972 1800 300 Two way

Table 1 The implementation of the country wide plan for broadband radio link for television and telephone traffic

* 1800 channel from Magneti-Marelli in 1967. ** Delivered by Elektrisk Bureau.

Ole Johan Haga

the Transmissionsection. Today, he isdirector of TelenorInternational.

In 1972, Einar Eke-berg was appointedhead of the Radio linkoffice, a position heheld until he retired in1991. After gra-duation at theNorwegian Universityof Technology in1953 he wasemployed at ElektriskBureau where he

worked as an R&D engineer on the radio link system that Tele-grafverket had ordered for the route Oslo – Arendal – Kristian-sand.

In 1954 Ekeberg was employed as a senior engineer at FFSB.He there had a central role in the co-operation with NATO, firstin connection with the FFI radio link system Oslo – Bergen, andlater on the communications infrastructure of NATO.

Also at NERA Bergen major changes occurred. The companywas bought by Elektrisk Bureau in 1977 and became a divisionin Elektrisk Bureau. In 1988 the Swedish company ASEA andthe Swiss company Brown Boveri were fused and renamedABB. ABB gathered all their Norwegian activities in ElektriskBureau.

In 1992, ABB bought all the shares in Elektrisk Bureau, and thelatter was renamed ABB Konsernet i Norge, and NERA becameABB Nera AS. Then, in 1994, ABB Nera became NERA ASA,an independent limited company, listed on the Oslo stockexchange.

Country wide radio link network II

The first broadband radio link system was implemented as acommon network for Telegrafverket and NRK, consisting of a2 + 1 system with 960 channels. The first systems using elec-tronic tubes were gradually replaced by solid state 1800 chann-els equipment.

During the decade 1970 – 1980 an alternative country wideradio link network was built in order to increase the capacity inthe transmission network in connection with the automation ofthe telephone exchanges and also to improve the reliability ofthe network.

The radio link network for NRK and Televerket was now builtas two separate networks:

• A network with 1800 or 2700 channels for Televerket

• A separate network with 960 channels for the NorwegianBroadcasting Company, NRK.

Planning started in 1972 – 1974 and the new network for NRKwas implemented from 1976. This network was built by usingequipment from NERA, which had won the contract after


25

20

15

10

5

0

1955 1960 1965 1970 1975 1980Year

Equivalent telephone circuits

in mill kilometres

During the 1955 – 1980 period the broad band radio link network in the SHFband 4 – 6 GHz expanded exponentially

The Tryvann tower, Oslo, completed 1961

Einar Ekeberg

Pho

to: F

rank

Aar

hus


The country wide radio link system for telephone and programmelines, 1980

Snow track at Tronfjell Vassfjellet radio link station in Trøndelag (Photo: Frank Aarhus)

Røverkollen radio link station in Oslo, main station in the alternativeradio link network (Photo: Frank Aarhus)


Tifjell radio link station in Trøndelag(From photo file of NØ.EA)

As a consequence of the major implementation of thedigital radio link network, new masts had to be erected on

several stations, as for instance at Ålfjell

Saurdalseggiradio linkstation in Sognog Fjordane

international competition withGTE (formerly MagnetiMarelli) in 1975. The completerenewal of the NRK networktook more than 10 years, as thefinancial resources madeavailable by NRK were limi-ted.

The new NRK network forradio and TV was supplement-ed with equipment from GTEon the spur links. In total,NERA has delivered 80 – 90 %of the radio link systems usedin the NRK network.

A new pioneeringera

In parallel with the renewal ofthe NRK network, the back-bone radio link network fortelephone traffic was extendedand renewed all the way toKirkenes. An alternativenetwork was implemented, inparticular in northern Norway.In the southern part of thecountry a ring structure wasestablished. Televerket thuspossessed two parallel radiolink networks covering thewhole country.

“This introduced a new pio-neering era for Televerket,because we had to start afresh,building new stations onalternative mountain peaks allover Norway. Nevertheless, thework was completed within tenyears”, says Inge Vabø, whoheaded the Planning groupwithin the Radio link officefrom 1974 to 1993.

After obtaining his B.Sc.degree at the University ofStrathclyde, Glasgow, IngeVabø was employed at theRadio division of ElektriskBureau in 1961. ElektriskBureau had then quite anextensive production of radioequipment, including e.g. FMbroadcast transmitters, HF shiptransmitters, military equip-

ment and radio link systems. Through L.M. Ericsson (LME) inSweden, which was a majority shareholder in the company,Elektrisk Bureau had a substantial export of radio link systems,in particular to Latin America. Telegrafverket and FFSB werealso important customers of UHF radio link systems.

In 1967, LME startedto transfer more andmore of the radio linkknow-how to Gothen-burg and in the beg-inning of the 1970sElektrisk Bureaustopped producingradio link systems. In1969, Vabø wasemployed in theRadio link office ofTeleverket, at firstworking with islandconnections and lowcapacity systems. In1974 – 1976 he wasactively engaged inthe implementation ofTeleverket’s firstsatellite earth station at Eik in Rogaland.

2700 channel systems developed

In the 1970s, when Televerket was extending its broadband ana-logue radio link network, Televerket negotiated a contract withNERA Bergen for the development of systems with a capacityof 2700 channels in the upper 6 GHz band. The contract wassigned in 1974 and was financed by the Research Establishmentof Televerket and the Ministry of Communications.

The first 2700 channel system was installed between Oslo andHamar in 1978. Televerket had also bought a system betweenOslo and Bergen with 8 hops, and several smaller systems inboth southern and northern Norway. On some of the stations theNERA equipment had stability problems, and it was thereforedecided to issue an international tender for 2700 channel equip-ment. The contract was won by the Japanese company NEC andthe NEC equipment was installed on some of those stationswhere the NERA equipment had stability problems. The NECequipment was taken down in 1995 after 15 years of almostfault free operation.

The transmission becomes digital – on a high ambition level

The first digital radio link had been put in operation betweenSvelvik and Drammen in 1969. It was produced by the Italianmanufacturer Telettra and was operating in the 13 GHz band witha capacity of 24 channels, in accordance with the US hierarchy,and was one of the first digital radio links in use in Europe.

Televerket was now entering its digital era. Televerket’s Re-search Establishment was established in 1967. The researchdirector, Dr. Nic. Knudtzon, had new ideas from his previouswork with radio relay systems at FFI and his experience intransmission network planning as head of the Telecommuni-cation Division at SHAPE Technical Centre in the Hague. Hisambitions were so high that many – including NERA and othertelecommunication companies – were of the opinion that he wastoo far ahead of the technological possibilities. But the scepticswere wrong!


The NERA system NL180 with 2700 channels,1 + 1 radio and modem

Inge Vabø

Former technical director at NERA, Per Fremstad writes:

“Televerket had at an early stage a clear strategy for digi-tizing the network. This strategy was not taken seriously atNERA. The opinion was that Televerket never followed theirplans. With hindsight one might say that Televerket did justthat. The new general manager of Televerket, Kjell Holler,well backed up by technical director Ole Petter Håkonsenand the research director Nic Knudtzon, did a great job tomodernize the telecommunication network during the ten yearperiod 1978 to 1988 ...”

Telettra/GTE won the first round, but NERA the next

The implementation of the new high capacity radio link networkin Televerket followed as a consequence of the digitization ofthe main telephone exchanges.

Standard Telefon og Kabelfabrik won the tough internationaltender for the main digital telephone exchanges with its System12. After a substantial delay the first digital telephone exchangewas put in operation in Trondheim on 13 June 1986, makingTeleverket one of the most advanced telephone companies inthe world. The next System 12 exchange was opened at Økernin Oslo in September 1986, and from then on everything wentrather fast.


At the end of 1987 Televerket had in operation two transitexchanges, 100 subscriber exchanges and 140 concentratorswith more than 400,000 telephone lines and some 130,000 com-munication lines. In the following year, in 1988, altogether 152exchanges and 275 concentrators were in operation, and System12 had a total of 640,000 subscriber lines. At the end of 1993more than 50 % of the telephone subscribers in Norway wereserved by System 12. And already in 1989 Televerket had

13 GHz mast link mounted on the roof of Drammen telephonestation

Antennas mounted on an antenna tower at Rønvik-fjellet in Bodø 1988 (Photo: Truls Arnesen)

Digital radio link equipment mounted in a closed container

issued a new tender foradditional digital exchanges to bein operation 1991 – 1994. Thetender was won by Ericsson –and delivery was again delayed.

Between all digital exchanges common channel signalling wasused, in accordance with CCITT Recommendation No. 7.

140 Mbit/s 16QAM system is developed

The digitizing of the telephone network involved very strongrequirements to the transmission network. Televerket thereforesigned a contract with NERA already in the early 1980s for the

development of a 140 Mbit/s, 16QAM radio link system in theupper 6 GHz band to be used for connecting the digitalexchanges to be ordered. The development at NERA, however,took longer than expected.

As a consequence, Televerket had to order equipment fromTelettra/GTE after an international competition. Also this equip-ment was delayed, but as even System 12 was more than a yearlate, the Telettra/GTE delay was not the critical one. The first


From the installation work inDrammen

NL100, NERA’s first digital radio link

Planned broadband digital radio link network 1985 – 1989 with capacity 1920 channels (2 x 140 Mb/s + 1x140 Mb/s common backup)

high capacity digital system was put in operation in the back-bone network in 1985.

NERA had its high capacity 140 Mbit/s system ready for deliv-ery in 1987 – 1988, and the system was implemented on a largescale in the backbone network. Equipment was also boughtfrom NEC, in particular in the 11 GHz band, and from Siemensin the 18 GHz band, where NERA could not supply such equip-ment at that time.

An auxiliary system with capacity 34 Mbit/s was implementedin parallel with the high capacity systems. Both NERA andGTE supplied equipment for these systems.

From 16QAM to 64QAM

A new contract was signed with NERA for the development ofa 64QAM, 140 Mbit/s system to be used in the lower 6 GHzband. This development was ready in 1990. In the meantime,Televerket had bought similar equipment from Fujitsu, Japan,for the Oslo – Drammen route.

Several 64 QAM systems were ordered and delivered fromNERA in the early 1990s.

Today, the complete backbone radio link network for telephonetraffic is digitized with equipment delivered from NERA,Telettra/GTE, NEC, Fujitsu and Siemens. The NRK radio linknetwork is still analogue for TV transmission.

Regional and local networks

In parallel with the implementation of the backbone network, anextensive number of radio links in the regional and localnetworks have been implemented. Altogether more than 1800stations are used in these networks in addition to the more than150 stations in the backbone network. The regional and localnetworks have analogue links with capacities from one tele-phone channel and up to 2700 channels and digital links withcapacities from 30 to 1920 channels. A majority of the radiolink manufacturers in the world are represented.


The radio link site at Ulriken in Bergen (Photo: Frank Aarhus)The Tyholt tower in Trondheim (From photo file of NØ.EA)


No. of analogue radio links No. of digital radio linksCapacity, Circuit Commun.channels length, km circuits, km

New Dismounted Total number New Dismounted Total numberin operation in operation

1 3,373 3373 11 9 151 0 0 0

2 412 949 7 4 29 0 0 0

8 34 275 0 0 0 0 0 0

10 52 523 0 0 0 0 0 2

12 1,376 16,507 3 2 70 0 0 0

24 220 5,278 2 1 10 0 0 0

30 4,946 153,672 0 0 0 37 13 252

60 1,236 74,142 0 0 16 6 3 7

120 4,524 550,020 0 7 55 23 7 131

300 652 195,510 0 6 12 0 0 0

480 6,865 3,373,248 0 0 0 29 6 133

960 7,820 11,208,960 8 10 90 0 0 0

1800 6,045 11,511,360 1 3 39 0 0 0

1920 4,447 19,600,320 0 0 0 21 0 53

2700 2,170 5,859,810 0 0 19 0 0 0

Total 44,172 52,553,947 32 42 491 116 29 578

Table 2 Number of radio links as per 31 December 1989. In addition to the numbers listed, The Norwegian Broadcasting Company,NRK had two digital radio links (each with 30 channels) and 46 analogue links (33 with 960 channels, 11 with 30 channels, onewith 60 channels, and one with 24 channels). The NRK radio links are included in column 2 (circuit length) and column 3 (com-munication circuits, km)

Altogether, there are more than 8000 transmitters/receivers inthe radio link network, in frequency bands from 140 MHz up to38 GHz.

In parallel with the implementation of the fixed networks,special mobile link systems have been supplied from USA foremergency and restoration purposes. The systems are built tomilitary specifications and are very robust.

Optical fibre cables take over in thebackbone network

Today, the activity on new radio links in the backbone networkis negligible. The optical fibre cable systems are taking over. Asan example, it may be mentioned that in 1988, 49 new digitalsystems were put in operation. 12 of them were using coaxialand fibre cables (565 and 140 Mbit/s) and 37 were radio links(140 Mbit/s). From then on, the number of optical cables hasbeen steadily increased, at the expense of radio links.

In this connection it may be of interest to note the cooperationbetween Telenor and the Norwegian State Railways, NSB.Under this agreement Telenor installs optical fibre cables alongthe railways. The capacity is so high that Telenor has now rent-ed the excess capacity from NSB in order to keep new operatorsaway from the Norwegian market as from 1 January 1998, whenthe competition becomes quite open. However, the said contractis contested by other companies, stating that it is in conflict withEU rules. The matter may eventually be solved by the ESAcourt.

The phased out radio links are used inother countries

As from 1992, the analogue radio links which had been inoperation since the 1970s were gradually taken down after ser-ving their purpose and earning a lot of money for Televerket.The equipment was now offered to East-European countries. Alot of the equipment was delivered to the mountainous Albania


140 Mb/s 64QAM radio link,3 + 1 terminal

Passive reflectors may connectstations which are not withinline of sight of each other

Risdalsheia radio link station (Photo: Truls Arnesen)

Andfjell radio link station at Saltfjell (From photo file of NØ.EA)

which had a primitive and quite insufficient transmissionnetwork. The delivery and the installation of the equip-ment in Albania was sponsored by the European Develop-ment Bank, EBRD in London. Some equipment has alsobeen sent to other developing and newly industrializedcountries, for example Vietnam. Thus, the Norwegianradio link know how is also shared with others.

Today, NERA is one of the leading manufacturers ofradio link systems in the world and has delivered equip-ment to a great number of countries.

Several Norwegians, among them Inge Vabø, have beenconsultants on radio link systems in a number ofcountries in Europe, Africa, Asia, and Latin America.

Televerket and the powercompanies

In the 1960s, the power company Oslo Lysverker (nowOslo Energi) decided to use radio links in its own trans-mission network. After applying for a license Oslo Lys-verker got permission to build its own network. Thefirst systems were installed between Oslo and Hall-ingdal in 1965. Later in the 1960s building of the hydro-electric power plants in the Aurland valley was begun,and Oslo Lysverker again applied for a new license forestablishing radio links for the new transmissiondemand.

But now, the general manager of Telenor, Per Øvregard,put his foot down: The power companies had already anextensive radio relay network thatcould be used for telephone tra-ffic. But Televerket had themonopoly! Previously, in the1950s, the military had built theirown separate network, and nowthe power companies representeda new threat to the monopoly. Hetherefore decided to have acommittee under the chairmans-hip of Haakon Nymoen to evalu-ate the situation.

When Oslo Lysverker in coop-eration with other power compa-nies in addition applied for a lice-nse to implement radio linkstowards Gudbrandsdalen and furt-her on to Møre in connection withthe building of a hydro-electricplant at Aura, Øvregard said no.Instead, it was decided to set upan agreement to the effect thatTeleverket should implement theradio link network for the powercompanies at their expense. Aquite extensive network was built,following the valleys Hallingdal,Hemsedal, Aurlandsdalen andGudbrandsdalen, and further onto Møre on the west coast.


Salten radio link station and TV/FM transmitter in Nordland (Photo: Håkon Ringdal)

Also the state electricity administration, NVE, was included inthe agreement, and when NVE needed transmission capacity topower plants in Telemark, Tokke, Sira Kvina and to Sweden, aradio link network was built by Televerket.

At first, the capacity was 24 analogue channels, but later digital2 Mbit/s systems were implemented. The suppliers of these sys-tems were NERA and GTE for equipment in the 1.5 GHz band,and EB and Telettra in the 400 and 800 MHz bands.

In 1980, the personnel resources needed for this work wasincreasing, and the management of Televerket had to stop it. Asa result the power companies were again granted a license tobuild and operate their own networks. At that time no one couldforesee that the monopoly would ever end for Televerket.Today, Telenor is faced with the consequence that the powercompanies are becoming formidable competitors in the trans-mission infrastructure.

Epilogue

Norwegian radio links during 50 years. The radio links gaveTelegrafverket a new life, but the road was long and the fightfor success was tough. At the beginning, the economy was badand the technology was obsolete.

The engineers knew what they had at hand, but one would haveto be a Jules Verne to foresee what was hidden in the future. Forexample, leading scientists visualized a future with a worldwide network based on a number of short wave radio stationsfor telephone connections between the continents. More daringwas the technology leading British telephone company GPO,which developed the first transatlantic cable. To be on the safeside, they used thermionic valves of pre-war design, having aproven life of some 30 years. The life time of modern tubeswould only be known a couple of decades ahead. And no sen-sible person could dream of transistors, satellites or optical fibrecables.

Since 1899 Telegrafverket had the monopoly and the respons-ibility for all telecommunications in Norway, an almost


Storhogen radio link station at Sogndal 1173 metres above sea level, winter1981 (From photo file of NØ.EA)

The tallest construction in Norway was put up by Televerket atHamnefjell near Båtsfjord in Finnmark, 1988. This radio link

mast is 242 metres high and weighs 270 tonnes. It is put togeth-er by cylindrical sections of galvanized steel tubes. The

construction has an internal maintenance elevator in the 2metre wide cylinder. Steel wire guys of 40 to 66 mm in diameterprevent the top of the mast from swaying more than 2 metres at

wind bursts of twice the strength of a hurricane

impossible task. And, as with GPO, one dared not to take thereally great risks – it was safer to use the equipment and techno-logy at hand, with which one was well familiar. And with limit-ed resources, e.g. rather simple radio connections and radiolinks were established in the more isolated areas. With today’stechnology in mind, the connections were very primitive, but inreality an amazing pioneering work was done, starting fromscratch with a poor economy.

During World War II several Norwegian engineers were attach-ed to British military research establishments. Immediately afterthe war The Norwegian Defense Research Establishment (FFI)was established, and later on The Norwegian Joint SignalsAdministration (FFSB). The engineers entered the arena withnew and unorthodox ideas, and they pinched the first hole in themonopoly. Suddenly there were two – and in fact three if thePhilco radio link of The Royal Norwegian Air Force is included– operators on the arena. The competition which followed wastough and the discussions sometimes noisy, but became in factan important stimulus for all parties, including the industry.Later on, Televerket got its own research establishment whichcould concentrate its activities on planning for the future, with-out bothering about the then long telephone waiting lists andother trivialities. Later on, also more funding became available.The technology soon reached an international top level, and theself confidence increased.

Today, Norway has one of the best radio link networks in theworld. Telenor shares a place in the front line with the world’sleading telephone companies. Norwegian telecommunicationindustry is the market leader in important telecommunicationsectors. Who is to receive the laurels? The honour must be shar-ed between FFI/FFSB, the industry, and Telegrafverket/Tele-verket/Telenor, because the results have been generated throughthe competition, the interaction and the cooperation between theparties involved. But what really matters is the pioneering workof the many individuals who took the lead and opened newpaths into the future. The number of individuals is so high that itis an almost impossible task to name them all. Being a closeobserver through a few decades, the author therefore decided todescribe as objectively as possible what actually happened,without mentioning special achievements and arguments.Names of individuals have been carefully omitted, with theexception of the very first pioneers and those who have beeninterviewed to establish certain facts.

A family of new technologies was born

Today, the activity on new radio links in the backbone networkof Telenor is negligible, as optical fibres and satellite technologyare rapidly gaining bigger shares. In local and regional networksthere is still great activity, for Telenor as well as for newoperators. The époque of the radio links is thus far from ended.

More important, however, is that the radio links have raised abig family. The technological know-how directly and indirectlycreated by the radio link development has been a catalyst and insome cases decisive for the development in adjacent areas. Oneexample is the Råö project near Gothenburg, Sweden – the veryfirst earth station for satellite communication in Europe basedon European technology. The project was the result of a threeparty cooperation between the telephone administrations inDenmark, Norway and Sweden.


Radio link know-how: From Arctic conditions in Norway to a tropical site inTanzania. (Photo to be published in an ITU handbook in 1997)

Radio link station for Oslo Lysverker in the Aurland mountains

Soon followed the Tanum earth station south of the border be-tween Norway and Sweden for international satellite communi-cation. The fatherhood was shared between the Norwegian andthe Swedish telephone administrations.

Then followed earth stations at Eik in Rogaland and at Nittedalnear Oslo and satellite communications to the oil and gas fieldsin the North Sea. Next came Isfjord satellite station, makingSpitzbergen an integral part of the Norwegian network.

Mentioned must also be maritime satellite communication,where Telenor has played a leading part. And optical fibrecables, where a Norwegian company holds a couple of worldrecords.

A thrilling future

What about the future? The telecommunication monopolies inthe world are withering away. New competitors are entering theNorwegian market, and some of Telenor’s former suppliers andcustomers are now competing with Telenor. On the other hand,Telenor is trespassing into other countries. New markets arewon in Europe, Africa, Asia, including China and Indonesia.And Norwegian industry and consulting companies have mark-ets almost all over the World, often including complete com-munication systems over land, under oceans and by way ofsatellites.

The radio link has got offspring who carries the developmentfurther ahead. The protecting monopoly umbrella has beenreplaced by more efficient offensive weapons: know-how andfront line technology.

Quo vadis, where are we going? Without making definite pro-phecies, it feels safe to assume that future development will beeven more exponential, the competition increasingly tougher,and it is better not to slumber behind the wheel. Certainly: thefuture will be a true thriller!

Acknowledgement

The present paper is based on material made available by IngeVabø and interviews with Torbjørn Forberg, Ole Johan Haga,Harald Heskestad, Asbjørn Nilsen, and Inge Vabø, and the auth-or’s own observations during four decades. The author is inparticular indebted to Inge Vabø, who has also assisted in trans-lating the present paper into English.


The radio link backbone network, 1996

Documents

Contents · – a presentation of the ITU-T T.120 series recommendations, Trond Ulseth..... 95 Charging of ATM services, ... In today’s society investment in competence is said