OpenAccess Africa eBook

Copyright © 2005 United Nations ICT Task Force

All rights reserved. Except for use in a review, the reproduction or utilization of this work or part of it in any form or by electronics, or other means now known or hereafter invented, including xerography,

photocopying, recording, and in any information storage, transmission or retrieval system, including CD-ROM, online or via the Internet, is forbidden without the written permission of the publishers.

The views expressed in this book are those of the individual authors and do not necessarily reflect the views or positions of the United Nations ICT Task Force, the United Nations itself, any of its organs or agencies,

nor of any other organizations or institutions mentioned or discussed in this book, including the organizations to which the authors are affiliated.

Published byThe United Nations Information and Communication Technologies Task Force

One United Nations PlazaNew York, NY 10017

[email protected]

Acknowledgements

I would like to express my gratitude to the many individuals and organizations that haveworked to bring this publication to fruition. I am especially appreciative of the peopleand organizations that have carried the “Open Access” workshops forward from thefirst occasion in Stockholm, Sweden to the latest one in Maputo, Mozambique;Professor Björn Pehrson at the Stockholm Royal Institute of Technology, AmericoMuchanga of the Eduardo Mondlane University, the whole ICT secretariat at Sida,Steve Song and Heloise Emdon at IDRC and Ambassador Astrid Dufborg, theconvenor of the United Nations ICT Task Force Working Group on the EnablingEnvironment. Without the sincere effort of all of them the workshop series wouldnever had taken place, and we would not have had such a rich pool of resources forproducing this publication.

I would also like to convey my high appreciation for the United Nations ICT TaskForce secretariat team, especially Ms. Cheryl Stafford for all her English lessons, Mr.Robert de Jesus for his excellent formatting skills, Karin Wenander for her helpfulscrutiny and Ms. Enrica Murmura for assisting me with the supervision of the printingprocess. I would also like to direct a special thanks to Mr. Rocco Callari and Mr. MatiasDelfino at the Graphic Design Unit of the Outreach Division of the Department ofPublic Information for working with me to create a beautiful cover design for this book.

Lastly, but not least, I would like to thank all the contributors to this publication for the privilege and the opportunity to bring all their ideas together under one cover. Theirgraciousness in responding to my hard deadlines and pestering e-mails will always befondly remembered.

i

Contents

Preface iii

Introduction 1

SECTION I: CHALLENGES AND CONCEPTS

Open Communication – Open AccessBJÖRN PEHRSON

5

Networking for Africa – Open Access and Other IssuesRAHUL TONGIA

16

A Layered Market ParadigmANDERS COMSTEDT

19

SECTION II: ACTORS AND EXAMPLES

Nurturing Entrepreneurship in MozambiqueDANIEL MANNESTIG, CONSTANTINO SOTOMANE, JAMO MACANZE

25

Promoting African Research and Education NetworkingROY STEINER, NYASHA TIRIVAYI, MIKE JENSEN, KARANJAGAKIO, PAUL HAMILTON, JACK BUECHLER

34

A New Survey of Investment in Education and Research Networking in Africa by Development Agencies and other Organizations – SummaryKATE WILD

45

Problem-Oriented Capacity-Building based on the Introduction of Information and Communication Technologies as an Enabler of Socio-Economic DevelopmentBJÖRN PEHRSON

58

Appendix IFight Against Malaria in Africa through ICT by MIMComGODFREY CHIKUMBI

74

Appendix IIMzOpen.Net: Wireless Hotspots in University Residences in MozambiqueENEAS HUNGUANA, QARIN HJORTZBERG-NORDLUND,ALBERTO MUCHANGA, ERIK STACKENLAND, JON ÅKERGÅRDEN

81

Appendix IIIRINEX – Building a Bridge to Reduce the Digital Divide and Enhance the Use of ICT as a Tool Eradicate PovertyISSA NKUSI, CLAUDE HAKIZIMANA, COCO MUSANINGABE

86

ii

SECTION III: OPPORTUNITIES

Open Access Networking in Africa: The FiberAfrica ProposalRAHUL TONGIA

93

Open Access: How Good is it for Africa?STELIOS PAPADAKIS

114

Developing a Fibre Optic Backbone for AfricaJABULANI DHLIWAYO

122

Innovation to Improve Access to ICTMARTIN CURLEY

134

Open and Closed Skies: Satellite Access in AfricaPolicy Reform and Regulatory Issues in Bridging the Digital Divide through Satellite TechnologiesMARTIN JARROLD

140

Economic Development in Africa Powered by Mobile TelephonyOLOF HESSELMARK, ANDERS ENGVALL

155

List of Abbreviations 167

About the Authors 171

iii

Preface

Communication lies at the heart of every society. Communicating with others does notonly allow us to share knowledge, goods and to build our livelihoods but it also gives us the opportunity express affection and emotions with our close ones.

The advancements in information and communication technologies have over the years, from the invention of the telegraph to the introduction of the Internet, gone hand inhand with modernization and development.

Yet, millions of people have never made a telephone call. The absence of modern tools for gathering information and communicating is particularly evident on the continent of Africa. Poverty in Africa is widespread and without the ability to communicate thecontinent will remain poor and isolated, lacking the means to participate in the globalsociety. Poor people in rural areas have to travel for days to trade their products, to get food and water as well as to receive government services. Such travel is often risky and expensive and the outcome uncertain. With the use of information and communication technologies (ICT) this risk can be diminished substantially and services can bedelivered effectively. This relation has been proved over and over when we see thatpoor people often chose, if they have the opportunity, to make use of ICT for servicesand economic information gathering. The expansion and deployment of ICT istherefore an essential part of any comprehensive effort to achieve the MillenniumDevelopment Goals and should be part of every country’s poverty reduction strategy.

The obstacles to greater access in Africa are many and complicated ranging fromdistorted pricing to archaic regulatory systems. With the advent of the second phase of the World Summit on the Information society the time is ripe to address these issueswith a comprehensive approach. In order for the situation to change meaningfully onthe ground we can not shy away from any aspect of the delivery chain. Be it political,economical or cultural.

The times are indeed opportune for Africa to become part of the global informationsociety. New technologies like wireless broadband (Wi-Max), cellular networks andenergy-efficient computers are appearing on the market and can if deployed properlybring real opportunities to the women, men and children living on the African countryside. The opportunities of internet protocol (IP) based networks for deliveringeverything from text to voice and image are astounding and should not be overlooked.

In our efforts to bring access and connectivity to Africa we have to be bold andcreative. The United Nations ICT Task Force working with the legitimacy of the 191United Nations member stated works hard to promote such creativity. By bringingtogether stakeholders from different fields and backgrounds, by linking businessestogether with academia, governments and civil society through its Working Groups,reaching across the globe, those creative ideas are allowed to flourish.

iv

The “Open Access” workshops organized by the Task Force’s Working Group on theEnabling Environment are a prime example of this effort. The workshops have proven to be successful forums for discussing and developing the concept of Open Access.People from all across the globe have gathered to share experiences, develop conceptsand form partnerships.

As the convenor of the Working Group it is my sincere wish that the thoughtsdeveloped at the workshops get shared with a larger community. The book you nowhave before you is an attempt to enrich the debate about ICT for development in thehope of forming greater consensus about appropriate action to bring Africa into thefold of an inclusive information society.

Astrid DufborgSwedish Ambassador Special ICT Adviser

Member of the United Nations ICT Task Force

1

Introduction

“Open Access for Africa” is the second publication of the United Nations ICT TaskForce Working Group on the Enabling Environment. The purpose of this book is toraise awareness among stakeholders in the development field of the opportunities andpossibilities that exist in bringing access and connectivity to the African continent.

The Working Group focuses on issues concerning low-cost access and connectivity. Inmany developing countries, and especially in Africa, national and internationalconnectivity is in short supply: optical fibre may not be available; satellite links arelimited and expensive; internal telecommunication infrastructures are typicallyconcentrated in a few main cities and present severe shortcomings in the rural areas.These technical problems, together with unclear telecom policies and regulations and an internal market that is often closed to competition, lead to lack of investment and highly priced services, thereby hindering penetration of communication services. Theseelements are interconnected and together form a "vicious circle". The objective of theWorking Group is to facilitate a productive dialogue among potential partners to tackle issues concerning low cost technical and business solutions for deployment of ICT indeveloping countries.

Apart from engaging in electronic discussions, members of the Working Group havecome together to organize a workshop series entitled “Open Access”. To this date three workshops have been held. The purpose of the workshops is to identify successfulsolutions and standards for access networks as well as new value chains, opening up for new actors and business models making connectivity and first mile access moreaffordable and available.

The articles in this book build on the third workshop in the series, organized by theUnited Nations Information and Communication Technologies Task Force WorkingGroup on the Enabling Environment in partnership with IDRC, Sida, the EduardoMondlane University and the Royal Institute of Technology in Stockholm, which tookplace in Maputo in May 2005 (www.openaccess.uem.mz). The main themes of theworkshop were open access solutions, a pan-African fibre backbone, financing modelsand regulatory environment for Open Access technologies. The goal was to share views and ideas on how to provide open access, including identifying the users, the relevanttechnologies and the regulatory framework, and how to support the entrepreneurshipthat is necessary to build sustainable networks in Sub-Saharan Africa. After thesuccessful conclusion of the workshop several members of the Working Groupexpressed an interest in capturing the essence of the debates in a format that could beshared with others.

“Open Access for Africa” includes contributions from presenters at the workshop aswell as other authors with interesting perspectives on the issues. The different authors

2 | Open Access for Africa

represent a variety of stakeholders, including governments, academia, NGOs and theprivate sector.

The book examines different aspects of how to provide increased availability of ICTinfrastructure and services for Africa, including identifying the appropriate actors, therelevant technologies, and suggestions for how to appropriately reform regulatoryframeworks as well as how to promote African entrepreneurship in the field of ICTinfrastructure and services.

The first section of the book attempts to grab the reader’s attention through a numberof shorter articles written in a provocative style and discussing the main problems ofAfrican telecom markets as well as how the telecom market has changed, and continues to change, over the past two decades with the further introduction of the Internet and IP networks. The section will, however, begin with an attempt to introduce the conceptof an “Open Access” model as well as giving a brief report on the three “Open Access” workshops that have been held until this date and a brief description of the workshopplanned for 2006.

In the second section the reader will come across examples of current initiatives in thefield of providing low-cost access and connectivity in Africa as well as supporting those actors that are important for a sustainable deployment of modern ICT. The articles will discuss success criteria, challenges and recommendations to others.

In the third and last section of the book the reader will encounter several articlesdescribing different models for providing low-cost access and connectivity to Africathrough different means such as through optic fibre deployment, wireless broadband,mobile telephony and satellite communication. The articles will deal with the challenges in achieving this goal but most importantly they will show the opportunities that exist if key actors can come together in collaboration.

It is the hope of the Working Group that this book will spur discussion about howAfrica on a grand scale can deploy information and communication technologies inorder to improve the livelihood of its inhabitants as well as provide inspiration and ideas for all those who believe in the power of ICT.

Samuel DanofskyUnited Nations ICT Task Force Secretariat

Section IChallenges and Concepts

5

Open Communication – Open Access

Björn Pehrson

1. Introduction

A communication infrastructure is as essential for the development of society astransport, power, water and sanitation infrastructures. Although this fact is widelyacknowledged, the policies and regulations controlling access to the availablecommunication resources are not always designed accordingly.

Communication systems require passive communication media, such as optical fibre orcopper wire or radio spectrum for wireless communication, to establish communication links, and networks providing services to user applications. The passive medium, links,networks and services define different layers of communication resources making upthe communication system architectures.

Open Communication, or Open Access to communication resources, means that thereare openly available technical specifications of interfaces between the different layersrequired to connect to and use the resources as well as open business models indicating under what conditions the resources are accessible. Open access requires an openregulatory regime allowing anyone to access communication resources at one level andprovide added value services at a higher level. Scarce resources, like radio spectrumneeds some sort of arbitration, while essential resources requiring large investments and owned by dominating actors needs to be opened up for others to access. The following discussion will mainly focus on access networks, backbone networks and trafficexchange points, involving different actors and different open access issues.

Traditional telecommunication regimes are not open. They require different kinds oflicenses or permits to get right of way, to communicate over administrative boundariesand borders and to provide different kinds of services. Closed regulatory regimes set up barriers to new market entrants and protect incumbent operators from competition,which leads to inefficiencies and expensive services. Due to the importance ofcommunication to the development of society, policy makers need to legislate andregulators need to define the detailed procedures and enforce them.

The communication market is defined by three types of actors: users on the consumerside, service providers and operators on the producer side while policy makers andregulators are observers and define the rules of the market. The consumers are oftenless organised than the producers and need support from the regulators or user agentsof various sorts.


Taking Sweden as an example, several major landmarks in the development towardsopen access regimes were made in the 1980s due to the establishment of academicnetworks extending the Internet and in the early 1990s during Swedish deregulation and decisions made by municipalities and the apartment housing industry to take control of the passive communication infrastructure on their premises.

• The housing industry, in cooperation with the tenants association, claimed ownership of the passive infrastructure in their buildings, outsourced the operation to neutral communication operators giving their tenants a choice of service providers independently of each other [NSRC].

• The City of Stockholm formed Stokab (www.stokab.se) to deploy, manage and lease dark fibre in the Stockholm region. Open to anyone that wanted to become a service provider, public or private. This development introduced new actors, new value chains and required new technical solutions. The new models have been studied and replicated all over the world [Stokab]

From the research side, the Swedish Royal Institute of Technology (KTH) followed and contributed to this development in cooperation with Stokab and the housing industryfrom the start and eventually applied parts of the models in projects in developingcountries. Two successful first projects in Mozambique 2002 led to an expandedprogram supported by the Swedish international development cooperation agency (Sida) with several projects in Africa, Asia and Latin America.

In 2003, the annual international workshop Series on Open Access was started,sponsored by Sida and the United Nations ICT Task Force, to promote the concept of Open Communication regimes [OC], and to review best practices implementing those[OAWS]. We will briefly present the outcomes of these workshops.

2. The first Open Access workshop, Stockholm, June 2003

The purpose of the first workshop was to identify successful solutions and standards for access networks promoting open access (first mile), as well as sustainable businessmodels for operation and maintenance of such networks that could be disseminated asbest practices.

The workshop presented and discussed successful examples of communicationapplications in developing countries and rural areas with an emphasis on schools andsmall businesses and their role as primary adopters in communities. The focus was onregional prerequisites, communication needs, and choice of technology, businessmodels, competence for operation and maintenance and other factors that affect thesustainability.

Universal access to Information and Communication Technologies (ICTs) has beenidentified as an important part of the efforts to improve the living conditions in theworld. It is consequently part of the United Nations’ Millennium Development Goalsand the Roadmap to reach them [MDG]. As the goals are formulated, it is evident that Education, Healthcare and support for rural entrepreneurs are areas to focus on.Connected schools can also serve as telecenters and healthcare centres/hospitals and do

Open Communication – Open Access | 7

therefore seem to be central projects to be supported, besides education and training. In the end, to get sustainable solutions, the implementation work has to be done by locally rooted entrepreneurs, millions of them on a global scale. An important question is: how can these entrepreneurs be identified, stimulated and supported? Among other things,they need regulatory frameworks allowing them to do what they need to do; they needaccess to appropriate technologies, and education and training in technology, economics and business administration, as well as access to funding.

The workshop program included sessions on:

2.1 Network topologies and link technologiesThe challenges discussed have a common denominator: to transport IP-packets,containing e-mail, web-information, voice, video, or whatever payload, between a localaccess point and the Internet, using whatever link technology.

The concepts of Open Communication and Open Access were introduced anddiscussed and an overview of link level technologies illustrating how the differentcharacteristics of the various technologies can be exploited to solve different problemsto establish a link for providing access to the Internet as exemplified in many of thepresentations at the workshop.

In areas where there is no telecommunication infrastructure, in terms of backbonenetwork, physical transport between servers can be used. Two such cases werepresented: Daknet in India, where buses on a regular route are used as a “mailman”picking up and delivering email via wireless communication between a server on the busand a server at the bus stop, and Samenet, aimed at even more infrastructure-less areas, serving reindeer herders in the north of Sweden where there are no roads, let alonetelecommunication networks.

2.2 Remote and sparsely populated areasThe population density and distribution are important parameters when deciding howto provide affordable universal access, both from technical and business aspects. Theworkshop started out discussing sparsely-populated areas. If the end points for a radiolink are beyond the horizon from each other, even if masts are erected, it is not possible to establish a line of sight. To reach beyond the horizon, if radio is to be used, youeither need to use a HF range frequency, which is reflected in the ionosphere, or to govia a satellite transponder. HF-range systems are cheap but can provide low capacityonly (1-2 kbps) while satellite links are high capacity but expensive (up to 34 Mbps pertransponder at approximately 1 MSEK/Mbps/year).

UHF/VHF requires more or less line of sight, It can sometimes be used over distancesup to 50 km and provide a capacity up to 16-20 kbps. If you can have relay stations with line of sight between them you can use low cost WLAN/Wi-Fi or Wi-MAX links (up to 50 Mbps).

• Wino Bushlink in the Songea district in south Tanzania, sparsely populated and at a large distance from any Internet PoP, well beyond the horizon. There is no wired infrastructure and no line of sight. There are too few users to afford a satellite link.


The solution at hand is using short wave (HF) communication exploiting the ionosphere to reflect signals back to earth beyond the horizon. It is relatively cheap but unfortunately also a very low capacity, up to about 2kbps.

• A telemedicine network in Peruvian Amazons with more or less line of sight over larger distances (50km). The link level technology chosen to transport IP-packets is UHF/VHF achieving up to 17 kbps.

• Jhai Remote Village project, presented by Vorasone Dengkayaphitchith, needed to connect a few remote villages in a valley shadowed by high ridges about 1-2 km away. The closest Internet point of presence is a hospital 5 km away on the other side of the ridge from which the village server could get a dial up connection. Since the territory is under the control of the villages, the solution chosen was to place an 11 Mbit/s Wi-Fi relay station on the top of the ridge with line of sight in each direction.

2.3 Densely Populated areas (Bangladesh/India)In densely populated areas, it is easier to find sustainable business models for providing a network infrastructure, including a fibre backbone, distribution and access networksand wireless access points. The challenges to provide affordable access includepolicy/regulatory issues allowing the use of open, shared network topologies. Caseswere presented from• Bangladesh, one of the most densely populated areas in the world. Recent

developments on the regulatory side and public e-Government services were discussed.

• India, also densely populated and with a large population living in 600.000 villages. The presentation described the government program to increase the penetration of telephone and Internet services.

2.4 Open Access Network case studiesThe purpose of this session was to illustrate how the open communication conceptcould be implemented in practice, by presenting successful pilots:• The Stokab model was presented illustrating the roles that municipalities and public

administration can take to stimulate the communication market by investing in passive infrastructure to lower the threshold for new actors, and at the same time get income, save costs, stimulate regional development and increase the quality of life of the populations.

• The SwedenOpen.net project provided a concrete model of a cheap, open, neutrally operated, city network in Stockholm that could be generalized and adopted into other areas. The model is based on a neutral cost-driven foundation, setting rules for sharing infrastructure and supervision to ensure that users comply with these rules. A system similar to road societies enables an organic growth of the network based on end user demand.

• Borderlight AB presented the business models of a company that makes its living from building operator neutral networks.

2.5 Local Internet Exchange PointsInternet exchange points (IXPs) are needed to allow local Internet Service Providerspeering in order to exchange traffic locally rather than via choked uplinks. The benefits include lower delay on local traffic, diminished loads and possibly cost saving on


uplinks. The challenge is to make fierce competitors cooperate. Many ISPs ratherexchanges traffic abroad than speak to a competitor. A neutral trusted go-betweenorganisation is useful.Three IXP-projects were presented:• Mozix, the Mozambique Internet Exchange Point gave an insight of the

importance of Internet Exchange Points (IXP) and how to build them describing the technology as well as reflecting on the driving forces behind their development, the economical, organization, politics and operations behind the IXP. He also reflected on the experience of the development of the first Internet Exchange in Mozambique, called MozIX (www.mozix.org.mz/) that was designed and implemented in 2002 by KTH and UEM students under a sponsorship of Sida and DFID.

• Tixp, the Tanzania Internet Exchange Point (TIXP), an initiative resulting in the establishment of the Tanzania ISP Association (TISPA) in 2002 (www.tix.or.tz/). The experience is that the process of agreeing among competitors can take time.

• Laonix, the Lao National Internet Exchange implemented 2004 supported by Sida, using the same approach as when implementing Mozix.

2.6 Empowering the disadvantagedEmpowerment of disadvantaged persons, particularly women, is an important item onthe global agenda. Another item is the problem of digital divide. • ICT and Gender Empowerment, an InfoDev/World Bank sponsored pilot Project

SITA (Studies in Information Technology Applications: A computer-skill training program for socially disabled women). The project has developed a multimedia-based strategy for training individuals with inadequate educational background and limited communication skills. It explores if ICT can play the role of a catalyst for empowering needy women.

• The Kista Divide, an ethnographic study of the Kista region, based on interviews of its dwellers, the idea of the network society and its problems is further explored.By describing the area as divided between those "connected" and those "off-line",between "haves" and "have-nots", it illustrates that the digital divide is present also in developed countries.

2.7 Connecting SchoolsSchools play a central role in any scheme for providing open access to the public. Thenew generations are educated there and they can function as telecenters for parents and the public after hours, providing a valuable community service and source of income to the school. At the workshop, two African projects were presented together with World Computer Exchange (WCE) that provides refurbished computers to schools.

• Uconnect provides Uganda primary and secondary schools with computers. Network training workshops are provided for selected teachers and students at education ministry headquarters. The Internet connection is by satellite, distributed to schools via dialup, both land line and GSM, fibre optic connections and broadband wireless, 2.4 and 3.5 GHz microwave.

• SchoolNet Namibia has adapted Wi-Fi wireless telecommunications standards and open-source software to provide affordable Internet access. SchoolNet Namibia


targets policy-makers, to make them aware of the opportunities and challenges involved in rural access, and reinforce the public sector efforts via a national alliance of local and international stakeholders. The alliance is expected to provide an enabling environment to expand the availability of ICT technologies in the health, agricultural, education and civil society, develop strategies for overcoming barriers in a policy environment that should support open entry and innovative approaches to wireless access and development of appropriate wireless telecommunications and alternative energy solutions and make provision for the implementation of such solutions in Namibia’s rural areas.

2.8 ConclusionsIt was generally agreed that there is an abundance of opportunities for localentrepreneurs to provide Internet access in developing countries but that the challenges still are large. It was recommended on how to support such entrepreneurs be the focus of the 2004 Workshop.

3. The second Workshop Stockholm 2004

The 2004 workshop pursued the issues of universal access and open networks,identifying and involving new relevant actors in the discussions and drafting of openconnectivity programs for developing countries. The workshop set out to identifysuccessful solutions and standards for access networks and as well as methods toempower new actors, business models and value chains critical for the development ofsuch infrastructures and services in cases where traditional models do not work. Whenthe traditional models fail to make the last mile, local entrepreneurs and communitiesshould be able to go the first mile to connect to a point of presence of serviceproviders.

A main purpose of the second workshop was still to build local skills and knowledgecapacity by inviting, to the workshop, representatives from developing countries - toparticipate in the discussions, to learn and to bring in their information on the status of infrastructures and local connectivity in their particular environments.

Another purpose of the workshop was to initiate cooperation with nationalorganizations in the least developed countries and in international organizations todisseminate knowledge about and promote deployment of open access networkssupporting the WSIS (Geneva Dec 2003) declaration "Information and communication infrastructure - an essential foundation for the Information Society" [www.itu.int/wsis]. In connection with the 2004 workshop in Stockholm, a discussion on strategies forpromoting an African fibre infrastructure was organised with African stakeholders,infoDev and Sida resulting in infoDev sponsorship of an in depth study [infoDev].Already in the study phase this work has had an impact on the discussion about theemerging fibre infrastructure in Africa and the demands for open access to it. Thepolicies in some of the involved countries have turned out to be surprisingly openalthough the associated regulatory frameworks are still undeveloped. In other countries in Africa and in South East Asia the policies are not as progressive or independentregulators.


3.1 Models and processesThe first session was designed to set the scope of the workshop and raise the issues tobe discussed in presentations and breakout sessions. This introduction included thefollowing presentations:• Towards an access for all, providing a survey of the work within the United

Nations system towards the Millennium Development Goals, focusing on the United Nations ICT Task Force and the World Summit on the Information Society (WSIS) process.

• Implementation models involving all stakeholders based on local entrepreneurs and their need for support systems, such as access to technology and education, regulatory frameworks, financial support.

3.2 Education, Healthcare and Emergency responseThe United Nations ICT Task Force’s Global e-School and Communities Initiative(GeSCI) was presented and discussed, including activities in Andhra Pradesh, Bolivia,Ghana and Namibia.

Support for healthcare is one of the prioritised areas working to achieve the MillenniumDevelopment Goals. Grameen Communication provided an overview of ICTdevelopment projects with involvement from the Grameen group in Bangladesh. Theprerequisites for healthcare in rural areas in Bangladesh were discussed. EricssonResponse, a division within Ericsson providing emergency communication services onthe request from the United Nations.

3.3 Rural entrepreneurship and local marketsLocal entrepreneurs from Laos, Rwanda and Tanzania were invited to witness about the environment they work in and express their needs for support.

Three presentations were selected to initiate a discussion on support systems for localentrepreneurs on financial services, regulatory challenges in developing countries andleapfrogging opportunities in the telecoms' markets of developing countries.

A few examples of organizations supporting developments projects in the ICT area,including rural access for various purposes, were presented, including SPIDER, aSwedish resource centre for ICT in developing countries at KTH supported by Sida, the development project award and database Stockholm Challenge now owned by KTHwith Stockholm, Sida and Ericsson as partners, the Multilateral Initiative on Malaria,MIM, and its communications arm, MIMCom, and the World Computer Exchange.

3.4 ConclusionsThere was a strong support for the model to use the communication needs of thehealthcare system, educational organisations and public administration as the basis when building a sustainable communication market, to leave the service provisioning to local entrepreneurs and to give the public sector a role providing a passive infrastructure for telecommunications, in analogy with roads, railways, electrical power lines, to get a more dynamic development of society.


In connection with the workshop, a special meeting on Open Access and Backbonenetworks, with a special emphasis on a fibre infrastructure for Africa was organised with representatives from infoDev, AfrISPA, Universidad Eduardo Mondlane (UEM) inMozambique, Rwanda IT Authority, Juasun, a ISP active in rural Tanzania, KTH andSida. It was proposed to use the idea of an open infrastructure for Africa as a case topush for the second phase of the WSIS in Tunis. InfoDev took on the task tocommission an investigation and report proposing an action plan for regulatory changes while UEM and KTH decided to push the regulatory framework using academicnetworking as a spearhead, putting the fibre link Maputo-Johannesburg, owned by thepower utility companies in South Africa and Mozambique, that could not be used dueto regulatory reasons in the spotlight.

4. The third Workshop, Maputo 2005

The 2005 workshop was organized by Universidad Eduardo Mondlane in Maputo,Mozambique, in May 2005. The focus was on a continued discussion about the VSATprocurement consortium earlier suggested, surveying the availability of fibre and thenew discussion how to use academic networking as a spearhead in the process toestablish a Pan African fibre Backbone.

Other issues also discussed at the workshop included the regulatory environment forOpen Access technologies in the context of developing regions, in particular the Sub-Saharan Africa, as well as financing models and education of entrepreneurs.

4.1 Feasibility of a Pan-African fibre infrastructureDifferent perspectives of the feasibility of a pan African fibre infrastructure wereprovided by international experts as well as by African fibre owners, not only telecomoperators but also alternative fibre, such as those provided by power utility companies. IDRC and NEPAD have supported a survey of the availability of fibre also alongrailways, pipelines, etc. It turns out that there is a lot of fibre deployed and more in the process of being deployed as part of other infrastructure projects.

4.2 Using African Academic Networking as Spearhead In the short term perspective, it was argued that satellite may be the only technology for providing Internet access to most of Africa, but land-based optical fibre infrastructureswas the only long term solution. For the time being, both tracks should be pursued.

SatelliteA number of donor organisations have cooperated with African universities and theirorganisations to organize a procurement consortium to press prices on satellitebandwidth [ATICS, PAREN]. The African Virtual University (AVU) is active in thisarea and AAU, the Association of African Universities, discussed the plan at its annualmeeting in Cape town in February 2005. At the same meeting, SARUA, the SouthernAfrica Regional University Association was formed. SARUA was presented at theworkshop and has identified access to ICT as their highest prioritized agenda item.The already operational MALICO VSAT consortium providing connectivity foruniversities in Malawi was presented and a study of the feasibility to set up a dedicatedVSAT hub connected to Géant discussed.


All the studies were presented and discussed.

A representative of the Global VSAT Forum discussed the regulatory issues concerning access in each country.

FibreAt the AAU meeting in Cape Town, a parallel track to the VSAT consortium wasdiscussed, suggesting fibre-based National Research and Education Networks (NRENs) connected to a regional fibre backbone that could provide Internet connectivity as well as peering between universities in Africa and via connections to academic backbones on other continents [AAU1].

At this workshop the availability of fibre owned by different actors, such as powerutilities, railways, pipelines, etc and the feasibility of a regional backbone was discussed.

4.3 Open Access Projects and TechnologiesRural telecommunications under the open access paradigm was discussed fromregulatory, technical and economical perspectives.

New emerging services, such as Voice and Video over IP were also discussed and a pilot project presented and demonstrated.

4.4 ConclusionsThere was a broad consensus that procurement consortia for pressing prices for VSAT capacity is the right way to go and that such consortia also could be used to lobby formore open communication policies and regulations and to establish fibre-based national research and education networks. As the result of discussions alongside the workshop,the planning of projects for the establishment of NRENs in Malawi and in Mozambique with trans-border connections to form a regional backbone started. There were alsodiscussions about how to implement a regional fibre backbone.

5. The fourth Workshop, Stockholm May 2006

The 2006 workshop will be organized by KTH in Stockholm in conjunction with theStockholm Challenge Award event, May 2006. The plan is to identify suitable projectsamong the finalists for the Stockholm Challenge Award that could be taken to the next level by exploiting open access methods supported by the organizational and humannetworks constituting the Open Access workshop community.


6. References

[AAU1] Björn Pehrson: KTH, Broadband Internet Access for African universities, Aparallel track to coordinated VSAT procurement, Presentation at the AAU AnnualMeeting in Cape Town, South Africa, 2005-02-23

[AAU2] Conference on African Research and Education Networking Infrastructure,Tunis, 14-15 November, 2005 (www.aau.org/tunis/)

[AREN] International Workshop on African Research and Education Networking,CERN, Geneva 25-27 September, 2005 (http://event-africa-networking.web.cern.ch/event-africa-networking/program.html)

[ATICS] Roy Steiner at al, African Tertiary Institution Connectivity Study,Commissioned by the World Bank Institute

[CSD]Björn Pehrson: Problem-oriented capacity-building based on the introduction of Information and Communication Technologies as an Enabler of Socio-EconomicDevelopment

[InfoDev] Anders Comstedt, Eric Osiakwan, Russell Southwood: Open Access Models, infoDev, Final Draft, September 2005 (www.infodev.org)

[MDG] UN Millennium Development Goals www.un.org/milleniumsgoals/ and Roadmap towards the implementation of the UN Millennium Declaration, Report of theSecretary-General 2001-09-06http://ods-dds-ny.un.org/doc/UNDOC/GEN/N01/526/07/PDF/N0152607.pdf?OpenElement

[NSRC] Business Development and ICT usage in Homes and Civil Society, NetworkSociety Research Centre Seminar (www.nsrc.se/seminar-2005-02-15.html)

[OAWS] United Nations ICT Task Force, Sida and KTH International WorkshopSeries on Open Access, Stockholm, June 2003 and May 2004, Maputo, May 2005(www.wideopenaccess.net)

[OC] Björn Pehrson: Open Communication, 1st International Workshop on OpenAccess, Stockholm, June 2003, (www.wideopenaccess.net)

[PAREN] Roy Steiner at al, Promoting African Research and Education Networking,Commissioned by IDRC and Connectivity Africa (www.connectivityafrica.org)

[SCA] Stockholm Challenge Award (www.stockholmchallenge.se)

[Stokab] Anders Comstedt: The Stokab Model for Fibre Deployment, 1st InternationalWorkshop on Open Access, Stockholm, June 2003 (www.wideopenaccess.net)


[VSAT1] Michael Jensen et al: Open and Closed Skies: Satellite Access in Africa, Policy Reform and Regulatory Issues in Bridging the Digital Divide through SatelliteTechnologies

[VSAT2] Olavi Trydhed: African Academic Backbone - Satellite Hub in Europe, March 2005, Commissioned by Sida

16

Networking for Africa – Open Access and Other Issues

Rahul Tongia

One of the major debates raging in the run-up to WSIS Phase II is the issue of Internet Governance. Clearly, current Internet Governance systems can be enhanced to include greater participation, not only from around the world but also from civil society.However, if one critically examines the poor penetration of the Internet in manydeveloping regions such as Africa (let alone broadband penetration), one realizes thatmany of the reasons for the low penetration are because of domestic policies or ground realities of developing countries (such as small markets), instead of an inherent biaspushed by the developed country elites.

If one carefully considers a major challenge for developing countries, namelyinternational connectivity, there are certainly some concerns about how dataconnectivity is priced (with developing countries paying for traffic in both directions, in contrast to telephony whereby the current system allows them to earn revenues forcompleting calls). However, much of the high costs are due to the largely monopolistic behaviour of the entities setting up international gateways and fibre links. This isespecially the case for many African countries who may have an international fibrelanding, but this has done precious little for making connectivity affordable for thepopulation at large. At the Mozambique Workshop on Open Access, one participantpointed out that SAT3 was one of the most expensive fibres in the world. Despitebeing utilized to only a fraction of its capacity, it was profitable in only around threeyears.

The new fibre consortium for East Africa (EASSy) might become a similarly under-effective infrastructure unless we ensure there are steps towards Open Access and away from the “closed club” structure seen elsewhere. Else, we will have incumbents orselect providers with if not an official monopoly then a de-facto monopoly, andinternational connectivity will remain prohibitively expensive.

Of course, international connectivity is a bogey cited by many groups to rationalize high connectivity costs. Truth be told, if one had a viable local demand, with local content, one could leverage satellite connectivity to increase affordable penetration, at least in the interim. Land-based fibres can be significantly less expensive than oceanic, soleveraging existing oceanic fibres through cross-border terrestrial fibre systems wouldbe a much less expensive proposition. This would also have the benefit of being routed through or near many population centres, instead of being a fast pipe on the beach.(The FiberAfrica proposal in this volume presents such a model).

Networking for Africa – Open Access and other Issues | 17

Much of the billions of dollars of annual investment in telecom over the last few yearshave been for mobile telephony, especially by private participants. But this has beencomplementary to meaningful data connectivity efforts, 3G systems and their claimsnotwithstanding. Throughout discussions at Mozambique and elsewhere, there was a(near) universal consensus that the challenges in bringing data connectivity to Africahave less to do with technology than with market and regulatory structure.

This relates to a deeper and more fundamental discussion on how telecoms and dataconnectivity are designed and spread. In the west, this debate has been called “net-heads vs. bell-heads.” The Internet-centric vision is one of decentralized control, where applications run from the edges, and voice can itself be thought of as just another (albeit important) application. The telephony-centric view, favoured by incumbents (who inthe US were called the Bell Operating Companies), is more hierarchical and involvesmore centralized control.1 The question remains if the Internet is different or special,and can it be treated differently from voice telephony (or should it?). Equallyimportantly, given that the underlying infrastructure can be shared (a fibre running a bit is a fibre running a bit – voice bit or data bit look the same), how can we ensure thatdominance in one domain does not lead to dominance in another?

The ITU’s NGN (next generation network) initiative is important as it implicitly brings governmental (and thus regulatory) support at some level. Like the net-head view, itassumes a packet-switched world, but it adds complexity and control through carrier-centric control with the purported aim of ensuring quality of service and security.However, such features, including regulations or frameworks for inter-carriersettlements, might inadvertently support the larger players and incumbents. A trulyopen infrastructure, beyond just open access, leaves many such issues to the end-pointsor to the different operators who might operate at different layers (such as a fibreinfrastructure provider interacting with a retail operator interacting with a contentprovider). Of course, given that in large parts of Africa, the largest players are not fixed line carriers but mobile operators, this provides an opportunity for the fixed lineoperators to consider new business and business models (such as data services and open access, respectively).

How real are open access models? The answer depends on whom you ask. There aremany municipalities and communities building out such networks around the world, but the incumbents have tried to fight back (in the US, a number of states have enacted or proposed legislation banning those types of networks.) Luckily, these moves are beingrecognized as ones that limit choices or services for consumers, and a number ofregions are supporting for such networks.

However, when it comes to the developing world, it appears there is the familiar issueof “do as I say, not as I do.” Specifically, the push by the West towards privatization

1The history of the role of ITU versus Internet development professionals is long and deep. The ITU had

advocated the OSI layered model as the foundation for data connectivity, but this was trumped by the use of TCP/IP, which did not conform to that “global standard.” Instead, it became the open standard for the Internet of today.


and markets ignores the strong (and potentially positive) role governmental support has had for extending universal service (rural telephony) or new services (such as inSingapore or Korea today). We have to recognize that meaningful penetrations ofaffordable broadband across Africa will not trickle down from market-driven measures.If we do want special entities for data connectivity, ones that receive specials waiversand concessions such as rights of way, customs waivers, etc., these should be public,open access networks, instead of monopolies (either public or private, regulated or de-facto). This would treat basic connectivity infrastructure as a public utility.

This regulatory muddle, present even in developed countries, is one reason we do nothave inexpensive broadband for the masses. Who owns what infrastructure and theservices they can they provide are burdened by cumbersome if not conflictingregulation. This is the case in Africa, where a large amount of optical fibres havealready been drawn by power companies, who either do not use them at all, or run low-speed, inefficient communications over them (largely for internal control andmonitoring uses). They might want to provide telecom services, but they either are not allowed to, or are unsure of how they would be treated by regulators. The solutionwould be to leverage the infrastructure for open access networks, thereby providinglow-cost connectivity without either helping the incumbents gain market power orcreating a new monopoly player. In fact, given that Africa is expanding electrificationinto rural and new areas, all new power transmission systems should include opticalfibre draws. With new designs that embed fibres, the incremental cost can beexceptionally low, only a few hundred dollars per km (or less). This would make most sense when there is regulatory clarity on how to use these assets for the public good.Developing countries must be aware that traditional designs and business models maynot serve them well, and they do not need to follow existing or “approved” paths only.

How does one move ahead in building open access and other disruptive networks?One suggestion to help finesse the policy issues has been to build such a network as aresearch and educational network, perhaps only for non-profit usage. Another optionhas been for a country or region to try building out the network, to see what the costsand benefits are, as a sort of low/medium cost experiment. For that type of effort, aid or grant money could provide the funds. International support would be crucial forsuch developments, which is tricky if they themselves do not advocate open networks.Ultimately, no amount of prodding or good wishes will be enough until the countriesthemselves take action to bring affordable, meaningful connectivity to their masses.

19

A Layered Market Paradigm

Anders Comstedt

Smoking among teenagers in the metro areas of the developed world is down. Onerecognized reason is that they cannot sustain their smoking habits and pay their mobile phone bills at the same time. The choice to quit smoking shows how strong the telecom needs are for these teenagers in order to stay in touch with their community, otherteens.

Is the situation the same in developing countries? Basic voice service is by many stilllooked upon as being best and most cheaply satisfied by an expansion of the old wireline telephone network service. This is an opinion aired even by seasoned peopleworking with development. They will, however, be disappointed if they try toimplement this strategy since it is cheaper to deploy a mobile phone system by a factorof at least three compared to a wire line one! You only have to look at the investments made by new operators and divide it by the number of customers and compare thatnumber to the USD 1,000 normally quoted as the cost for a wire line access. Will a USD 100,000 base station provide you with more than 100 regular users? In anything but the most rural areas it will. Furthermore, the service provider enjoys a lower risk usingprepaid phone cards, handsets provided by the user and no wires to be stolen. With avery limited competition from substitutes, providers in densely populated areas indeveloping countries are very profitable, not only thanks to higher margins than in thedeveloped world, but also because they are generally well managed, much better thanthe old incumbent telephone companies. Does this mean that there will be competition between two or three cellular companies, all having totally self supporting systems,including backbones?

The metro kids of developed countries are not only big users of mobile phones but also of the Internet. The Internet industry with its roots in a more co-operative world ofpeers, and not in the vertically integrated centralised telecom industry world, has growndifferently. There is no doubt that with the help of a different technical architecture,paired with developments in optoelectronics, signal processing, and other enablingtechnologies we now have an Internet Protocol (IP) based technology platform alsoembraced by the entire telecom industry. The reason is that using this new platform ischeaper and more efficient than relying on old analogue technologies.


Behind the front desk, every operator now uses Internet technology, even if it is notvisible to the average consumer. Virtually all vendor development resources are going in this direction. The IP based building blocks are simply cheaper to use. Although newservices, like mobile phones, have been high margin commercial successes, cost isalways an issue. And as every market matures cost will become even more of an issue, as will the question of how to widen the market. How can a mobile phone service provider use its structure to provide other types of Internet like services? Can it easily share some infrastructure resources with other mobile phone providers without diluting its brandname and appearance in the consumer market? May a separate company sell basictransport service using the infrastructure of the mobile phone company? Can thisseparate company even sell upstream connectivity and transport to others than themobile phone companies, such as Internet Service Providers (ISPs)? ISPs that in theirturn sell e-mail, web surfing and now also voice service over the Internet (VoIP). Theanswer to all of these questions is yes.

The very structure of the IP architecture enables this sharing of infrastructure. 2 Thegroups of bits and packets, sent over optical fibres, wireless links or even a barbed wire, may look the same when they are transported, but they will be converted into different things by the multitude of different operators. Instead of having completely differentand separate systems everyone is using an architecture where the same transportmechanism, the transport layer, can send packets of bits representing a telephoneconversation, a web page, an e-mail or a TV show. The content and final format will be decided by the next layers, the service and application layers, a combination of thesoftware and hardware closest to the user. This layer is also run by a multitude ofdifferently skilled companies.

The clear horizontal layering of the IP technology can be viewed as a division of themarket responsibilities and tasks with service companies on top of infrastructurecompanies. Companies with an old vertically integrated technology platform may come to have a cost and flexibility disadvantage, especially in markets where they just lost amonopoly and didn’t realize the change in market paradigm in time. They will have ahard time introducing new services simply because of the rigid structure of theirtechnology, but they also face organisational challenges as the management structurereflects their technical platform. From the outside it is just an overstaffing problem, but inside it is worse than that.

Economies of scale, customer density per area as well as utilisation are all importantfactors for an infrastructure company. The company should therefore strive to have asmany users as possible transporting different packets of bits on its road. Not havingaccess to a shared infrastructure could be fatal for any service provider, no matter if its service concept is the most interesting in the whole market. Duplicating infrastructurewhen you have a small market share is simply not possible. The investment per user will be prohibitive. It would be like building a high-way for your own use.

2To learn more on this see for instance

http://global.mci.com/about/publicpolicy/presentations/newamericalayerspresentation.pdf

A Layered Market Paradigm | 21

The future will most certainly bring about an increased horizontal specialization wheretelecom regulators will have to pay more attention to the lower layers of the systemwhere scale and density matters, and leave the higher levels to authorities dealing withgeneral consumer issues.

The reason for this is that we will have few players on the infrastructure levels, say three to five, maybe even one or two providing the optical fibres or the wireless towers. But we will have dozens of service providers facing the end user in a much more open anddynamic competition. Policy makers need to adopt this perspective too, and swiftlyimplement rules that curb any misuse of a dominant power inside a layer towards upperand lower layers. There is less need for policy makers and regulators to engage inactivities focusing on the relation between the service provider and the consumer, likeprice control, since many of these efforts will prove to be both wasteful and pointless.This is not to say that consumer protection laws are not needed. On the contrary, bydesign, slow sector specific legislation and regulation, today trying to hopelessly micromanage a fast changing, diverse and dynamic market of service operators will be bettertraded for good general consumer protection. The focus for telecom regulators shouldbe on providing consumers with choice by asking themselves how their actions can getfour to five different offers to the user, and then stay out of the way to look at the next lower layer, asking the same question. Ultimately they will find a layer where inherenteconomy of scale prevents several players. This layer, with few players, will needongoing attention and regulation to prevent misuse of power.

Those who think that this horizontal layering is not already happening just need to look at where the mobile phone industry is going. Being the most successful part of thetelecom business in developing countries, it now looks at how to cost effectivelyprovide services to the next billion users worldwide. This means that major systemvendors will become engaged in establishing service companies running the entireproduction for mobile phone service providers. In doing this, they will of course haveeconomy of scale due to running not one but maybe four or five networks, plus two or three also in a neighbouring country. Many of the components in these networks will be shared through the service company. The industry will have become global or at leastregional. It will still be on the look out for who is going to provide basic transport in the backbones though, since this is very much a local factor.

An important question to ask is who these infrastructure providers will be. Will it betransformed old PSTN wire line companies or new entities like power utilities stringing fibre on their pylons? The technology shift towards optical fibre, inherently non-conductive to electricity, opens up an interesting perspective to run optical fibres on the electrical high voltage power pylons and poles out to towns and villages where different technical means, both wire and wireless will be considered for the final access. This iseven more interesting as we are facing significant rural electrification programmes inmany countries. The fact that power and telecommunications reinforce thedevelopment effects inherit in both of these two services provide for an even moreinteresting scenario. Government intervention and donor support is needed to makethese fibres open on non-discriminatory terms just to avoid a situation with limitedaccess in the first place. It is equally important that other structural decisions are not


focused on what the old incumbent telco exclusively likes, but more on what newentrants like wireless ISPs may desire.

Will the kids in developing countries be given a reason to give up smoking as they grow a somewhat bigger income and prices of telecommunication services come down? Andwill they have a choice to get Internet and voice connectivity from someone else but afew dominating mobile phone companies, with limited competition? I hope Africanregulators seize the current opportunity to make it so.

Section IIActors and Examples

25

Nurturing Entrepreneurship in Mozambique

Daniel MannestigConstantino Sotomane

Jamo Macanze

1. Introduction

Small drops of sweat are dripping from the face of Mr. Constantino Sotomane,coordinator of the MICTI implementation team. It has been another tough day at work, and this day has not been different from any other day during the last two years. “But I can see the light at the end of the tunnel now”, confirms Mr. Sotomane. “Progress hasbeen made, and MICTI is growing stronger every day”. “We recently became OracleAcademic Initiative partner and we are giving courses for 20 people, being busy from08:00 to 20:00”. “More courses will come in November and December, and theincubator is developing and will soon graduate successful companies”.Entrepreneurship is the driving force for initiating business ideas and for mobilizinghuman, financial and physical resources, and the attitude and capacity to innovate andtake initiative. This article looks into entrepreneurship in Mozambique and finds outmore about the MICTI initiative that step-by-step is making a difference in the country.

2. Mozambique Today

Mozambique is a country in transition endeavouring to consolidate peace, democracy,and economic growth and development after more than 25 years of armed conflict;namely the liberation struggle between 1964 and 1974 and a devastating civil warbetween 1976 and 1992. In 1990 a new Constitution was adopted introducing, amongothers; a free market economic system, a multi-party political system, the abolition ofthe death penalty, and freedom of the press and of association.

Mozambique is a poor country with almost 70% of the population living in a state ofabsolute poverty. In addition, almost 50% of the adult population is illiterate, just under 60% of children in the age group 6-12 years are out of school, and only 1% of studentsenter Higher Education Institutions. Coupled with few technical/vocational institutions, these realities entail tremendous challenges for long-term empowerment, skillenhancement and employment opportunities for the individual and, sustainabledevelopment for the country. The country’s development needs are evident from its168th ranking – out of 174 countries – as reported by the UNDP’s 2000 HumanDevelopment Report.


Over recent years Mozambique’s economy has exhibited impressive growth – albeitfrom a low base – mainly driven by substantial infrastructure ‘mega-projects’. Theeconomy is in the early stages of industrialization, with trade and services (42%) theprimary contributor to GDP, followed by agriculture (32.1%) and manufacturing adistant third (12.3%).

The Government of Mozambique’s primary objective is to “reduce the incidence ofabsolute poverty by about 30% over the next ten years”. To this end, the country’sPoverty Reduction Strategy (PARPA) is dependent on the private sector to generateeconomic growth with the public sector fulfilling the country’s development andpoverty reduction objectives. PARPA has evolved after extensive and transparentconsultations between the Government, the private sector and civil society, and hasbeen approved by the World Bank and the IMF. Policy initiatives and prioritization aredriven by their underlying contribution to PARPA and its objectives.

Mozambique is characterized by a relatively small private sector. Manufacturing andfishing contribute to more than 90% of the industrial sector (manufacturing isdominated by a few companies and fishing by prawns). Various institutional andfinancial constraints – bureaucratic environment, lack of access to finance and highcosts of borrowing – hinder the development of the private sector, in particular SMEs.For example it takes on average, 3 to 4 months to register a company in Mozambiqueagainst for example 2 days in Australia. Furthermore, the cost of registration amounts to 116% of GDP per capita in Mozambique against for instance 1.4% in Canada.

The Mozambican ICT sector, including infrastructure and human resources is small, but experiencing rapid growth. In terms of Internet access, the number of Internet ServiceProviders (ISPs) has risen from one in 1995 to ten in 2001, with approximately 50,000subscribers in 2002. Exposure to the Internet is likely to be higher than the numbersindicate, as there are normally several users per subscriber. But even so, the numbers are still relatively small.

Looking into some selected Higher Education and ICT Characteristics in Mozambique:• Only 1% of students enter Higher Education Institutions (HEIs), less than 9% of

enrolments graduate. Less than 1% of the population has information and technology skills or experience.

• There are three private HEIs and one public (University of Eduardo Mondlane) offering ICT courses. Estimates are that a total of 30 to 40 ICT graduates are produced per year. Figures are not available for technicians, but the low number of vocational institutions indicates very little, if any, expertise at this level.

• ICT in the private sector and parastatals is dominated by a small number of organisations. The major users are the banks, utilities (Water, Telecom, andElectricity) and Mozal. Demand for ICT expertise and skills are greater than the supply. There are few research facilities in the private sector.

• Apart from basic software development (accounting and resource management), there is very little software expertise and capabilities.

• No telecommunications equipment is produced in Mozambique, but most major suppliers operate commercially in the country.

Nurturing Entrepreneurship in Mozambique | 27

• There is little ICT capacity in the public sector with manual procedures and processes being the norm. Substantial investments – process design, systems, hardware, and skills, among others – are necessary to facilitate efficient, effective and transparent services.

3. MICTI – a Vision for an ICT-literate Society

On the 12th December 2000, following a two-year nationwide debate that involved allstakeholders – public and private sectors, civil society, NGOs and developmentagencies – the Government of Mozambique approved an Information andCommunication Technology (ICT) Policy. Its key objective is to establish theparameters that will lead Mozambique into the Information Society, thus promoting the general use of information and communication technologies as a development orenabling tool.

The Mozambique Information and Communication Technology Institute (MICTI)under the leadership of Eduardo Mondlane University and the Ministry of Science andTechnology, is a priority program and plan of action for the Country’s ICTImplementation Strategy.

MICTI is the brain-child of Prof. Dr. Venâncio Massingue, Minister of Science andTechnology and former vice-rector of Eduardo Mondlane University, a futurist praising entrepreneurship and new ideas. Prof. Dr. Massingue has both the vision and thenecessary drive to put this mega-project on the right way to success, and he would like to change the minds of the Mozambican University student. Their first choice is a“safe” job in one of the Ministries. Why do they not dream of starting their owncompanies, securing their own salary and to employ others?

MICTI is a long-term vision for building ICT skills and capacity in Mozambiquefacilitated by combining the energies and contributions of the Government, Academia,domestic and international private sector, as well as development and donor agencies.Building an ICT–literate society is not a short-term endeavour. The Mozambique ICTInstitute is at the core of a ten to fifteen year vision to build ICT skills to contribute tothe growth of income generation opportunities and employment and to create a core of locally trained ICT experts ready to apply their knowledge to real world problems inMozambique.

The core objectives are:• Contributing to fight against poverty through expanding citizen’s access to global

knowledge resources and attracting foreign investment and new partnerships into Mozambique.

• Creating in Mozambique an anchor project in ICT related fields that reinforce the linking of the country into the Global Information Society.

• Acting as an important implementation project of the Country’s economic strategy.• Encouraging and supporting start-up, incubation and development of innovation-

led, high growth, knowledge-based enterprises.


• Providing an environment where larger and international businesses can develop specific and close interactions with a particular centre of knowledge creation for their mutual benefit.

• To have formal and operational links with global centres of knowledge creation such as universities, higher education institutes and research organizations.

The Program is made up of three inter-related components:The Research and Learning component address the direct need for skilled personnel inMozambique and the region. It provides quality structured learning and relevant work-based experience in the related fields of ICT, Management and Systems Processes.Access to global expertise is sourced through sabbaticals and exchange programs withinternational centres of excellence.

The Incubator component nurture entrepreneurial skills providing employment and wealthgenerating opportunities for the Country’s student and academic community. Stronglinks is forged with the private and public sector facilitating an alignment of appropriate action program and Mozambique’s industrial and development needs.

The Science and Technology Park component complement the Research and Learning, as well as the Incubator components. Participating international and domestic organizationsprovide expert input to the Research and Learning component, seed and nurtureMozambican knowledge and innovation capacities by the utilization of the Incubatorcomponent and benefit from the skilled and trained human resources which haveprogressed through the first two program components. Organizations are able toestablish formal and operational links with the University of Eduardo Mondlane, as well as all of the Country’s Higher and Further Education institutions (public and private),gaining access to their research, scientific and technological expertise. The emphasis isto source technology and industry sector organizations relevant to Mozambique’s needs and opportunities.

Fig 1. The MICTI concept


To this end the Government has provided three hundred and sixty five hectares ofpristine land located outside of Maputo in Moamba, Southern Mozambique, earmarkedfor the development of the MICTI campus. The area will be a ‘special economic zone’, enabling investment environment including incentives and benefits of a ‘specialeconomic zone’ status granted to Moamba.

Fig 2. A computer model of a Moamba complex

Moamba offers opportunity for a “greenfield” development where specific tenantrequirements can be optimized, situated along the Maputo Development Corridor, amajor regional infrastructural development between Mozambique and South Africa.

4. Business and Technology Incubator – Nurturing Entrepreneurship

The vision of the MICTI Business and Technology incubator is to promote economicdevelopment in Mozambique through encouraging research and learning as well as theentrepreneurial spirit in ICT related areas, creating wealth for the tenants and for thecountry as a whole. The incubator has created a dynamic environment from whichparticipants are able to engage with academic and research endeavours, whilesimultaneously contributing to their own and the country’s economic development,providing space, training and facilities in privileged conditions.

The MICTI Business and Technology Incubator host five small ICT companies at theUniversity Eduardo Mondlane main campus. Pilot operations started in September 2003 after an independent selection panel chose innovative and economically viable ideaswith focus on ICT and with growth potential that will satisfy the Mozambican market,having dedicated team with entrepreneurial spirit and sufficient time available.

“One of the problems the incubated companies are facing is to be dedicated enough todevelop their companies”, Mr. Sotomane says. “How can you dedicate yourself 100% to develop your business idea if your family or people close to your need your support?”“And the African family is big and needs your support”. “People are very flexible andcreative but not always dedicated”. “That is a big challenge for us and ourentrepreneurs”.

The incubator assists small and medium sized ICT businesses, through the provision of high quality facilities and services, to develop into successful and economically viableentities that will contribute to the economic and employment growth of Mozambique”.


The objective of the Incubator is to foster entrepreneurship in the ICT sector inMozambique by:• Promoting the development of creative and innovative ICT based companies • Encouraging the development of ICT based businesses throughout Mozambique • Increasing the rate of success of new and small ICT based businesses • Reducing the business development times of tenant client companies • Encouraging cooperation between and within small and medium sized ICT

businesses throughout the country • Providing tenant clients with access to high quality, secure and reliable physical and

communications infrastructure • Providing value adding services to tenant clients that will minimize their operating

and development costs • Facilitating the development of business and technical know-how

Since the start, the selected companies are:

Tenant Employ. Business Idea Web pageDDJ Law online

2 Juridical services online www.ddjlaw.co.mz

Webcom 3 National search engine and web development

www.webcom.co.mz

Real Soccer 2 Online sport magazine www.olance.co.mzE-Novar 3 Tourism portal of

Mozambiquewww.coco.co.mz

Informóvel 2 “Mobile” education

The incubator provides infrastructure and equipment, business and support services and shared resources. Each tenant has a room with 2 computers connected to the Internet, furniture, air condition, white board etc. Tenants have free access to resources andsupport services such as scanners, data projectors, digital cameras, meeting rooms,library and IT support. On a nominal fee basis, tenants can use telephone, fax and print. In order to help the companies to develop and become sustainable, the incubatorsupports the tenants to develop a business plan, register the company, get in contactwith potential customers, get access to funds and loans, produce marketing material,participate in fairs, seminars and other useful events. The incubator provides training,networking, education, workshops and seminars in management, accounting andfinancial planning, marketing, business planning, strategy, entrepreneurship and othertopics.

“We have managed to get a nice mix of different companies that also can help eachother to develop”, says Mr. Sotomane. “If one company has juridical competence andanother can make home pages, they swap services, a win-win situation”.

The incubator has established an evolutional model that describes how tenantcompanies can develop. The model is divided into four stages; the first is to make abusiness plan, second to develop a prototype of the product/service, third to get clients and fourth to become economically viable. Each step also has a clearly defined level ofknowledge that the tenant is supposed to reach before proceeding to the next stage.


The MICTI Incubator is interventionist, which means that it actively takes part in thedevelopment of the companies, but never takes ownership. The tenants are totallyresponsible for their own operations. The tenants have regular meetings with theincubator manager in order to follow up on executed tasks and activities, plan comingactivities, talk about current problems and their possible solution and to follow up onthe general work plan and future of the company.

The MICTI team is very flexible and receptive, responding quickly to needs andquestions from the tenants and, if deemed necessary, try to help the tenants to developtheir businesses. And this is something that is highly appreciated by the companies.To establish the incubator and carry out its activities during one year, MICTI received a grant from IDRC, and IDRC will also support the Research and Learning centre. Inorder to help new start-ups in early phases and through its entrepreneurship facilitation project, UNDP subsidized the monthly fee of the incubator tenants and created another loan fund to expand their businesses.

Webcom – a company that develops web sites, portals and a national search engine forMozambique – has been in the incubator from the start. “We pay a small, fixed,monthly fee for our office room, with computers connected to the Internet and someshared services”. “That’s what we need and we are very happy to be established in theincubator”, says Leonardo Xerinda, one of the founders of Webcom. “We have beenable to expand and to get customers faster and cheaper than have otherwise been thecase”. “And relations to the other companies in the incubator have also beenimportant”, he confirms. “Being established in the incubator gives us credibility”.

5. Requirements for a successful incubator

What does it take to put up an incubator and how could it be replicated?

It should be clear that there are various kinds of incubators; general or specialized in acertain area; focused on research, high-tech or general purpose, etc.The MICTI incubator faced many challenges and there are several factors that wereimportant in the successful implementation of the incubator.

A strong leader and a strong management team that can visualize the final product – i.e. the incubator – and keep the project on the right way are of uttermost importance.Without strong leadership, the project might lose speed or deviated from its originalobjectives.

The biggest challenge for the MICTI team was probably to find enough resources to set up and maintain the incubator, i.e. resources to buy equipment and pay salaries foremployees. The support from IDRC was crucial, which guaranteed the opening of theincubator. If it is not possible to find donor support, a partnership with local businesses and organizations is another possibility. Because of the nature of the incubator, it isdifficult to get the incubator itself sustainable, especially in the short term.

Another big challenge is to find qualified employees to run the incubator. Incubatorstaff needs to be multi-disciplinary and dynamic to give the necessary support to the


incubated companies. Qualified staff is rare in developing countries, especially ifresources are scarce. There is no formula on how to overcome the challenge, but thepersonal contacts and connections of the leadership and management team played acentral role in the case of MICTI.

It is also not easy to find mentors or a resource pool of skilled people that will give afew hours per month for free in order to support a specific company. Again, personalcontact and networking with business associations are important.The implementation team needs to have a lot of energy and be very flexible anddynamic. It needs to be dedicated and to have dedicated resources, and to have realistic view and perspectives.

Another factor worth mentioning is that the mix of selected companies is important for their development. A fruitful mix of companies means that synergies between them can be obtained and that different ways of cooperating can be established. This is especially essential in the early stages, when new companies do not have a lot of resources fortheir own development.

Networking and establishing relationship with key organizations and cooperationpartners should not be forgotten. There are normally more benefits to be obtained than initially thought.

6. Contributions to be made

The information society is changing everything about the world we live in, including the practice of development — ICT is now a basic component of development, not aluxury. ICTs have the potential to create jobs, improve access to basic services, increase the effectiveness of governments and facilitate the sharing of information with peopleliving even in remote parts of developing countries. This transformation of the globaleconomy is creating new networks that cross cultures as well as great distances,although access to, and use of, these technologies remains extremely uneven.

The scale of the ICT challenge is immense. Despite the forces of market liberalizationand globalization and efforts at public policy reform, the goal of achieving universalaccess to ICT and Global Information Infrastructure has remained elusive, and thedisparity in access to ICT is growing. Today 96% of Internet host computers reside inthe highest income nations with only 16% of the world’s population. There are moreInternet hosts in New York City than on the entire continent of Africa.

ICT has become an indispensable tool in the fight against world poverty. ICT providesdeveloping nations with an unprecedented opportunity to meet vital development goals such as poverty reduction, basic healthcare and education far more effectively thanbefore. Those nations that succeed in harnessing the potential of ICT can look forward to greatly expanded economic growth, dramatically improved human welfare andstronger forms of democratic government.

The risks are great, but so are the rewards. ICT can serve as a critical enabler to achieve many of the development goals agreed to by the world leaders.


New global and local competitive opportunities for ICT put a premium on skillsdevelopment. Developing countries need to address the capacity gap to secure not only a critical number of technical qualified people but also to acquire the expertise to assess, design and implement national ICT for development strategies.The scale of this challenge is immense. Yet there never been a better time for collective action to connect all regions of the world.

Information and Communication Technology (ICT) is universally recognized as thedriving force of the global information society and the knowledge-based economy.Economic growth may be accelerated by technological innovation and diffusion.Technological transformation is dependent on the skills and productivity of humanresources and sustainable development is greatly affected by the higher education andresearch capabilities of a country.

It is envisaged that the development of an initiative like MICTI will be a significantvehicle for expanding the Country’s skills and expertise base – in this case, ICT related –and contribute to the diversity and strengthening of Mozambique’s economic capacity.These developments will contribute to the growth of income generation opportunitiesby individuals and facilitate the Country’s participation in the global economy.

Fig 3. Minister Venancio Masingue and Constantino Sotomane, coordinator of MICTI.

34

Promoting African Research and Education Networking

Summary of prepared for the International Development Research Centre (IDRC) - Connectivity Africa – Promotion of African Research and Education Network Program (PAREN) program

Roy SteinerNyasha Tirivayi

Mike JensenPaul HamiltonJack BuechlerKaranya Gakio

1. Introduction

Internet connectivity in tertiary education institutions in Africa is in general tooexpensive, poorly managed and inadequate to meet even basic requirements. As arecent ATICS (Africa Tertiary Institutions Connectivity Survey) survey showed, theaverage African university has bandwidth capacity equivalent to a broadband residential connection available in Europe, pays 50 times more for their bandwidth than theireducational counterparts in the rest of the world, and fails to monitor, let alone manage, the existing bandwidth (ATICS 2005). As a result, what little bandwidth that is available becomes even less useful for research and education purposes.

However, initiatives within the continent point the way to a different future. NorthAfrica is the most advanced of all regions in Africa because universities in thesecountries have just recently become members of the EU MED Connect project, whichlinks them to high-speed undersea fibre networks. The potential for these types ofarrangements hold out the possibility to dramatically alter the bandwidth landscape intertiary institutions in the near future.

These experiences and those around the world argue strongly that there is an imperative to examine the potential to create initiatives to improve bandwidth access for Africanuniversities. A range of options are available. At one end relatively simple buyingconsortia can be created for even small groups of universities. On the other end acontinental not-for-profit telecommunications service provider or organization, whichwould provide a much broader range of services and ensure that effective network and bandwidth management practices are in place, can be created. Such initiatives would

Promoting African Research and Education Networking | 35

need to provide cost-effective and well-managed bandwidth services to the research and higher education sectors across Africa.

Average Bandwidth Costs/Kbps by Region

Source: ATICS 2005

The International Development Research Centre (IDRC) commissioned this studybecause it became aware of many parallel initiatives that would benefit greatly fromworking together both in terms of sharing information and in developing a commonvision. The economies of scale for bandwidth purchasing are straightforward – themore bandwidth gets purchased the less expensive it becomes. Only through unitedaction can these economies of scale be realized.

This is not by any means a new concept. Most of the organizations profiled in thisreport understand this principle and several excellent reports have already addressed this issue (Bandits, ATICS 2005). In addition, many donor organizations have taken theissue of improved connectivity very seriously and many of their efforts are reflected inthe success of the described networks. It is hoped that this study builds upon theseother initiatives and assists the community to address the many constraints to improvedconnectivity. These include political and regulatory barriers, the challenges of workingcross-regionally, the challenges of providing equal opportunities to Francophone andAnglophone and Lusophone institutions, to name but a few.

The vision that motivates all of us in overcoming these obstacles is that of a fullyconnected African research and education system. Through establishing low cost highquality networks a platform for generative discourse can be created leading to improved policy advice, more effective cross pollination of best practices and lessons learned aswell as encouraging an affinity towards cost sharing and partnership engagementmodels. Other benefits would include:• An increase in African research material on the Internet - content that is directly

relevant to the socio-economic development of African environments.

A v e r a g e B a n d w d i th C o s ts (USD pe r K b p s ) b y R e g i o n

0.12

0.52

3.18

4.38

4.70

5.46

8.00

0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00

U S U n iv e rsity

N o rth Afr ica

C e n t ra l Afric a

E a st Africa

S o u t h e rn Afr ica

Av e ra g e

W est Afr ica

R e g i o n

Mea

n C

ost

USD

/Kbp

s


• Improved educational standards of African universities. By accessing the same research materials, students will be able to make meaningful contribution to their areas of research.

• Literature searches truly at the fingertips of the researchers. • Enhanced peer review processes (both participation in reviewing and submissions)

that need to be facilitated through electronic communications networks.• Increased collaboration and partnerships among individuals and research

institutions and enabling regional comparative studies with localized partners. • Real time collaboration with other international researchers using video

conferencing. This aspect is very crucial especially in Medical research centres where critical theatre procedures can be conducted with the participation of other international experts.

• Enhanced institutional effectiveness.• The creation of knowledge networks - crucial in co-coordinating world and Africa

wide efforts in combating social ills such as HIV/AIDS and food insecurity.• Enabling regional centres of competences for local issues.

This survey was conducted over a period of four months and developed several keyobservations for improving connectivity to institutions in Africa. The study profilesseveral planned and existing regional and interregional groupings as well asintercontinental groupings. These are profiled separately from the national consortiums.

The study also looks at the following aspects:• Type of services being provided by the academic networks • The state of Internet infrastructure in Africa• Bandwidth cost models based on bandwidth source (terrestrial vs. satellite)• Governance and staffing in the academic networks.

The study concludes with recommendations for the formation of bandwidth consortia.

2. Key observations from the study

2.1 Network ProfilesThe following organizations were profiled:• National Research and Education Networks (NRENs): These are national

organizations in Africa representing groups of tertiary education institutions whose mandate includes improving connectivity for its members. In several cases these organizations are also involved with other priorities but all of them are actively involved with Internet connectivity.

• Regional and Continental Networks: These are often project specific networks in Africa that have a variety of mandates. However in all cases, bandwidth provision is critical to the success of the project or organization and in most cases a type of bandwidth consortia has therefore been created.

• Proposed Regional Continental Networks/Organizations: These are still in the proposal or concept note stage and they intend to develop bandwidth consortia to serve a variety of purposes.


• Organizations committed to increasing connectivity for tertiary institutions: These are the wide variety of organizations that have recognized increased connectivity as an important issue and have expended resources and supported initiatives to do so. These organizations do not have physical networks but are involved in one way or another in addressing connectivity.

• International Research and Education Networks: These are networks around the world that are also addressing the issue of connectivity for universities.

A complete list of the organizations profiled in the PAREN study is available in the full length report and can be accessed by visiting the following URL:http://www.connectivityafrica.org/page.php?file=PAREN_Report_final.pdf

Profiles of these organizations focused on:• The motivation behind the formation of the grouping. This includes the objectives

that the group wants to achieve. In most instances, these include the provision ofbandwidth to tertiary institutions at relatively lower prices than those in the industrial market.

• The dates when the establishment was set up.• Membership of the organization and the requirements for becoming a member.• The quantity and quality of bandwidth being made available to member

institutions.• Existing Network topology.• Description of the nature of the organization, paying attention to organization and

staffing structures.• How the organizations are financing their operations. • Services that are being provided to members by the consortiums.• The challenges that are being faced by these networks in carrying out of their

duties, including in their actual establishment.

The profiles indicate a wide range of different network structures, managementorganizations, membership and services provided to members. The majority of theAfrican NRENs have as one of their key objectives the improvement of connectivity of their member institutions. Of the nine operational academic networks in Africa, theNigerian and Malawi networks, NUNet and MALICO are VSAT based while theEgyptian networks, EUN and ENSTINET, having been established in 1986 and 1987respectively, are fibre based. The majority of the member institutions pay subscriptionfees to remain part of their networks. Some networks are currently oversubscribed with more member institutions than the network was intended to support as is the case with for instance KENET in Kenya.

As expected, networks in Europe, Asia and Latin America seem to be more advancedthan their African counterparts in terms of network infrastructure, bandwidth capacityand membership numbers. However, it is encouraging to note that several regionalinitiatives and networks are in the making and some national academic networks inNorth Africa have already fostered beneficial partnerships with European networks e.g. NRENs in Tunisia, Egypt, Morocco and Algeria now have high speed networks linkedthrough EUMEDConnect while GEANT and TENET South Africa have recently


established a link at the Telkom UK hub that will enable peering between the twonetworks.

2.2 Network servicesMost of the networks in Africa focus on the provision of basic Internet services, billing and network management. However, more effort is needed to address issues such asbandwidth monitoring and management, negotiating for the deregulation of policies,which affect bandwidth provision, and bargaining for better pricing.

The following chart indicates the frequency of services provided by the currentacademic networks in Africa.

State of Internet Infrastructure in Africa

2.3 State of Internet Infrastructure in Africa The state of Internet infrastructure can be briefly explained as follows:• Existing terrestrial links reflect a historic layering, which started with the Panaftel

analogue network of the 1970s and 80s. In many cases, the Panaftel network within a given country comprises the national transmission backbone today. Many of the links have deteriorated due to lack of funding and maintenance, and have fallen into disrepair, and some sections have been damaged or destroyed during wars. In

Frequency of Services in African Networks

0 5 10 15 20

QoSDial Up IPMirroring

LBEIPv4

ATM/PVCVPN

FundingBilling

IPv6Research

IPR & HostingConsultancy

MonitoringVO IP

License NegotiationNetwork Mgmnt

TrainingInformation

MulticastVideo Conf

SupportBasic IP

Serv

ices

Networks


some countries, the analogue network has been upgraded and extended as the incumbent operator has built out its national backbone. The full extent, operationalstatus, or capacity, of national backbone links across all African countries is not at this stage fully known.

• All of Africa is currently covered by satellite bandwidth. By December 2003, thirteen satellite operators had 51 satellites deployed with coverage over Africa. These satellites have pointed 39 Ku-band frequency beams and 30 C-band beams throughout the continent.

• There are a number of planned regional connectivity projects that include; the East Africa Digital Transmission project, COMTEL, Intelcom II, COM 7 and the submarine cable systems SEAMEWE II, SEAMEWE III, Sat-3/WASC and EASSy which surround the Northern, Western and Eastern parts of Africa. The following map shows the SAT-3 cable and its landings as well as existing inland fibre Networks.

SAT-3 Undersea Fibre Submarine Cable and Inland Fibre Networks

Source: IDRC 2004

This map demonstrates that, in the near future, it will be possible for most Africancountries to have external-level connections to the general Internet via a high capacity


link. There is, however, a lot of ground that has to be covered before this becomes areality.

• The overall supply pattern of bandwidth leaves a considerable gap in the market that can be filled by satellite-based suppliers in the short to medium term,particularly in areas (the interior, Eastern and central Africa) not yet reached by regional fibre projects, and in those countries which do not have national transmission backbones which reach hinterland locations (like Nigeria, Angola and DRC).

2.4 Bandwidth Cost ModelsThe costing structure of any African network will be influenced by the choices ofnetworking made, since terrestrial-based and VSAT-based networks have very differentcost structures.

Given the bandwidth situation already existing in Africa, it appears that both VSAT and terrestrial networks will have to be used to cover all institutions. This situationcomplicates the generation of a uniform pricing model for the network. However, on a terrestrially delivered multiple site network it may be difficult to calculate the pricingmodel without the correct tools in place. The simplest way to do this anyway is to group sites that are connected back to the point of delivery, divide their bandwidth allocationby their costs and charge accordingly.

A satellite-delivered network is easier to cost out. The service is directly delivered andthe cost can therefore be determined directly according to the service provided. Once a determination is made for the bandwidth required, the vendor would calculate the price of services supplied and charge accordingly. The bandwidth volume has a dramaticeffect on bandwidth pricing and savings of over 50% are possible if institutions bandtogether to purchase their bandwidth in larger volumes.

2.5 GovernanceAll information technology networks have some form of an oversight body (usuallyrepresented by a board), an advisory body (usually represented by an advisory committee), executive management (usually represented by secretariat managers) and operations (usuallyrepresented by technical staff). Models of governance and management found in theprofiled networks included:• The Reciprocity Model: The majority of existing networks are set up as reciprocity

networks, with members expecting mutual and equal benefits after making equal investments e.g. KENET (Kenya).

• The Donor Centred Model: Some networks or initiatives are funded by a single organization that takes the dominant role in management and governance e.g. USAID Leland Initiative, SPIDER, Sida/SAREC initiative and the Partnership for Higher Education.

• The Special-Member Model: Finally, there are networks where two or three parties make a greater investment in the project than the rest of the members. Those members with ‘extra’ amounts of investment then form a special group of members with certain privileges, as defined in their AUP for example CUDI in Mexico.


Staffing patterns observed from the study include:• Full Operational Staffing: The organization sets up a complete operations structure,

with both administrative and technical support offices for instance APAN (Asia Pacific).

• Decision Focused Staffing: This pattern consists of a single decision-making group that has the responsibility of outsourcing the requisite specialist services. Under this model, members are often tasked with maintaining their own link to the network. The group has responsibility over the usage of both the facilities and bandwidth as is the case with for instance TIEN and TENET.

3. Recommendations for Bandwidth Consortia

3.1 Establishing a Pan African VSAT ConsortiumDue to the limited availability of international fibre gateways and national backbones,the establishment of regional or continental consortia for bandwidth purchasing is likely to focus on the purchase of satellite bandwidth but may also need to consider emerging international or national backbones in the near future. Given the fact that terrestrialinfrastructure is still disjointed and sparse across much of Africa, the only immediateway to deploy a world-class educational network within an initial 12-month period is to implement a satellite-based network. The hardware for these types of networks isavailable at relatively low cost. It is also of higher quality and is more easily available.There are essentially three ways in which a VSAT consortium could evolve:

• Creation of several regional and/or research-focused networks: For example, East, Southern and West African tertiary institutions could each create their own VSATconsortium. This way, each network would be able to deliver significant value because critical volumes would be obtained. Additionally, other specialty networks (i.e., for medical research, such as MIMCom, or language-focused networks [French West African]) could also be established.

• Organic growth of existing established network: The obvious example of this is the initiative being taken by the African Virtual University in sourcing bandwidth for the five universities participating in the Partnership for Higher Education. Over time, other institutions could join the consortium, and it could become the dominant bandwidth provider.

• Establishment of an independent entity focused on bandwidth provision: African educational institutions and committed donors could decide to follow the example of most other networks in the world and create an independent organization that is focused on providing the lowest cost bandwidth to its members.

3.2 Governance structure of Pan Africa VSAT ConsortiumThe study recommends that any Pan African VSAT consortium adopt the followinggovernance structures:

Independent Board of Directors: This body would serve the primary oversight functions of a board: strategic direction and fiscal responsibility. Ideally, its members would represent a broad range of stakeholders, including representatives from governments, educationinstitutions and the private sector. It must be seen as being independent and acting in


the best interests of the membership. The selection of the board would be theresponsibility of the Membership Advisory Group and the funding agencies. Membership Advisory Group: This would comprise a group of not more than nine electedrepresentatives of the educational institutions making up the membership of theconsortium. The primary responsibility of this advisory group would be to work withthe donors in selecting an appropriate board of directors and providing feedback asneeded from the membership.Donor Advisory Group: This is envisioned as a group of donor representatives who would ensure that the strategic interests of the project are being met. The primary role of thisgroup would be to select the board of directors and to ensure that funding is channelledto the consortium as needed.

3.3 Staffing structure for a Pan African VSAT ConsortiumThe staffing of the consortium organization should be minimized in order to ensureefficiency. Ideally, the technical management of the network (i.e., the management ofthe Network Operations Centre [NOC]) would be handled by the consortium’soperations department (at the consortium headquarters), which would monitor network performance.

3.4 Network Configuration for a Pan African VSAT ConsortiumThere are two possibilities for the network configuration:• The network configuration would be a meshed network with a central hub in either

the US or Europe. The network should not be over-complicated. Whilst the geographical spread is wide, if similar terminals – with only a few permutations on the hardware – were deployed from a central African store, it should not be too difficult to achieve a large number of deployed sites per month.

• The network configuration would still be a meshed network but with a central hub in Africa. In this case member institutions have ownership of the whole initiative.

3.5 Funding and Sustainability StrategyIn order to achieve sustainability, the consortia will have to pursue a strategy of agradual reduction of bandwidth subsidy over the course of 5 to 10 years. The exampleof TENET SA is instructive: the organization provided subsidies on increasingquantities of bandwidth over three years. At the end of this period, member institutions were covering all of their own bandwidth costs. The consortium will need to be a legal

Operations(IT Departments)

Executive ManagementGroup

Expert CommitteesTechnical Service Providers

Special Projects


body that will sign the bandwidth contracts with the VSAT supplier in order to obtainthe volume discounts desired.

The consortium must then enter into binding contracts with each of the educationalinstitutions for the provision of their bandwidth. It is expected that some memberinstitutions will default on their payment obligations, and strategies must be in place tohandle this eventuality. There are four primary mitigation strategies:

• Member institutions must prepay their bandwidth fees. If payment is not received on time, services will be immediately cut off. Usually, this creates a crisis within the institution, and funds are immediately prioritised for bandwidth.

• If member institutions create a history of non-payment, the amount of bandwidth they could purchase would be reduced.

• Estimates of expected default rates could be made and a “self insurance” fund created to cover the expected shortfall. Such a fund would be built into consortium charges.

• Insurance companies are invited to provide insurance against default.

3.6 Regulatory StrategyAn important role of the consortium will be to negotiate with governments to allow the use of satellite and/or to eliminate license fees and monopoly pricing for educationalbandwidth. A well-conceived diplomatic strategy would have to be pursued in order toaccomplish this. Precedents, both within and outside of Africa, indicate that this ispossible. EUMEDCONNECT, for example, was able to assist universities in NorthAfrica in lobbying their governments for educational waiver on fees and pricing. Also,the Partnership for Higher Education has not had much difficulty in doing the same for the sub-Saharan countries in which they operate.

3.7 Location of ConsortiumFrom a technical standpoint, it would be best for the consortia headquarters to belocated near the central hub in other words, in Africa, the US or Europe. Politicallyhowever, it would be best if the headquarters, together with the central hub, werelocated within Africa in a country with; good relations with the rest of Africa, technicalinfrastructure advanced enough to meet the needs of an international office, facilitiesfor easy travel to and from headquarters, politically stable and where highly qualifiedstaff are willing to be located.

3.8 Partnerships with Private SectorsStrong partnerships will need to be forged with the various private-sector serviceproviders. They will include:

• The satellite operator• The satellite equipment suppliers• The networking equipment suppliers

Any bandwidth consortium would have high visibility, a large number of sites,significant bandwidth and important political value. Given these characteristics, and the fact that most private-sector players have some sort of social responsibility program,


favourable terms could be negotiated for their services. However a cautious approachwould need to be taken and strong legal/contractual relationships would probably bemore reliable than sole dependency on preferential deals. This would mitigate againstprivate sector partners defaulting from a preferential deal where they would have theupper hand. GEANT, ALICE, EUMEDconnect and TEIN2 are handled in thismanner.

3.9 Partnership with Regional Academic Networks in Other ContinentsInternational networks such as Geant and Internet2 have already indicated their interest in assisting with linking their networks to similar emerging African networks. Thesecould provide relatively low-cost upstream connectivity for African institutions. Also, it would be important to foster South-South knowledge exchange by ensuring thatAfrican networks are linked to their counterparts in Asia and Latin America.

3.10 Recognition of Evolution of Bandwidth SourcingThe consortium will have to actively monitor the bandwidth landscape in Africa and, as terrestrial options become more attractive, be able to quickly shift connections forcertain institutions to the most competitive option. This will mean that, over time, theconsortium will evolve from providing bandwidth via VSAT only to providingbandwidth via terrestrial sources as well.

4. ConclusionThe proposal to create bandwidth consortia comes at a time when changes in theconnectivity landscape in Africa are creating dramatic opportunities for increasingonline access. Not only is there increased content that is directly relevant to Africa, but also there is the fact that bandwidth volumes have increased at the same time thatsignificant policy reforms are enabling access to this bandwidth.

This report shows in detail how a potential consortium can be formed. It is now up tothe African academic institutions in consultation with the donor community todetermine how best to turn this idea into reality.

The full PAREN study is available at:http://www.connectivityafrica.org/page.php?Documents.html

5. Reference

Steiner R. et al, African Tertiary Institutions Connectivity Survey, AVU, Nairobi. 2004,http://www.atics.info/index.html

45

A Survey of Investment in Education and ResearchNetworking in Africa by Development Agenciesand other Organizations – Summary

For the International Development Research Centre (IDRC) -Connectivity Africa – Promotion of African Research and Education Network Program (PAREN) program - August 2005

Kate Wild

The author wishes to thank all those who responded to e-mails and phone calls in order to contributeinformation to this report. Any errors or omissions are the sole responsibility of the author.

1. Purpose of the study

This investment study was commissioned by IDRC as part of its program to promoteeducation and research networking in Africa (PAREN).

It complements earlier IDRC work that identified research and education networkinginitiatives in Africa and models on which they might draw. The network study(Promoting African Research and Education Networking, A study sponsored by IDRC, Steiner et al, January 2005) concluded that connectivity in tertiary institutions in Africais expensive and totally inadequate to meet basic education and research requirements.It concluded, however, that new initiatives in the region could bring about a landscapein which sufficient bandwidth would be available to enable African universities toexchange information and data among themselves and to contribute to and draw from global developments in education, research, science and technology.

This present study is intended to identify external partners working with Africaninstitutions to achieve those ends.

The full report is available at:http://www.connectivityafrica.ca/page.php?Documents.html.


2. Methodology

The terms of reference called for a profile of organizations contributing to educationand research networking in Africa and a measure of the extent of their contribution inrelation to:• bandwidth subsidies• national research networks• regional research networks

The profiles were developed through a combination of web research, phone interviewsand e-mail exchanges conducted mainly in a six week period in March and April 2005.

Many organizations were examined, with the most relevant multilateral organizations,donors, foundations, academic and non-governmental organizations and businessesincluded in the report. Organizations providing content and technical training andsupport are covered as well as those focused mainly on connectivity.

Initiatives that may have a bearing on future networking possibilities – the most visible of which is the Africa Commission – have also been included.

Since the research work was completed a number of programs have advancedsubstantially and new ones have emerged. Examples of the former include the India-Africa Partnership Project and the ITU/UNU collaboration on AFUNet (AfricanUniversities Network) supported by a feasibility grant from NORAD. CERN – notincluded in the study – is now taking an active role in stimulating debate prior to theTunis phase of the World Summit on the Information Society. DFID is launching amajor new content initiative entitled Research Africa.

The report that is summarized here therefore needs to be seen as a picture, at aparticular time, of an ongoing process.

3. Coverage

The study combines a compilation of profiles of organizations and initiatives withanalysis of how they can contribute to building a more effective networkingenvironment within the relatively short term. The analysis is intended to lay thefoundation for future joint action to expand bandwidth access.

Many of the organizations profiled here are bilateral or multilateral developmentorganizations, foundations or multinational companies based outside Africa. Others are key African stakeholders whose commitment will determine the success of thebandwidth enterprise within the African research and education community.

Information is provided on over 60 programs or organizations.

A Survey of Investment in Education and Research Networking in Africa | 47

Section in Report Organization or program Website

3. The Africa Commission

Commission for Africa www.commissionforafrica.org

4. Organizations involved in current consortium

African Association of Universities

www.aau.org

Partnership for Higher Education in Africa

www.foundation-partnership.org

African Virtual University www.avu.org5. Multilateral Organizations

World Bank www.gdln.orghttp://info.worldbank.org/ictwww.infodev.org

UNESCO www.unesco.orgITU and UNU www.itu.int/wsis/docs2/NEPAD www.eafricacommission.orgAfrican Development Bank www.afdb.org

6. Bilateral Development Agencies

Sida/SAREC www.sida.se

Swedish Program for ICT in Developing Regions (SPIDER)

www.spidercenter.org

USAID www.nettelafrica.orgwww.aascu.orgwww.dot-edu.org

DFID www.dfid.gov.ukNORAD www.norad.orgDANIDA www.danida.dkCIDA www.acdi-cida.gc.caIDRC www.idrc.caNUFFIC www.nuffic.netIICD www.iicd.orgItaly www.innovazione.gov.it/itaGTZ www.gtz.de/en/InWent www.InWent.org

7. The European Union Geant www.geant.netDante www.dante.netEUMEDConnect www.eumedconnect.net

8. La Francophonie AIF www.agence.francophnie.orgAUF www.auf.orgINTIF http://intif.francophonie.orgSIST www.sist-sciencesdev.netAFD www.afd.fr

9. InfrastructureProjects

Intelsat www.intelsat.com

Inmarsat www.inmarsat.comNew Skies www.newskies.comPanAmSat www.panamsat.comRascomStar www.rascomstar.comIEEAF www.ieeaf.org


Nectarnet www.nectarnet.orgORASCOM www.otelecom.comCOMTEL www.comesa.int/ict/projectsEASSy http://eassy.orgFibre Africa proposal [email protected]

10. Content Providers WHO www.who.intFAO www.fao.orgINASP www.inasp.infoeIFL www.eifl.netJSTOR www.jstor.org

11. Research Networks MIMCom Net www.nlm.nih.gov/mimcomAIMS www.aims.ac.zaSARUA www.sauvca.org.za/sarua/

12. Technical Capacity NSRC www.nsrc.orgWiderNet www.widernet.org

13. Private Sector CISCO www.cisco.comGeo International www.geointernational.netGlobal VSAT Forum www.gvf.orgHP www.hp.com

14. Global Scientific Collaboration

NSF www.nsf.gov/cise

Internet2 http://international.internet2.edu

MSI - SIG www.msi-sig.orgGUS [email protected] http://www.nmiscience.org/

15. Other Initiatives Ibaud www.ibaud.orgNethope www.nethope.org

4. Analysis of findings

The analysis aims to:• identify challenges facing development partners as they confront networking issues.• provide an overview of who is contributing what and where in terms of:

o connectivityo capacity buildingo contento national networks ando regional networkso recommend approaches to framing the next phase of debate on education

and research networking in Africa.

4.1 Development ChallengesThe meaning of national research and education networks in AfricaAccess to high speed, high quality bandwidth in African universities will be anincreasingly important component of both African and global science. But it may need to be balanced, for some time to come, against the need of student and teachingpopulations in universities and research centres, in rural as well as urban areas, forfunctional levels of connectivity. The concept of closed networks dedicated to scientific


endeavour which underpins national research networks in the north (NRENs) may not prevail in Africa for some time.

A relatively small number of core donors but many interested onesFew development organizations have a substantial commitment to funding bandwidthprograms within universities. The core donor community includes the Partnership forHigher Education in Africa, the Agence Universitaire de la Francophonie (AUF) andSida/SAREC. Although in the case of Sida/SAREC it is difficult to identify preciseconnectivity investments both the Partnership and AUF plan to invest more than $1million annually in connectivity as part of broader programs to support universityreform. While important players, these are by no means the only players on the stage.Many others, organizations mentioned in this report (the World Bank, OSISA,WHO/HINARI, etc), and probably some left out, fund connectivity as a tool tosupport their knowledge-based programs.

The diversity of donors with varied agendas has led some national partners to call formore collaboration among international development partners: • at the regional level to share information and rationalize resources. • at the national level to collaborate on specific projects.• to minimise duplication of efforts at all levels.

MainstreamingEnhanced connectivity for the tertiary sector may not be seen as a priority within ICTprograms that have been integrated into core development sectors to support theachievement of the MDGs.

But access to higher levels of bandwidth is a mainstream requirement of the tertiarysector, which itself is a strategic instrument of national science and technology policy.There is room for elaborating the bandwidth case in the context of education andresearch, science and technology and national development goals.

A variety of modelsMany satellites circle Africa and fibre cables surround all but the East coast of thecontinent. Arguably Africa is well connected now to the rest of the world. Connectivity within the region is much more problematic.

Debates - about the relative merits of satellite and fibre and the priority of linking to the external world or networking within the region - will ensure that there is no single right model. They argue in favour of flexible project design so that programs can beinterconnected as technical and commercial circumstances evolve.

A Pan-African approach?Connectivity in tertiary institutions was addressed initially on a country by country basis. Now geographic (SARUA) and interest-based (AUF) programs are emerging.

While there is not likely to be a single education and research networking model for the continent there is value in comparing experiences, linking sub-regional networks andseeking opportunities to aggregate demand where it makes sense to do so. Defining


what constitutes a constructive program in support of a continental framework is thenext challenge.

4.2 A regional overview of programsThe following 12 programs were used as sample through which to assess the geographic spread of connectivity programs in the region: • African Virtual University – AVU has transformed itself from a World Bank

project to an inter-governmental organization with 34 learning centres in 20 countries.

• World Bank’s Global Development Learning Network – GDLN networks learning institutions to promote knowledge sharing, training and consultation on development topics.

• CISCO Networking Academies – CISCO has programs in 40 countries in Africa to teach networking and other information technology related skills.

• The Partnership for Higher Education in Africa brings together the Carnegie, Ford, MacArthur and Rockefeller Foundations to assist with the renaissance of higher education in Africa inter alia through enhanced ICT skills and connectivity.

• Sida funds, often through its research arm SAREC or the university-based SwedishProgram for ICT in Developing regions, the deployment of ICT in universities, including connectivity.

• MIMCom is the communication arm of the Malaria Research Network established to meet scientists’ needs for e-mail and fast communications in the early Internetdays.

• WiderNet is a program based at the University of Iowa to improve educational networking mainly in Africa and support access to content through the e-GranaryDigital Library.

• NetTel Africa is a public/private capacity building alliance linking regulators and universities in Africa and the US.

• EUMEDConnect is an EU project to network R&E communities of the Mediterranean including North Africa with 3500 European establishments served by Geant.

• AUF promotes the creation and circulation of French-language scientific knowledge inter alia through the integration of ICT in higher education and the establishment of a network of “campus numeriques”.

• SIST is a program of the French Ministry of Foreign Affairs to build a system for scientific and technical information in Africa.

• INASP works with partners globally to strengthen local capacities to produce, manage, access and use scientific information and knowledge.

In six countries – and for different reasons - there is no discernible presence ofconnectivity programs: Republic of Congo, Equatorial Guinea, Libya, Mauritius,Seychelles and Sudan. Programs tend to cluster in a few countries.


Number of programs Countries9 Uganda, Tanzania8 Mozambique7 Ghana6 Burkina Faso, Ethiopia, Nigeria, Senegal5 Mali, Cote d’Ivoire0 Republic of Congo, Equatorial Guinea, Libya, Mauritius,

Seychelles, Sudan

The broad brush geographic picture shows concentrations of activities in support ofeducation and research networking in ten countries of the 53 in the region: Uganda,Tanzania, Mozambique are at the top of the list with substantial presence also inBurkina Faso, Ethiopia, Ghana, Nigeria, Senegal, Mali and Cote d’Ivoire.

A number of countries, and institutions based outside metropolitan areas generally, arevirtually untouched by the advance of externally-funded programs to supportconnectivity.

4.3 ConnectivityThe table on the next page shows what the study has been able to establish aboutsubsidized bandwidth. It illustrates different approaches.

The Partnership for Higher Education has a long term commitment to expandedbandwidth for universities in the five countries with which it works. The program hasbeen ongoing for close to five years – it has involved strong links with both universitymanagement and technical staff and a significant capacity building effort. Moreimportantly perhaps it has committed funds (approximately $4.6 million) into the future to develop and implement bandwidth pooling and management programs. ThePartnership member universities have opted for a satellite solution. While thebandwidth agreement that is presently being negotiated covers only five countries more may be included in the next phase.

The Agence Universitaire de la Francophonie works in 15 Francophone Africancountries to develop digital campuses (10) or information centres; the ten digitalcampuses are intended to stimulate national networking, content sharing andconnectivity. The quantities of connectivity are relatively modest but the program isbroad-based. Connectivity is provided through local ISPs – satellite solutions areconsidered only when no other options are available. Connectivity and related staffcosts within the universities are covered through AUF’s operational budget at a level of approximately 1.3 million euros annually. Training and other program activities arefunded as projects also by AUF.

WB-

GDLN

AUF

CNSi

daHI

NARI

PHE

plann

edEU

MED

Gean

tAV

U

Alge

ria15

5mbp

sBe

nin38

4kbp

s25

6kbp

s12

8/64k

bps

Burki

naFa

so,O

uaga

Bobo

-D38

4kbp

s plan

ned

512k

kbps

128k

kbps

128/6

4kbp

s

Buru

ndi

128/6

4kbp

sCa

mero

on25

6kbp

sCh

ad51

2/128

kbps

DRC

256k

bps

Cote

d’Ivo

ire38

4kbp

s25

6kkb

psEg

ypt

384k

bps

45mb

psEt

hiopia

384k

bps

??

Gabo

n2m

bps

Ghan

a38

4kbp

sne

gotia

tion

CC: 5

12/12

8,GI

MPA:

256/1

28Ke

nya

384/9

6 kbp

sMa

lawi

?Ma

li51

2kbp

s12

8/64 k

bps

Maur

itania

384k

bps

128/6

4 kbp

sMo

rocc

o34

mbps

Moza

mbiqu

e38

4kbp

s1m

bps?

nego

tiatio

nNa

mibia

384k

bps

Nige

r12

8/64 k

bps

Nige

riane

gotia

tion

Rwan

da12

8/64 k

bps

Sene

gal

384k

bps

2mbp

sSa

int-L

ouis:

128/6

4 kbp

s,UC

AD: 1

28/64

kbps

South

Afric

aGe

ant

Swaz

iland

Tanz

ania

384k

bps

1mbp

s?ne

gotia

tion

Tunis

ia45

mbps

Ugan

da38

4kbp

s?

nego

tiatio

n


The World Bank’s Global Development Learning Network covers 12 countries –connectivity is delivered through the Bank’s own network. Emphasis is on the localdevelopment of services that deliver the benefits of connectivity – collaborative learning and a wide variety of national, regional and global consultations - rather than onbuilding networking skills per se.

AVU, funded initially by the World Bank and now by a number of developmentagencies, supports connectivity in the distance learning centres through which it delivers its programs in member universities.

EUMED Connect provides substantial connectivity to four countries in North Africathat border on the Mediterranean. Of all the programs discussed in this report EUMED is the closest to connecting African countries into the global research network.

Sida and HINARI provide connectivity in countries where it is important to the delivery of other programs. In the case of Sida connectivity is significant and intended tosupport university wide education, research and collaboration goals. In the case ofHINARI connectivity support is limited to providing small amounts of bandwidth tofacilitate access to information tools.

Like a number of bilateral aid agencies, USAID does not generally support connectivity but does so on occasion where it is a requisite of nationally developed strategies.Namibia and Uganda are examples where connectivity is provided to teacher traininginstitutions in the context of its dot-EDU program.

These different models for the provision of connectivity reflect the different programgoals of development organizations.

4.4 CapacityAlmost all the programs identified involve capacity building. The main difference isbetween the majority of programs that build networking skills in the context of theirown programs and the few organizations that make their services available to buildnetworking capacity in support of a broad range of African initiatives.

The Sida/SAREC approach to training is comprehensive in coverage but focused onfew universities: it includes practical programs that join Swedish experts with localteams to implement solutions to local problems; professional training in Swedishuniversities; and ICT research through sandwich programs at the graduate level. Sida iscommitted to exploring open e-learning environments for networking and ICT training.

The Partnership for Higher Education has initiated workshops and research projectsdesigned both to expand capacity and to move the connectivity agenda forward through collaboration among the African partner universities.

EUMEDConnect offers a mix of formal training programs for network engineers andthe facilitation of study tours and participation in seminars and conferences to stimulatenetworking among individuals.


AUF offers national and regional workshops on various aspects of the application ofICT in a university environment: network and system administration; design andimplementation of information systems, web publishing, research and multi-media; and education technologies – the capacity building activities of both AUF and SIST have astrong content development and dissemination component. GTZ is also active in thisarea.

Not surprisingly the content providers provide training on how to use their ownresources but some also have broader focus on building capacities to access, produce,process and disseminate information – HINARI and INASP are examples. INASP hasalso developed methodological tools for bandwidth management and will continue tobe involved in building capacities in this area.

The Network Startup Resource Centre is specifically focused on capacity buildingthrough training programs, technical assistance and trouble shooting. It has been astrong contributor to the training workshops run first by ISOC and increasingly by theAfrican Network Operators Group (AFNOG).

More coordination of networking training could establish good practice, fill gaps incoverage and help ensure that sufficient African capacity is available to meet expanding needs.

4.5 ContentMost of the content initiatives have dual objectives of facilitating access by Africanresearchers, educators and students to the body of scientific knowledge which is nowelectronically at the fingertips of academics in the North and processing anddisseminating knowledge generated in Africa. The Francophone initiatives – AUF andSIST – are particularly focused on building and disseminating local knowledge.

Some services offer broad subject coverage – INASP, eIFL, JSTOR – and workthrough national and university libraries; some cover specific sectors (HINARI forhealth and reproduction, AGORA for agriculture) and work also through specializedfacilities: medical and nursing schools, tropical disease research centres, nationaldepartments of health and agriculture. Because of their longstanding relationships withspecialized libraries and information centres, sometimes outside major metropolitanareas, they can be a useful source of information and contacts about research and datainitiatives relevant to core health and rural livelihood issues.

The content services are important partners in making the case for bandwidth from auser perspective.

4.6 National networking Algeria, Egypt, Morocco and Tunisia in the North; Kenya, Malawi, Nigeria, SouthAfrica and Tanzania in sub-Saharan Africa provide examples of African nationaleducation and research networks.

The closest approximation to northern models exists in North Africa as a result ofEUMEDConnect’s activities in Algeria, Egypt, Morocco and Tunisia. Progress has not


been as fast as intended partly because the package of measures incorporated in theEUMED program (applications and regulatory reform as well as connectivity) could not all be delivered simultaneously. But NREN nuclei are in place and will expand duringthe next phase of the project.

The gradual evolution of national networks is central to the AUF "campus numeriques" program; the campuses work with universities to build ICT capacity and infrastructureto facilitate cooperative programs and information sharing nationally. The SISTprogram of the French government also works through national committees whichinclude universities and research centres. SIST and AUF coordinate their programs atthe national level.

TENET (South Africa) – linked to GEANT - is the only broadband NREN in sub-Saharan Africa.

KENET (the Kenyan Education Network) – originally supported in 1999 by theUSAID’s Leland Initiative but now operating independently – provides training anddiscounted access via leased lines and dial-up facilities to over 30 academic institutionsin the country.

A number of donors are supporting networking in Nigeria. Six Nigerian universities are members of the Partnership consortium. The Carnegie and MacArthur Foundations are working specifically on a national bandwidth capacity building network which will focus first on technical and management issues then on policy. SIST is working in Nigeria todevelop skills related to information processing and access. WiderNet also has programs in a number of Nigerian universities to build Internet connectivity and capacities.

MaLICO has very strong local roots within a consortium of Malawi university librarieswhich developed the network as a tool to improve access to electronic informationresources. MaLICO was developed with only a small amount of start-up support fromOSISA but is now operating independently. It is the only African example of a library-driven network which serves NREN goals.

This varied pattern conforms to the view on African NRENs expressed by the manager of South Africa’s TENET: NRENs in Africa do not focus only on the needs for highquality, high volume bandwidth but on the more mundane needs of students andteachers to connect with a wider world.

4.7 Regional networksThere is a host of regional networks – falling outside the purview of this study - thatcould take advantage of improved bandwidth infrastructure on a regional basis if it were available. These include CODESRIA – the Council for the Development of SocialScience Research in Africa, AERC – the African Economic Research Consortium andRIA – Research ICT Africa.

Many of the programs described here have coordinating centres or focal points incountries of the region forming networks of common interest which would similarly


benefit, as would the regional economic coordination organizations (ECOWAS,COMESA, SADC, IGAD).

MIMCom and NetTel are subject focused networks which recognize the strategicimportance of connectivity to the success of their substantive agenda – and the African Mathematics Network, while only in the planning stage, shares this view.

Through EUMEDConnect, the North African countries (with the exception of Libya)are linked functionally within a regional network to all Mediterranean countries.

The proposal developed by SARUA (the Southern Africa Regional UniversitiesAssociation) for a regional network covering universities in SADC countries could – if it materializes - provide a model for other sub-regional networks. TENET in South Africa is an important asset in the sub-region that is so far lacking elsewhere.

4.8 The next phasePrinciples drawn from current practice may be a useful guide for future bandwidthprograms:• African education and research networking may be driven, at least in an initial

phase, more by the need for bandwidth for all tertiary education operations than by the need for secure, closed research networks on the model of northern NRENs.

• Network infrastructure requirements must be located firmly within the higher education and research agenda – linked to national science and technology and development goals – as well as to the program requirements of individual universities.

• Senior university management as well as IT leaders must be part of the debate on network and bandwidth requirements.

• More collaboration among donors can help produce a wider geographic spread among programs and facilitate comprehensive capacity building strategies as well as minimize duplication and highlight good practice.

• Development organizations need to recognize that long term planning cycles are required and that resources need to be committed independently of the particular technical solutions that emerge.

• Program design needs flexibility to adapt to new infrastructure options (technical and commercial) as they emerge.

• Synchronizing regulatory reform, applications, content and capacity building with the provision of expanded bandwidth will help promote the use of the newly available resources.

The goal of the next phase should be to build support broadly within the region andamong external partners by:• engaging policy makers within the science and technology, communication and

education communities and• multiplying and connecting national and sub-regional programs on the ground.

Any new continent-wide bandwidth program may want to consider the followingactions to facilitate identification of opportunities and expand support.


• Maintaining a watching brief with respect to discussions on the higher education recommendations of the Africa Commission Report – through the AAU (African Association of Universities).

• Monitoring relevant global initiatives – Global University System, Nelson Mandela Institute MSI, etc.

• Establishing links with the higher education program in UNESCO through the African Regional Scientific Committee of the Forum on Higher Education or through the Working Group on Higher Education.

• Engaging La Francophonie through the AUF.• Seeking common ground with NEPAD.• Opening discussion with key African research networks to establish their needs for

expanded bandwidth.• Involving a few key private sector representatives – perhaps from Geo

International and Cisco – to try to identify concrete contributions in support of regionally defined initiatives.

• Reinforcing the content angle by incorporating WHO as the representative of HINARI and AGORA – two key development sectors where African research and education will make a contribution in the context of the Millennium Development Goals.

Several strands of program activity (advocacy; technology; capacity building; local andinternational content) will be needed to construct a firm foundation for broadexpansion of education and research networking throughout the region.

58

Problem-Oriented Capacity-Building based on the Introduction of Information and Communication Technologies as an Enabler of Socio-EconomicDevelopment

Björn Pehrson

AbstractThe Swedish Royal Institute of Technology (KTH) presents experiences from a modelfor conducting development cooperation projects. The model has multiple goalscoming together in a problem-based, project-driven framework involving stakeholdersfrom academia, public sector, industry and civil society.

Primary goals are• Involve all stakeholders in concretising, maintaining and working towards a vision

and action programme exploiting the potential of information and communication technologies (ICT) to improve the quality of life of individuals, and the efficiency of organisations.

• Define and conduct a number of development projects in the framework of the action programme leading towards a vision that can be used to establish a problem-oriented approach to capacity building in the ICT area at selected universities.

The activities should include • Selection of geographical areas to target and university partner(s) to involve in

development projects.• Education and training of academic staff in appropriate pedagogical models for

teaching and learning, such as problem-oriented learning, peer-learning, vicarious learning, and associated roles for different stakeholders, including teachers, students, project principals, external experts, etc.

• Strengthening of research and development at involved universities regarding technologies and entrepreneurship relevant to rural development, including design, development and deployment of pilot services and networks

• Problem-oriented education and training in technologies and entrepreneurship on the master level relevant to the establishment of sustainable business models meeting user requirements, including design, development and deployment of pilot services and networks.

• Review of the telecommunication policies and regulatory framework to outline a regulatory environment supporting a secure, dynamic and economically sound development in the area of telecommunications based on the view that the

Problem-Oriented Capacity-Building based on the Introduction of ICTs | 59

telecommunication infrastructure is a utility of the same importance for society as roads and railways for transportation, power distribution, water distribution, etc.

• Business development in organisations providing basic public services in rural areas, such as public administration, healthcare, education and support for local entrepreneurs and civil society. The communication needs of such organisations are likely to generate a substantial traffic volume forming the basis of a sustainable communication market in rural areas which will also benefit private sector.

• Research including analyses of the impact of ICT by socio-economic studies before and after the introduction.

The deliverables would include • Research, education and technology transfer to societies at selected universities,

including curricula and capacity to conduct programmes and/or courses based on problem-oriented, project-driven learning supporting development of rural areas based on ICT.

• Increased capacity to conduct socio-economic development projects, in terms of educated individuals and know-how related to ICT as an infrastructure for development.

• Pilot systems, services and networks illustrating best practices and providing services in selected rural areas in healthcare, education, support to local entrepreneurs and public administration.

1. Background

Universities have three major tasks: Research, Education and Community service. Theideas conveyed in this paper are relevant for all these tasks although our focus is set on how institutions of higher learning can serve as an engine of socio-economicdevelopment in their environments. The challenges involved include increasing theaccess to education as well as improving the quality of the research and educationactivities and increasing their relevance to national needs.

A bi-modal approach is necessary: a bottom-up approach to improve basic livingconditions like basic health, primary education, access to water, etc, and a top downapproach focusing on higher education, technical development and industrial linkages.Information and Communication Technologies (ICT) are important, both as aninfrastructure for organizational management, research and education in general and asan academic area of research, education and industrial development relevant to thenational needs. The development framework should include cross cutting themes, suchas strengthening the governance and management, introducing quality assurance andinfrastructure development, as well as core themes, such as faculty development,improving access & learning, excellence in research and relevance to national priorities. Within this development framework, proposals for Development Projects are solicitedsupporting the development goals as well as concrete project goals relevant to theprogram, activities, timelines, milestones, deliverables and budgets.

This paper discusses a development model in the ICT area taking the top downapproach to capacity building activities supporting the bottom up approach in terms of


introducing ICT as an enabler for rural development, focusing on healthcare, schoolsand public administration.

In section 2, a framework for problem-oriented, project-driven research, developmentand learning, based on a combination of peer learning, vicarious learning and coachingis described.

In section 3, a process for introduction of ICT in a developing country is described.

In section 4, the R&D framework described in section 2 and the development processdescribed in section 3 is combined and instantiated with local data.

2. Problem-oriented capacity-building

In this section, we describe a framework for research, development and learningdeveloped at KTH in cooperation with partners from all over the world, includingStanford University, National University of Singapore, and University College, UK.

2.1 A framework for problem-oriented research, development and learningThe framework is based on problem-oriented, project-driven learning in distributedteams, considering geographical distances and time zones, as well as organizational andcultural differences. The teams could be focusing either on entrepreneurial start-upventures or consulting for external project-owners.

The components of the framework include:• Work processes in the teaching team, the project team, and external stakeholders as

well as discipline/problem-oriented content in designated areas.• Pedagogical support systems based on problem-based, project-driven learning, peer

learning among team members and vicarious learning from teams that have dealt with similar situations previously.

• Technical and spatial support systems in terms of distributed interactive learning spaces supporting audio/video conferencing with shared work spaces, data sharingand connections to mobile team members.

The framework has its roots in the work described in sections 2.2-2.5

2.2 Experiences from existing programsExperiences from existing master level programs taught in English with internationalrecruitment in the area of Internetworking (www.it.kth.se/intms) and ICTEntrepreneurship (www.it.kth.se/entms), both including a course in ICT System Design fitting the description above can be found at: (http://csd.ssvl.kth.se username=guest,passwd=welcome).

2.3 Pedagogical modelsTheory and practice of the pedagogical models mentioned can be found by visiting the following websites: • Problem-based learning: Students learn when motivated by problem-solving to

achieve a task (www.udel.edu/pbl/, http://interact.bton.ac.uk/pbl/ )


• Peer learning: Students with different competences, experiences and cultural backgrounds learn from each other(www.questia.com/popularSearches/peer_learning.jsp)

• Vicarious learning: Students learn by studying the learning challenges and results of students in earlier projects (www.informatics.sussex.ac.uk/research/appcog/vl/index.php)

2.4 Research on interactive learning spaces The research on interactive learning spaces was initiated at Stanford(http://iwork.stanford.edu/) and followed up by KTH, a few German universities andrecently also Ventspils University College (www.balticopen.net). KTH has developedand used interactive learning spaces the last decade and helped establish interactivelearning spaces at several universities in different parts of the world.

2.5 Research on support for mobile learners Research and development of wireless access and support for mobility, includingmobility of terminals and sessions, have been conducted at KTH for many years. Theresearch has been taken into development projects resulting in spin-off companies aswell as a leading edge learning environment extended from the KTH campus tohotspots in the city within the StockholmOpen.net project, one of the pilot networksimplementing an open network architecture discussed in the next session.

3. ICT as an enabler of development of socio-economic development

In this section, we describe our approach to the introduction of ICT in developingcountries in different stages, based on our experiences from several countries on Asia,Africa and Latin America, including Bangladesh, Laos, Mozambique, Tanzania, Namibia and others.

3.1 Computer systems and communication networksICT includes two major components:• Computer systems used for a wide spectrum of information processing tasks,

knowledge management, and communication services, etc. • Communication infrastructures, transmission links and networks.

In the developed world, ICT has proven to be a powerful enabler of development,improving both social and economic parameters, improving the quality of life ofindividuals as well as the efficiency of organizations and productivity. Although some of the involved technologies have had an explosive development, the impact onproductivity statistics has been slower since the large scale benefits require changes inhuman and organizational behaviour including the way business is conducted. There isno doubt, however, that ICT is a very powerful enabler of socio-economicdevelopment.


3.2 Access to computersComputers are still expensive. Making computers available seems, however, a goodinvestment in socio-economic development. Access to computers in schools, Internetcafes and telecenters is a first step towards a larger pc-penetration.In some developed countries, the penetration is very high. In Sweden the pc-penetration in homes is over 70%, a development that has been supported by a taxreduction motivated by the government’s and parliament’s recognition that ICT is anenabler of development.

3.3 A national communication infrastructure An infrastructure for communication, including interactive voice (telephony), video and data communication, as well as broadcast media, is as important for the development of society as infrastructures for transport, including roads, railways, airports andinfrastructures for power and water distribution. As opposed to the latterinfrastructures, the communication infrastructure has not until recently received thenecessary attention of the public sector. Industry has made us believe that it can provide the services needed on a strict commercial basis. This has turned out not to be true.Even in developed countries, the public sector has had to support the basicinfrastructure to facilitate the development demanded by their citizens, at least on thebroadband side.

3.4 StakeholdersThe main actors on the communications market are the users buying the services thatthe service providers are providing. For several reasons, the picture is however muchmore complex. The user side is composed by different kinds of organizations as well as individual citizens with different interests and power. The service provider side isdivided in broadcast media operators, voice service providers operating on fixed ormobile phone networks, Internet service providers, etc. The balance between thedifferent actors is uneven and some of the basic resources, such as radio spectrum, arescarce, which calls for a regulator. We have chosen to identify the following stakeholder groups:

Policy makers and regulatorThe policy makers define the market rules from a political perspective in terms oflegislation and directives to the regulator. The regulator implements the policies set by the policy makers in terms of regulations, licenses, permits, etc.

Good policies and regulations promote a dynamic development of society, which isoften best accomplished by paying a balanced attention to the needs of all stakeholderson the communication market. As already mentioned, there is, however, a shift in theunderstanding of how the market is best supported giving public sector a moreimportant role than earlier in the establishment of a national passive infrastructure.

Another shift in progress is the deregulation of vertical monopolistic operators thatoperate everything from the passive infrastructure, transmission, networks, services and user access. The new paradigm promoted is a market based on horizontal competitionwhere several different operators compete on one level but where operators on onelevel do not compete with their customers on the next level.


An increasing number of policy makers take the position that communication should be regarded as a utility and be provided in abundance as a sustainable business, not as ascarce resource exploited for maximum profit.

User agentsThe user side of the communication market is a mix of large organisations andcorporations, small and medium sized enterprises (SMEs) as well as individual citizens.The large organisations can match the service provider side and influence the market alot while the small enterprises, civil society and individual private users can not.

It is therefore useful to look for user agents, i.e. large organisations that are neutral, oreven biased towards the users, to improve the balance. The regulator should be neutralbut still promote development on both sides. Public sector is paid by taxes and shouldwork for the tax payers. In some countries there are strong consumer organisations that can play a role. In Sweden, the apartment housing sector plays an important rolestandardising and pushing a fibre to the home concept which is very user-oriented andmaking it possible for tenants to select services from different operators independently from each other.

Public sectorTo build a sustainable communication market, the public sector plays several important roles, as a policy maker and regulator, user and user agent and as an owner and/orpromoter of the necessary passive infrastructure.

As a regulator, the public sector controls how the passive infrastructure can be used to the benefit of the society.

As a user, the public sector agencies, belonging to central, regional and localgovernment generate a substantial part of the total traffic, normally 20-40% of the total traffic, in phases of development often much more. The communication needs for thepublic sector is thus a key component when building a sustainable communicationmarket.

As an owner/promoter of passive infrastructure, the public sector can often include awider spectrum of values when calculating the return on investment than private sector companies.

IndustryIndustry in general has a role as user. As such, large companies often build their ownprivate networks for reasons of security, etc. To build a sustainable market, it isimportant to be able to provide the quality of service that the large users can accept atlowest cost.

ICT industry includes system manufacturers and integrators on one hand and operators on the other.


Civil societyThe civil society includes associations of private citizens with common interests ofdifferent kinds, such as athletics and sports, culture, religion, etc. This sector isimportant to the level of quality of life in a society but mostly financially weak. Civilsociety needs to piggy-back on resources provided by the larger public and privatesector organisations.

3.5 What is included in the Communication infrastructureThe different levels in the communication network architecture are described below and the market actors on the different levels are discussed. In general, in areas where themarket is small, the public sector may have to operate active links and optical networks. As the market grows one or more competing private companies can take over.

The passive communication infrastructureThe passive infrastructure includes:• Radio spectrum and physical locations for antennas on the wireless side• Right of way, ducts, fibre, physical spaces for termination and access on the wired

side• Regulatory frameworksThe passive infrastructure is a national asset. To support a dynamic development ofsociety, it should be managed and made available at lowest possible cost, or evensubsidised in development phases. To avoid inefficiency, parallel infrastructuresmanaged by competing organisations are preferable. In Sweden, there are manyindependent national fibre infrastructures owned by both public and private actors, such as Banverket (railway), different telecom companies, different power distributors, etc.Users are allowed to lease dark fibres.

Transmission linksThe transmission layer can include purely optical communication links and networks(CWDM/DWDM) with an optical interface to the user, or digital links with an electrical interface to the user (SDH/Ethernet). State of the art long haul links are now in the 2.4-40 Gbit/s SDH or 1 or 10 Gbit/s Ethernet. Users can get all or fractions of this toconnect their own network segments or to a service provider. Operators can use suchlinks to connect to each other, directly or via a traffic exchange point, to exchangetraffic.

IP-network and Operator servicesOn the network layer, Internet Service Providers provide IP-services connected to aservice provider or to the Internet. Network services include domain name service,time service, route servers, etc.

End-user servicesThe most common end-user services include email, web access, remote login, backup,databases, information services, web hosting, etc. These services are normally providedby Internet Service Providers (ISPs).


3.6 Open network architecturesOpen network architectures are designed to allow different actors to build networkstogether by connecting different network segments, such as access networks, servicenetworks, long haul links and traffic exchange points. To facilitate this, the differentsegments need open standardised technical interfaces and business models facilitatingcooperation. The advantage is that different actors may have different possibilities tocreate business in a specific area. For example, local entrepreneurs might seeopportunities to establish a sustainable business, which no national operator can takeadvantage of. In that case the local entrepreneur should be able to build an accessnetwork taking his users to a point of presence of the national operator, rather than the opposite.

3.7 The value ladder of public ownership of passive infrastructureThe investment required when deploying a passive fibre infrastructure, including ducts, fibre cable, termination spaces, etc, is in the order of 100 USD per meter. This cost can be reduced if the deployment can be made together with other infrastructures, such aswhen extending the power grid, building roads, along railways, pipelines, etc.

Step 1: Max return on investmentThe basic instinct of an investor is to maximize the return on investment to recover it as soon as possible. The strategy is to find an optimal price to maximize the profit.

Step 2: Cut own communication costsIf the price for leasing fibre is lowered and the regulatory framework is based on anopen regime, more operators are attracted, which stimulates competition and lowers the communication costs.

Step 3:Increase tax revenues and cut costs for unemploymentIf the price for leasing fibre is lowered even more, existing companies can improve their performance and new companies can be created, which increases tax revenues and cuts costs for unemployment.

Step 4: Reduce costs for social welfare and crimeIf the price for leasing fibre is made nominal, individual citizens and civil society willflourish, which potentially increases the average quality of life and lowers a widespectrum of social costs.

3.8 Business development of public servicesThe normally highest prioritized public services include healthcare, education, publicadministration, including support to local entrepreneurs. In many countries, these areasare considered as core businesses for the public sector. It is, however, not always thecase that these services are provided by the public sector. In some countries they arerather dominated by the private sector or by the civil society as commercial companiesor non-profit organisations. The form of ownership and funding is not important in this context. However operated the services mentioned are considered basic, both indeveloped and developing countries.


Rural healthcarePrimary healthcare centres in rural areas mostly have too limited resources in relation to the services they are expected to provide. There is mostly only one doctor, a few nurses and sometimes paramedic staff and a business manager. The doctor is mostly un-experienced in the medical profession, since more experienced doctors compete formore prestigious and resourceful positions in the urban areas. The number of patientsthat belong to the centre is often very large. The infrastructures for transport of patients that need more resources, and for telecommunication to provide expert support to rural doctors, are often poor and sometimes even non-existing. Few staff members, if any,have had any sort of contact with ICT during their training. The business managermight be an exception.Basic ICT systems that are useful to introduce in such environments include:• Computer systems supporting patient records, drug inventory and business

management procedures such as HR management including payroll and financial management

• Communication systems supporting email, voice and video conferencing for consultation with medical experts at secondary district hospitals or national referral hospitals, as well as access to literature databases and WHO information, etc

When introducing ICT in the healthcare process in rural areas, the challenges, besidesthe business development necessary to take advantage of the potential, also includeaccess to power, a communication infrastructure and staff that could be trained to beable to take responsibility for the sustainable operation of the ICT systems.

EducationThe education system normally includes primary and secondary schools and tertiarylevel universities and colleges, including teacher training colleges.

The challenges when introducing ICT in educations are different on the different levels but common issues include:• Providing access for teachers and students• Developing the curricula• Make learning material available via ICT• Training teachers in the use of technology and new pedagogical models

Although it is important to invest heavily in teacher training, it is not advisable to waituntil the training programmes are complete. In Sweden, the introduction of ICT inschools was delayed 10-20 years due to the strategy to make the teachers the agent forthe introduction. It was not until ICT became widely available in homes that thedevelopment accelerated. The students brought the usage to the schools rather than the opposite.

Public administrationCommon starting points when introducing ICT for business development in publicadministration include email and voice services, internal procedures for HR andfinancial management and external services to the citizens such as providing forms.


Support to local entrepreneursEntrepreneurs in general can benefit from Internet access to find business informationand market products and services. Access for such purposes can be provided byopening up schools for the local community during off-hours. Entrepreneursestablishing themselves as local network or service providers need an open regulatoryframework, access to appropriate technology and technical solutions, education andtraining, financing, etc.

3.9 How to build a sustainable marketSummarising this section we identify four steps:

Research and development as spearheading demonstratorsThe first step is to use concrete examples to demonstrate and convince the policymakers, regulators and users about the power of ICT as an enabler of development.Bottom up local entrepreneurs and top down academic research groups can unite to do this as research and development projects. Commercial aspects can wait a bit.

Public services to build volumeThe basis for a commercial market can be made by business development of publicservices to exploit the benefits of ICT. Procurements of communication services willactivate private sector, including local entrepreneurs. Schools can serve as telecenters off school hours.

Stimulation of private sectorIn a development phase, there are opportunities for local entrepreneurs. To make itpossible for these entrepreneurs to take advantage of these opportunities, by makingcomputers and network access available in all ways possible, will create new livelihoodsfor many.

Towards a better quality of lifeMaking communication services cheap will facilitate for the civil society to satisfy allsorts of needs of the citizens.

4. A Development Project model

In the subsequent sections, we outline a vision, goals, activities and deliverables for adevelopment model in the ICT area in a developing country. The model is being tested in practice in projects in Africa and South East Asia.

The project would combine the establishment of the research, development andlearning framework described in section 2 at selected universities and the ICTdevelopment process described in section 3.

4.1 VisionThe vision of the project would be to establish sustainable broadband markets in ruralareas. The strategy is to focus on the following development objectives:


• Demonstrate feasibility and build capacity in terms of human resources by non-commercial research, development and education in a neutral and competent academic setting.

• Adopt open policies and regulatory frameworks promoting open access to infrastructure transmission services and consumers to support user communities as well as new service providers and network operators.

• Have public sector invest in backbone infrastructure to establish basic public services accessible in rural areas, focusing on those that are supporting progress towards the Millennium Development goals:

o Education using e-learning applications and community access to support local entrepreneurs by making schools available as telecenters

o Healthcare, including email and access to journals for doctors,, patientrecords and telemedicine applications allowing inexperienced doctors and paramedics consult specialists at district and referral hospitals

o Public administration, such as email for civil servants, support forenvironment impact analyses, GIS-applications, support to localentrepreneurs, etc

o By generating a substantial traffic volume and an open accessinfrastructure, public services will provide the basis for a sustainable andexpanding broadband market

• Stimulate local business, households and civil society, to develop services and use broadband applications to create new livelihoods and conduct business more efficiently.

4.2 Approach1, 3, 5 and 10-year goals and a project plan will be detailed in a planning project in theR&D&L framework described in Section 2 during the academic year 2005-2006involving all stakeholders.

As outlined in section 3:• Use existing public sector initiatives in education, healthcare and public

administration to formulate requirements and business models. • Involve user communities and organizations to describe their applications and their

requirements as a basis for procurements• Identify services and access networks that local entrepreneurs can provide in a

sustainable way to meet the user requirements• Identify the support that the local entrepreneurs need, such as technology,

education, funding, regulations, etc • Define the requirements on a national infrastructure making it possible for the local

entrepreneurs to build their access networks • Select a few pilot applications from existing initiatives, in education, healthcare and

public administration• Include typical requirements for broadband Internet services that cannot be

provided via gprs or pstn networks.


4.3 ActivitiesThe general methodology used in the area of communication systems design is aniterative model involving all stakeholders with cycles of design, implementation,evaluation, redesign, etc.

The first cycle includes identifying all stakeholders and forming and consolidatingsteering and reference groups and a small pilot implementation on a carefully selectedarea that is considered to be representative but not too complex. In subsequent cycles,the number of areas is increased in order to identify variations that need to beaccommodated in a full scale program.

The activities in each cycle include learning, design of system components, procurement of equipment, implementation, testing, training of maintainers and documentation,dissemination and evaluation of results.

The capacity building is accomplished by recruiting new team members in each cycle,while team members participating in an earlier cycle can be involved as coaches or other roles in subsequent cycles.

4.4 DeliverablesThe deliverables include a development programme with development goals in a short-term, medium-term and log-term perspective, a plan for the establishment of aframework for research, development and learning as described in Section 2 a pilotdemonstrator for the use of ICT in rural healthcare, schools and local communities. The deliverables also include reports, presentations and demonstrations.

5. Experiences

Although the problem-oriented, project-driven learning framework has been developed in a global setting during a decade, the international development cooperation projectshave been conducted only the last four years.

5.1 Access networks and servicesAccess to computers and networks, in particular the Internet, is still one of the majorbottlenecks in the world. Since universities play an important role in the capacitybuilding in under-served areas, they are a priority. The finally get to the rural areas, thewhole communication chain needs to be considered, from central government toprovince, district and village administration, from referral hospitals to district hospitalsand rural health centres, from ministry of education to teacher training colleges andschools.

There has to be interplay between top-down and bottom-up activities. The followingproject examples illustrate both sorts.

Universidad Eduardo Mondlane (UEM)In 2002, a mixed team of students from UEM and KTH connected the UEM studentdormitories an operator neutral access network allowing several Internet Service


Providers to provide services to the users, the academic network being the first(http://csd.ssvl.kth.se/~csd2002-mozambiqueopen).

National University of Laos (NUOL)During the period 2003-2005, KTH, NUOL, NAFRI (the National Agriculture andForestry Research Institute of Laos), and STEA, the Science and Technology andEnvironment Agency of the Lao Prime Ministers Office in Laos, are cooperating toestablish campus networks with a wide spectrum of services, for NUOL and NAFRI.The cooperation also includes the planning of a national education and researchnetwork in Laos and an interconnection to the emerging regional academic backboneTEIN2 supported by EU. (http://csd.ssvl.kth.se/~csd2003-team1,http://csd.ssvl.kth.se/~csd2004-team4, http://csd.ssvl.kth.se/~csd2004-team5,http://csd.ssvl.kth.se/~csd2004-team6, http://csd.ssvl.kth.se/~csd2005-team1,http://csd.ssvl.kth.se/~csd2005-team2).

Vientiane and Maputo Gigabit NetworksThe Vientiane Gigabit Network grew out of the KTH-NUOL-STEA cooperation in2005 to provide a showcase high-performance city network as a first step towards anational networking effort involving e-governance, e-health and e-schools. The Maputo Gigabit Network grew out of a similar context in Mozambique.

Dar es Salaam Institute of Technology (DIT)In 2004, students from DIT and KTH established a campus network at DIT, including a fibre infrastructure, switches and routers and servers, providing basic Internet services to DIT students and staff (http://csd.ssvl.kth.se/~csd2004-team7).

Rural Access to ICT in TanzaniaIn 2005, the first phase of a development programme for rural access to ICT was taken by forming a steering committee with central stakeholders, including policy makers, the regulator, public sector development agencies and private sector service providers. Toconcretize possible projects in a larger scale programme, a few minor projects relatedwhere conducted also in the first phase (http://csd.ssvl.kth.se/~csd2005-team5).

Malaria Research Centres in AfricaThe Multilateral Initiative of Malaria consortium has, via its communication armMIMCom, cooperation with KTH to connect malaria research centres to the Internet to facilitate their access to information and colleagues. International students at KTH have been involved in deploying access networks for such institutes in Congo Brazzaville,Ghana, Malawi, Mozambique, Nigeria, Uganda (http://csd.ssvl.kth.se/~csd2004-team8,http://csd.ssvl.kth.se/~csd2005-team6).

Rural Health Centres in BangladeshIn 2005, a first phase of an e-Health and telemedicine strategy was developed indiscussion between KTH and the Grameen group. An experimental implementation isplanned during 2006 (http://csd.ssvl.kth.se/~csd2005-team7).


5.2 Internet Exchange points (IXPs)Internet exchange points make it possible for different local Internet Service Providersto exchange local traffic locally, rather than using expensive up-streams transitbandwidth, often via satellite, to send local traffic abroad and back again. So far, sixsuch IXPs have been deployed since 2003 and more are being planned.

MozixIn 2002, students from UEM and KTH designed and deployed an Internet exchangepoint in Maputo as one of the first in Africa. Mozix is a clear success and all ISPs,except the incumbent, are connected. Mozix was extended with new services in 2005,(http://csd.ssvl.kth.se/~csd2002-mozambiqueix, http://csd.ssvl.kth.se/~csd2005-team3).

LaonixLaonix was established in Vientiane in 2004, making it possible for Laos to fully exploit their domestic fibre infrastructure although the international bandwidth is veryexpensive due to monopolistic policies still in place. All ISPs except the incumbent areconnected (http://csd.ssvl.kth.se/~csd2004-team3).

BolixThe first IXP in Bolivia was established in 2004 in La Paz by a team of students fromUniversidad Mayor de San Andrés (UMSA) in La Paz and KTH. Unclear policies andprivate business strategies have prevented several of the ISPs to connect(http://csd.ssvl.kth.se/~csd2004-team1).

NicixNicix was established by two KTH students together with students from UniversidadNacional de Ingenieria (UNI) and staff members of ISPs in Managua, joining acontinuing education programme. The business climate among the Internet ServiceProviders has prevented several ISPs from connecting(http://csd.ssvl.kth.se/~csd2004-team2).

RinexRinex, the IXP in Kigali, Rwanda, was established 2004 by a team of students fromNational University of Rwanda and Kigali Institute of Science, Technology andManagement and with support from the Rwanda IT Authority (RITA). Theinauguration of Rinex has been an important milestone in the very active ICTdevelopment in Rwanda. All ISPs are connected and some of the students involved inthe design and deployment are now operating the IXP.(http://csd.ssvl.kth.se/~csd2004-team21).

NIXThe IXP in Windhoek, Namibia, was established in 2005. In the process, details aboutthe design of existing network and the current regulatory framework was revealed thatillustrate the need for more open communications policies in Namibia, if a moredynamic development of the society is wanted (http://csd.ssvl.kth.se/~csd2005-team4).


Malawi, Zambia, BurundiIXPs are currently being planned in Lusaka, Blantyre and Bujumbura and will deployed early 2006.

5.3 Regional backbonesThe backbone networks are part of the infrastructure of a society and should be treated by public sector as roads, railways, etc. This is generally not the case. Although themobile phone networks are growing rapidly in many developing countries, theirbackbones are mostly built using microwave technology that does not support thedeployment of broadband networks, such as fibre.

The Sarua-fibre project [Sarua] is making a case for a regional academic broadbandbackbone in Southern and East Africa implemented in a way that can be exploited alsoby commercial networks.

6. Conclusions and plans for development of the model.

The experiences from using the problem-oriented, project-driven framework forresearch, development and learning described in this paper are very encouraging. Themodel can be refined to improve the selection of projects to synchronize better withnational and international strategic development programmes and human resourcedevelopment programmes.Ideas for development of the model include• Extension of the scope of capacity building to include not only how to design,

deploy and operate broadband Internetworks and services, but also power generation and distribution , ICT Entrepreneurship and public sector business development.

• Cooperation with the Stockholm Challenge Award process (www.stockholmchallenge.se), a global evaluation of projects for development of under-served regions, in order to improve the screening process to find high-quality development efforts to support.

• Designing a benchmarking process based on a set of indicators reflecting the maturity of a region concerning the use of ICT.

• Coordinating efforts defining requirements of public sector business development, technical solutions, supporting entrepreneurs and analysing the impact of ongoing development efforts in a common framework serving as a think-tank for communities planning development efforts towards the network society.

7. References

Beside the web references providing earlier in the text, a few examples of earlier projects are described in appendices below by the project teams. More details about projects that have been conducted the last four years are available at http://csd.ssvl.kth.se/. Theappendices include:


[1] Godfrey Chikumbi: Fight against Malaria in Africa through ICT by MIMCom

[2]Eneas Hunguana, Eneas Hunguana, Qarin Hjortzberg-Nordlund, Alberto Muchanga, Erik Stackenland, and Jon Åkergården: MzOpen.Net: Wireless hotspots in UniversityResidences in Mozambique.

[3] Issa Nkusi, Claude Hakizimana, Coco Musaningabe, Innocent Nkurunziza: RINEX -Building a Bridge to Reduce the Digital Divide and Enhance the Use of ICT as a Toolto Eradicate Poverty

74

Appendix I

Fight against Malaria in Africa through ICT by MIMCom

Godfrey Chikumbi

1. BackgroundMalaria is a severe problem in many developing countries, especially so in the Sub-Saharan Africa, where the climate and environment is conducive to the spread ofMalaria. Multilateral Initiative on Malaria, MIM is an international alliance oforganizations aiming to strengthen African malaria researchers. The MIM secretariat iscurrently hosted by Karolinska Institutet and Stockholm Universitet in Stockholm,Sweden. As one of its main activities, the alliance has worked to connect research sitesto the Internet and to promote innovative usage of this connection, through itscommunication arm MIMCom. It believes that;

"We must develop a communications system so that the miraculous triumphs of modern science can betaken from the laboratory and transmitted to all in need." .. Senator Lister Hill, 1965

Currently 20 sites across Sub-Saharan Africa have been connected. The MIMCom-2005project aim was to provide or upgrade Internet connectivity at four locations in theAfrican continent, as well as to provide or improve their internal networks. The projectwas funded by Sida – the Swedish International Development cooperation Agency, and executed by students at KTH – the Swedish Royal Institute of Technology.

2. Objective

The MIMCom project was aimed at providing and upgrading Internet connectivity atmalaria research centres spread across four countries. Improved Internet access willenable African scientists to communicate freely, seek and retrieve scientific data andobtain timely information regarding grant opportunities as well as setting up morecomplex IT-based networks to forward their research agenda, all of which are vitalprerequisites for successful participation in and contribution to the global researchcommunity. Centres to be connected were:• Institute of Child Health (ICH), College of Medicine, University of Ibadan, Nigeria• University Training Hospital (UTH), Lusaka, Zambia• Mulago Hospital, University of Makerere, Kampala, Uganda• Centre d’etudes sur les Resources Vegetales (CERVE), Brazzaville, Congo

Brazzaville.

Appendix I: Fight against Malaria in Africa through ICT by MIMCom | 75

2.1 Primary GoalTo provide each site mentioned above with the following:• Stable Internet connectivity – at least 64kbps• System Operation Training to the local staff so that they can keep the system

working smoothly.

2.2 Secondary Goals• To provide or improve internal networking (LAN or WLAN) at the respective

sites.• To provide a server and/or document repository at each site for archival and

versioning.• To provide VPN connectivity across the sites in order to facilitate information

sharing and collaboration.• Setting up an internal IP-telephony system• Setting up an IP telephony system across the four sites• Checking availability of appropriate bio-medical software that can be used at the

various sites – and deploying the same.

3. Site-specific deliverables

3.1 Mulago Hospital - Paediatrics Department, Kampala (Uganda)The Mulago hospital is the largest referral hospital in Uganda and is located in Kampala, the capital. It maintains a symbiotic relationship with the Makerere Medical School.Most of the members of the malaria research project are in a single-story building that is part of the Makerere University's Paediatric department and is based at the teachinghospital. The building had one computer connected to the WAN of the university andabout 12 stand-alone computers. The malaria group wanted to have an independentsystem which they could control and identify with other research groups on theMIMCom network.

The requirements were:• Internet access for 12 computers• LAN installation• Server to handle mail, proxy• Other equipment: printer, scanner, digital camera

3.2 CERVE, Brazzaville, Republic of CongoCERVE _ Centre d'Etudes sur les Resources Vegetales _ is a research centre located in the Orstom forest area / science park in the capital, Brazzaville, of the Republic ofCongo. The site is the base for the Malaria Research activity.

The following specific requirements to CERVE were finalized after studying the sitereport, feedback from the principal as well as researchers on the site:• Internet Access for approximately 20 PCs • LAN Installation so that all PCs can be connected to the Internet• Printer and Scanner


To meet these requirements, we also decided to provide a server (to act as a Web Proxy as well as Email server) which would interface CERVE to the Internet. In addition, wealso needed an uninterruptible power supply (UPS) with sufficient backup capability due to the poor electricity supply to the site.

Choice of Internet Connection TechnologyBrazzaville has a fixed line telephone network, however it is very unreliable, and mostpeople rather use cell phones connected to one of the two providers. Also, as theOrstom Science Park is located inside a forest with very tall trees, getting a line-of-sightto the local ISP (Celtel) via Wireless (locally known as BLR) is also very difficult. Due to this, the only viable option was to use VSAT (Satellite) connectivity. OFIS andAITECH both were able to provide VSAT connectivity as well as LAN for CERVE.OFIS suggested a C-band solution, providing 1024 kbps downlink / 128 kbps uplinkbandwidth. The upstream provider is GeoLink, Italy. AITECH suggested a Ku-bandsolution, providing 1024 kbps downlink / 76.8 kbps uplink bandwidth. The upstreamprovider is Global T&T, Belgium.

Services ProvidedBecause of the lack of skilled IT staff at CERVE, we took a decision to provide onlythe most basic services to keep system administration overhead to a minimum. Thefollowing services were provided:• Transparent Web Proxy (Shorewall)• NAT + Firewall (Shorewall)• DNS Cache (Bind)• File Server (Samba)• VPN Server for remote access (not yet fully working) (PoPToP)• Webmin for easy-to-use-administration• DHCP server with as much auto configuration as possible (ISC DHCPd)• UPS Auto-Shutdown on low-power (apcupsd)

Problems Encountered• One of the first problems encountered was the choice of vendors. While in

Sweden, we decided to go with OFIS for the Equipment as well as the Internet connectivity. However, on reaching the site a lot of new information surfaced, due to which there was a last-minute change in the vendor for providing the internet connectivity.

• An ongoing problem was the erratic electricity supply, which often caused the installation work for LAN to be delayed. Normally, there was no electricity supply at least 2 days every week.

• Language problems: Congolese are francophone, and few people speak English. This got better over time, but still caused problems.

• Money Transfer issues _ There were some problems with money transfer to the VSAT provider (Global T&T in Belgium) and to AITECH (Brazzaville) for the LAN installation, due to which the work was held up for quite some time.

3.3 Institute of Child Health (ICH), Ibadan, Nigeria


ICH is a research centre as well as a children hospital located within the College ofMedicine, Ibadan, Nigeria. The Institute was established over four decades ago for thestudy of the problems of child health relevant to this environment. Over the years, theInstitute has, in collaboration with other institutions and departments, conducted a wide variety of laboratory, clinical and applied field research activities.

RequirementsThe primary requirement is to increase the bandwidth at ICH, either by providing a new connection or by boosting one of the existing connections.

Services ProvidedThe Ku-band solution was used to provide the upgraded internet connectivity. We used a gateway software called Clarkconnect in the server which will make the management of the network easy once everything is configured. Clarkconnect is a Linux-basedoperating system that transforms standard PC hardware into either a standalone serverin the network or a dedicated firewall /gateway to a network. The services providedwere:• Stateful firewall• Bandwidth management• VPN +Dynamic IP support• Content filter• Intrusion prevention• Resource and port monitoring• Mail server, Mail/Mx backup service• File server• Print server• Transparent proxy server• Web server• Webmail• Database• Antivirus and Antispam• System monitoring, to know when a system is offline

ProblemsWhen the VSAT was installed there was a huge latency of about 1100 ms which isunusual and we had to call a spectrum engineer to analyze the problem. He later fixedthe latency to 650 ms which is typical for satellite networks. One of the biggestproblems we faced was transfer of money and foreign exchange. Because of this, a lotof time was wasted. In Nigeria most organizations and companies are not allowed toreceive foreign money.

3.4 University Teaching Hospital (UTH), Lusaka, ZambiaThe University Teaching Hospital (UTH) is a 2000 bed tertiary care hospital located inLusaka, the capital city of Zambia. It is the major referral hospital in Zambia. It alsoserves as a District Hospital for Lusaka. Its intended function is to provide tertiary careto patients referred from clinics and hospitals in Zambia. However, due to lack of


secondary care provider institutions in Lusaka province, UTH also functions as aprovider of secondary and primary care for the district and province. In-patient services are provided to approximately 3,000 patients per month and outpatient visits areroughly twice this number. The University Teaching Hospital is situated approximately6 km south of the main campus, within the College of Medicine. The movement ofpatient information to the UTH is a facility that has received less attention than inpatient registration thereby leading to lack of information on patients outside and within UTH. For many years, University Teaching Hospital has provided health care servicesin deferent research units to the people of Zambia. The Malaria research unit is one of the services provided by the hospital.

RequirementsThe malaria research unit of the University Teaching Hospital (UTH) required increased scalable storage for its record archives but had no money to invest. With an alreadyextensive archive of patient histories and new patients being processed daily, the UTHmalaria research unit was in critical need of a storage system. The basic requirementswere:• Internet Services.• Guarantee of Power supply and backup.• The server for proxy, mail and DHCP• LAN and Campus connectivity between various buildings.• Firewall and Bandwidth Manager.• Further training for the staff in order to sustain the new systems.• User policy enforcements to be implemented.

Chosen Solution and ImplementationIt was agreed that the wireless solution proposed by MicroLink Zambia would be used to provide Internet to UTH Malaria research centre, using 5.6 GHz frequency band.The Dell PowerEdge 2600 server was used to act as gateway, mail, web, dhcp and proxy server. The Site required a big server in order to establish a Health Informatics Centre. The Health Management Information Systems both at the UTH and the clinics werevery weak, leading to the creation of an information vacuum in the health facilities. The manual collection of information and its compilation made it very difficult to obtainquality and useful data that could be utilized for planning and monitoring of patientcare, which is essential in guiding policy formulation and implementation of healthservices. The Fedora core 3 operating system was used and a lot of server freeware wasused, such as sendmail, Apache, Zebra, usermin, webmin, etc.

ProblemsThe whole project implementation took six weeks. The only major delay was due to the late delivery of the server; however everything was done according to the schedule. ADell PowerEdge 2600 server was used while waiting for the Dell PowerEdge 2800server. Apart from the funding for further training for IT staff, everything requested by the site was provided and implemented.

4. Stakeholder Analysis

4.1 MIM and MIMCom


The Multilateral Initiative on Malaria – Communication arm (MIMCom) expected us to improve Internet connectivity at Malaria research sites in Africa by implementing themost appropriate technology that is fast, reliable and sustainable. This is achievedthrough a partnership between MIMCom and the research sites, which ensures that allkey site conditions are taken into consideration, like financial resources, Internetinfrastructure if any, technical support availability, etc. MIMcom also expected us toassess the existing computing capacity at the site, including the network infrastructure(LAN) if one exists, how best to improve it as well as talking to the respective systemsoperators. MIMcom made sure that IT staffs were trained at the sites, so that they cantake care of any problems that arise. MIMcom extended its full support in all matters of importance to the project.

4.2 SidaSida, the Swedish International Development Cooperation Agency, is a governmentagency that reports to the Ministry for Foreign Affairs. Sida is responsible for most ofSweden's contributions to international development cooperation. In 2004, thecontributions amounted in total to SEK 21 751 million. Sida's task is to work toimprove the living standards of poor people in the third world nations. The Department for Africa has overall responsibility for cooperation with countries in sub-SaharanAfrica as well as regional support to Africa. The Department formulates and proposescountry and regional strategies to the Swedish Government and coordinates thedevelopment co-operation in line with these strategies once decided on by theGovernment. Sida has committed funds to this project as a part of cooperating in theGlobal Projects with International Organizations (MIMCom).

4.3 KTH / TSLABKungliga Tekniska Högskolan (Royal Institute of Technology) is responsible for one-third of Sweden’s capacity for engineering studies and technical research at post-secondary level. The IT University at Kista, cooperation between Stockholm University and KTH, has various engineering courses offered, including ICT Development. About 100 students from the IT University take part in the “Communication Systems Design” course offered by the Telecommunication Systems Lab (TSLAB) at KTH each year.

5. Lessons Learned

We had to deal with a lot of obstacles that do not appear in academic environments,such as state regulations, nepotism and not diligent providers, so we learned thatgrowing a project outside is much more complicated than at the university and it isneeded to deal with a lot more bigger obstacles, usually of a non-technical kind. We saw that plans can go haywire and there is a need to be prepared and balancing conflicts of interests, which is a lot more delicate than in academic environments, because there area lot more people to balance.

5.1 How did we try to be creative?In Congo, for example, when choosing the vendor, we had to be able to managepeople's desires effectively, while still having our own way. This is an example of non-technical creativity that is distinctly different to technical creativity. We have anotherexample in Nigeria, where while we were waiting for the funds to arrive from Sweden,


we tried to get an advance of money from the organization to which the site belonged,by convincing them of the commitment of the team with the project, and at the sametime we were doing the same thing with the provider, so they could begin the workbefore receiving the money. In Zambia, the malaria research unit have no recurrentfunds for monthly bandwidth fee, so we had to ring in other two projects (The humanherpes virus type-8 or Kaposi's sarcoma associated herpes virus ) working with MiamiUniversity in USA. These projects had funding for the next three years and made animmediate contribution in terms of 5 wireless access points to extend the LAN to their offices. While waiting for the actual database server, we asked the supplier to provide us with another server, although with less capacity, so that the implementation couldproceed on schedule and the IT staff training could be done. Also during the equipment procurement, the Dell PowerEdge 2800 server with Microsoft software was USD16,000, but we decided to use open source software (freeware) and the cost of theserver came down to USD 10,000. We were able to save USD 6,000 and if we have toinclude other sites, USD 18,000 was saved by using open source software. Regardingtechnical creativity we could mention the case of Congo as well, where they did nothave IT staff, so we had to select very carefully the services we implemented on theserver, to be as easy as possible, to avoid complications.

6. Conclusions

The MIMCom 2005 project was primarily focused on providing Internet accessinfrastructure to various sites in sub-Saharan Africa. The primary objectives weresuccessfully met at 3 sites, and are on the way to completion at the fourth (CERVE).The team successfully dealt with unique problems at each site, and was able to copewith working in an extremely geographically distributed environment, environmentalproblems, as well as technical issues that came up during the execution of the project.

81

Appendix II

MzOpen.Net: Wireless Hotspots in UniversityResidences in Mozambique

Eneas HunguanaQarin Hjortzberg-Nordlund

Alberto MuchangaErik StackenlandJon Åkergården

AbstractThe majority of university students in Mozambique cannot afford to buy a personalcomputer that have become indispensable tools for use in academic activities. For those who own a computer, few have access to the Internet due to the relatively high costs of this service in the country. Considering the also unaffordable costs of books in thecountry, Internet is a crucial resource as a complement in the teaching-learning process, since most of the existing bibliography in local university libraries is not enough and/or is not updated.

Most of the faculties have rooms equipped with networked desktop computers to beshared by students who may book them for their local use to perform academic general-purpose tasks. Some of these computer rooms are connected to the Internet, allowingstudents to make use of Internet for their academic activities. However, most university premises close around 5 pm, meaning that students do not have access to computers, to the Internet or to the possibility of carrying out research or other academic activitiesthat require access to these resources, after that time.

In 2002 a project designated “Mozambique Open” was proposed with the objective offinding a solution to minimize this problem. It was the result of an exchange programbetween the Swedish Royal Institute of Technology (KTH) and the Eduardo Mondlane University (UEM, Mozambique) and was sponsored by the Swedish InternationalDevelopment cooperation Agency (Sida). This project, which was part of aCommunication Systems Design course given by KTH, involved a team of twoMozambican and three Swedish students from the fields of computer science, electricalengineering and industrial management, who investigated and implemented a solutionto provide Internet connectivity in four UEM student residences in Maputo, theMozambican capital city. This document presents a brief overview of the outcomes ofthe project.


1. Objectives and goals

The aim was to provide Internet connectivity in four UEM student residences, with anaccess control mechanism based on the concept of operator neutrality, as a requirement to allow the network clients to be able to choose the services providers they want touse.

2. Major stakeholders

The MzOpen project had the following entities as stakeholders listed, the first threeplaying a more active role for the success of the project:

Eduardo Mondlane University - UEM (www.uem.mz)UEM is the main public university in the country. It is the owner of the deployedinfrastructure. Through its own ISP, Centro Informatica Universidade EduardoMondlane (CIUEM), that is also one of the first ISPs in the country, it contributed tothe project by providing technical support in the establishment of the Internet links.The MzOpen network control servers are hosted in CIUEM.

Royal Institute of Technology – KTH (www.kth.se)KTH is the entity responsible for the Communications Systems Design course. TheMzOpen project initial phases (design up to testing phases) took place at KTH. Theinstitution contributed with technical support provided by a teaching team, laboratoriesand all the indispensable resources for the research activities of the project.

Sida – Swedish International Development Agency (www.sida.se)Sida was the sponsor of the Mozambique Open project. The institution covered thecosts related to team project travels, equipment acquisition and other issues necessary to make the project a reality.

Local ISPsAlthough they did not get involved in the project, these companies can be seen asstakeholders given the nature of the MzOpen network that uses the concept of operator neutrality that allows several ISPs to provide services to the costumers by using acommon access network.

3. Overview

The MzOpen.Net network was built based on two technological concepts: WirelessLANs and Operator Neutral Networks. It consists of Wi-Fi hotspots in studentresidences located around the city (outside of the university campus), connected to theInternet through a wireless backbone to the University ISP (CIUEM), located in themain campus. The access to Internet services provided by CIUEM to the students iscontrolled based on authentication credentials, submitted by the clients via anauthentication services website. Although the backbone belongs to UEM, thetechnological solution is based on the concept of Operator Neutral Networks, allowing several ISPs to provide services using the shared access infrastructure as illustrated inthe diagram below.

Appendix II: MzOpen.Net: Wireless Hotspots in University Residences in Mozambique | 83

Figure 3.1: Open.Net Topology

The choice for wireless was motivated by the relatively high costs of the civilengineering with work associated with wired networks and the flexibility offered bywireless technology for future expansion of the infrastructure. The Operator Neutralconcept was implemented using the Open.Net Access technology(http://www.stockholmopen.net) pioneered and developed by KTH. The “PublicServices” block infrastructure, which is used to allow clients to dynamically choose their providers, is hosted at CIUEM premises. Although some informal contacts andinvitations have been made to another ISP during the project implementation, for thetime being CIUEM is the only services provider

4. Deliverables

A complete report describing the technologies adopted to implement the solution forthe problem being studied, and as well the step-by-step implementation guidelines waselaborated and made available online (http://csd.ssvl.kth.se/~csd2002-mozambiqueopen). The document can be used to replicate similar solutions in otherdeveloping countries or at least constitute a starting point to inspire others in thedevelopment of new solutions. Other set of documents such as the project plan and the lessons learned paper were also made available.

5. Results

As a result of the project implementation, each of the four student residences gained aroom equipped with desktop computers and Wi-Fi hotspots enabling students to access the Internet 24 hours per day. Students that have their own laptops can also benefitfrom the installed infrastructure by approaching the covered areas (some are alreadyaccessing the Internet from their rooms).


The network has the advantage of being easily expandable and given the fact that it was deployed based on the Open.Net philosophy infrastructure, different services providers can operate on the network, giving the students freedom of choice when accessing theservices.

Capacity building in terms of human resources can be seen as another result. Besidesthe experience and knowledge obtained by the project team members, two students per residence were trained in basic network troubleshooting process in order to help in thedaily management tasks. Wireless LAN skills were also shared with the university ISP(CIUEM) technicians, and nowadays they use it to provide solutions to several types of costumers.

6. Impact

The MzOpen.Net project had both direct and indirect impact mainly in the universitycommunity, for the involved team members and somehow in the civil society.

6.1 University CommunityUEM Students in general and particularly those who live in student residences directlybenefited from the installed networks, since they now have access to the Internet 24hours per day. Based on the experience of these networks, several universitydepartments started adopting Wireless LAN technology for expanding their datanetworks. A recent example is the existing plan to cover UEM’s second main campus,the engineering campus, with a wireless network with access to the Internet.

6.2 Team MembersThe experience gained by working in a real-life problem-based project, the contact with and the knowledge transfer among students from different cultures and technicalbackgrounds can be considered as the biggest impact the project had on the projectteam members. Students had the opportunity to deal with new technical topics, and aspecial emphasis was put on the acquisition of skills in developing technologicalsolutions using free and open source tools.

6.3 Civil SocietyBy the time the Mz.Open.Net project took place, the concept and the knowledge ofwireless LANs were not wide spread locally. Based on the acquired skills, theMozambican team members had the opportunity to make some presentations aboutWireless LANs on seminars to university students, in an ICT conference and as well incourses organized for the private sector companies in the country. The focus was more on the scientific point of view, since most of the available information is intended forcommercial purposes. It was (and it still is) noticeable, the interest of the audience(including ISP staff) in having a deeper understanding about WLAN.

7. Lessons learned

Working in such a project and as well the observation of the operation of the deployed networks produced several lessons to be learned. These lessons are invaluable both for

Appendix II: MzOpen.Net: Wireless Hotspots in University Residences in Mozambique | 85

the involved team members and also to be shared with other similar initiatives.Regarding the later, it can be concluded that:• The involvement of several actors is crucial to make Open.Net applicable. ISPs

should have a strong enough reason to convince them to provide services in a similar network, diversity of services (e.g. Internet voice calls, local email accounts, data storage) must be made available so that students also feel motivated to pay for the costs of the services (since other ISPs would provide services on a commercial basis). At the moment CIUEM is only providing the basic service, the access to Internet since its priority is to assure that the whole university community has access to this service. Thus, other ISPs have “room” to introduce value added services.

• It is indispensable, to involve of those who will benefit from the installed infrastructure in the associated management tasks. The operation of the deployed networks observed some constrains in its first months due to equipment damage and theft. These threats showed that the students living in the residences had an important role to play in the process of management and maintenance of the installed infrastructure. With their direct involvement in the management process, some successful operation models were proposed and adopted by the university residences’ community. A concrete example was the development of models that allowed them to independently (without requesting external financial support) introduce and run new “services” (e.g. printing) in a self-sustainable way and to ensure the replacement of stolen and/or damaged equipment. This shows that a well organized student community has the potential to independently generatefinancial resources that can be used to expand and improve their own networks.

86

Appendix III

RINEX – Building a Bridge to Reduce the Digital Divide and Enhance the Use of ICT as a Tool to Eradicate Poverty

Issa NkusiClaude Hakizimana

Coco MusaningabeInnocent Nkurunziza

1. Introduction

In 2000, the Government of Rwanda (GoR) adopted a national ICT policy to supportits social-economic development vision aiming at poverty alleviation. GoR recognizedthe role that Information and Communication Technologies (ICTs) can play inaccelerating the socio-economic development in Rwanda. The main mission is to enable the country to achieve a middle-income status by year 2020. In addition, GoR aims attransforming itself into an information-rich knowledge based society and economy bymodernizing its key sectors using information and communication technologies.It is within this preamble that GoR adopted the National Information andCommunication Infrastructure (NICI) Policy and Plan in late 2000 for which theRwanda Information Technology was created to coordinate and facilitate theimplementation of the NICI plan.

GoR believes that ICTs will remain a myth if the country can not be able to transformthe use of ICT into goods and services that Rwandans can benefit directly from. Thechallenges therefore is to ensure effective utilization of ICTs in areas such asgovernment administration and service delivery, broadband access to citizens, Tele-Medicine, e-learning, e-Services, etc, and to facilitate the access to information in urban and rural areas.

In 2004, a special cooperation between Royal Institute of Technology (KTH), RwandaIT Authority (RITA), Kigali Institute of Science, Technology and Management (KIST)and National University of Rwanda (NUR) fathered Rwanda Internet eXchange point(Rinex). Indeed, Rinex was implemented in line with NICI-2005 plan (reference to“ICT Infrastructure development” pillar).

As things are moving forward, a fibre optic backbone project is progressing at theNational level. Ideas have come up to develop network management and monitoringtools for the government network, network contents such as VoIP services, deployment

Appendix III: RINEX – Building a Bridge to Reduce the Digital Divide | 87

of applications in public administration, schools and/or healthcare centres, which mayadd value on the fibre optic backbone being laid out in Rwanda. A discussion betweenKTH, Rwanda IT Authority, and KIST is on the way in order to implement these ideas.

2. Objective and Goal

Rinex was established to allow peering between ISPs that are connected directly to asingle international upstream, thereby avoiding passing through their own internationallinks for their local communications, hence enhancing quality of service and improvingbandwidth savings.

3. Major stakeholders

• Rwanda IT Authority (RITA);• Internet Service Providers (ISPs);• Universities (KIST and NUR);• Rwanda Utilities Regulatory Agency;• Rwanda ICT Association;• KTH;• Sida.

4. Results and deliverables

4.1. ResultsA Communication Systems Design (CSD) course was the key engine which madeRwanda Internet eXchange point a reality. In 2004, KTH, offered Rwandan students ahigh level training, availing high technology lab environment with a twenty four hoursaccess to the Internet, exchanging experience with different people coming fromdifferent parts of the world (Europe, Nicaragua, Bolivia, Tanzania, Mozambique, Laos, Taiwan, Thailand, India, China, and so on).

Two ISPs currently operating and NUR are connected to the eXchange point. Thoughthe regulation on providing data service does not restrict any, NUR and KIST are more focused in using Internet for academic and research purposes rather than commercial.Human resource development is one of the key elements that Rwanda is emphasising to achieve its vision. KTH, with the CSD course, is contributing and supporting Rwandain capacity building.

4.2. Deliverables to both KTH and RITA (the principal)The project has a website that provides information about the project and the activities of the team. The deliverable documents were uploaded on this website.http://csd.ssvl.kth.se/~csd2004-team21/.

Project plan, presentations, topology design, video (expressing briefly the eXchangepoint project in Rwanda), press release, lessons learned paper, technical proceduresmanual, ISPs system administrator training, final written report produced after thecompletion of the project were all deliverables given to KTH and RITA.


5. Impact of the project

With the increasing demand for online applications (E-mail services, e-education,website hosting, telemedicine, apply for online passport application, etc), Rinex turnedout to have been established at the right time.

In addition, an online application -keeping track of HIV/Aids patients’ treatment and e-provision of medicine in Rwanda- was not easy for users to browse on the Internetwhen it was deployed, because it contains heavy clickable images. Rinex was animportant help for users to access information on the Internet without having to waitfor too long. People host web sites or co-locate servers freely to any ISP since theircontents can be accessed much faster. With such a national ICT infrastructureachievement, Rwandans are trying to improve its network infrastructure and at the same time thinking about contents development.

The Pan-Africa Virtual Internet Exchange (PAVIX) approach was the idea of creating a mesh of point-to-point interconnected African IXPs. Rwanda is now ready to connectto a regional exchange point, if it is going to be implemented. Nevertheless, developing content is crucial; otherwise the establishment of regional exchange points will not beuseful.

6. Lessons Learned of interest to a wider audience

6.1. Team Building:Good working team spirit is the first lesson learned by the students. The team members had different ways of seeing, approaching and solving problems. The team memberscould not get time to organize themselves but had to learn about each other and try tofind a way to work in a more professional attitude with each other. That impliedsacrifices at different levels of personality.

6.2. Practical ExperienceStudents quickly won skills of solving practical problem in the real Internet community. Good planning with defined timeframe makes a project move faster.

6.3. Difficulties Before the implementation phase, students did not have enough experience on working over the Internet. A lot of time was spent on reading materials in order to understandconcepts and it was not easy to cover a large amount of material.

6.4. OpportunitiesDuring the project lifetime, the team members got the opportunity to start their careerwith a lot of support and friendship from professionals from all around the world. The team got the opportunity to experience an international working environment, whichwas a challenge but turned out to be good experiences to bring back home.During the CSD course, students got the opportunity to meet important people during seminars, workshops, lectures at KTH but also when they went back home for theimplementation phase. In brief, this project brought the students on the stage of theICT scene in Rwanda.

Appendix III: RINEX – Building a Bridge to Reduce the Digital Divide | 89

To conclude, the KTH CSD course develops enough skills and hand-on experience tostudents to make a complete real life project, but of course students should do their part to contribute to successes.

In the CSD course of 2004 when Rwanda students participated for the first time, theytried their best and produced notable output (Rinex) which currently serves the localInternet community. In 2006, students hope to do more projects in networkmanagement and monitoring, and application deployment. Grateful thanks are due toKTH, Sida, RITA, Rwandatel, KIST, NUR, peers and the individuals who participatedto make Rinex happen.

Figure 1: The team during launching ceremony and Rinex equipment

Figure 2: Rinex equipment

Section IIIOpportunities

93

Open Access Networking in Africa: The FiberAfrica Proposal*

Rahul Tongia

1. Introduction

The trade-off for “bread vs. computers” regarding information and communicationstechnologies (ICT) and development has been debated extensively, and consensus isevolving to recognize that ICT is a powerful means for development, and its role iscomplementary instead of competitive.3 Certainly, ICT cannot replace otherdevelopment activities, but it has great potential to make development more efficientand sustainable. ICT is also a large and growing market. Indeed, the spectacular growth of cell phones in Africa, where the penetration is now 10%, indicates a strong demandfor such services.

However, when we consider data networking, especially broadband, we find a muchbleaker picture. For starters, most countries have very limited Internet usage. Theavailability and ownership of personal computers (PCs) is a limiting factor, but there are models of shared access (including schools, kiosks, cyber cafés, etc.,) that can helpreduce this issue. Nonetheless, data connectivity is just too expensive, if available at all.4

In large fractions of the developing world, broadband is available only in niche areasand dial-up becomes the only option for connectivity, which itself may be limited.

*Disclaimer: There are many assumptions made in this note that are not shown, and no claim is made for the

optimality or universality of the solutions or ideas. Nonetheless, the FiberAfrica proposal is an attempt that has been thought through at many levels by a number of professionals, and we hope that decision-makers will at the very least engage in a serious dialogue over proposals such as this one. This paper is not meant to be a traditional scholarly work (with minutia of detail over assumptions, references, etc.) Instead it is a paper meant to stimulate discussion and lead decision makers into serious dialogue over how networks are built in developing regions such as Africa. It builds upon discussions with various academics, professionals, and government officers, and a vision articulated by Professors Raj Reddy and V. S. Arunachalam at Carnegie Mellon University. More details on this proposal, dubbed FiberAfrica, can be found at: http://www.contrib.andrew.cmu.edu/~tongia/FiberAfrica--ending_a_digital_divide.pdf3 “The issue is whether we accept that the poor should, in addition to the existing deprivation of income, food

and health service, etc., also be further deprived of new opportunities to improve their livelihood.” Weigel, Gerolf and Waldburger, Daniele (editors). “ICT4D – Connecting People for a Better World. Lessons, Innovations and Perspectives of Information and Communication Technologies in Development.” Swiss Agency for Development and Cooperation (SDC) and Global Knowledge Partnership (GKP). Berne, Switzerland. 2004.4 Satellite connectivity could be available almost anywhere, but it is an expensive proposition.


Unfortunately, the total cost of basic dial-up based connectivity can be multiple timesthe average income in many countries in Africa.5

In this paper we show how for approximately 1 USD/person one-time capital costs, the majority of Africans could avail of (virtually) free data connectivity within walking or cycling distance.6

However, this leapfrog network requires a rethink in how networks are built, owned,operated, and utilized, and are based on an Open Access model.

2. Need for Connectivity and Open Access

If we consider the desired end-goals of empowerment and opportunities, access leads to information, which can lead to knowledge, leading to empowerment and opportunities. Of course, it is not linear, and one requires complementary capabilities, especially tointerpret information into usable knowledge.

Access >>> Information >>> Knowledge >>> Opportunities and Empowerment

While it would be a stretch to claim improving access would solve the problem, it would be a necessary (though not sufficient) condition. If we consider the digital divide, it can be at four levels: Awareness, Availability, Accessibility, and Affordability.7

• Awareness– People must know what can be done with ICT (information and communications technology); they must also be open to using ICT

• Availability – ICT must be offered within reasonable proximity, with appropriate hardware/software

• Accessibility – relates to the ability to use the ICT (spanning literacy, e-literacy,language, interfaces, etc.)

• Affordability – All ICT usage together should, ideally, be only a few percent of one’s income (under 10% maximum); this involves life-cycle costs (total costs of ownership—TCO), spanning hardware, software, connectivity, education, etc.

One concern, linked to the affordability and relevance arguments, is “why broadband?”A leading academic published an article during the World Summit on the InformationSociety (WSIS) Phase 1 titled “Let Them Eat Megabits.”8 While there are many meritsto the arguments, these overlook several issues (in addition to some such argumentsbordering on patronizing). Connectivity is not a binary phenomenon (yes or no) butrather has many facets or dimensions. Using what hardware/platforms? Whattechnologies? What speeds? What uses? If one could make it similarly priced, almost all

5 ITU World Telecommunications Development Report Dec. 2003 data.

6 This excludes end-user equipment such as computers or wireless modems, which would be a distributed

cost that scales with usage. 7 R. Tongia, E. Subrahmanian, and V. S. Arunachalam, “ICT for Sustainable Development: Defining a Global

Research Agenda” (2005) Allied Press, Bangalore. ISBN : 81 - 7764 - 839 – X. Book prepared for the National Science Foundation, World Bank, and United Nations.8 The Financial Times article and discussion are captured at http://www.citi.columbia.edu/elinoam/FT/11-

25-03/megabits.htm

Open Access Networking in Africa: The FiberAfrica Proposal | 95

optimization solutions would call for broadband. The real opportunity is that throughleapfrogging, one can achieve broadband at much lower costs than conventionalwisdom would have indicated.

Mobile phones should only be viewed as complementary to an optimal broadband(data) solution. Mobile penetration is limited in rural developing areas, even if there isthe theoretical coverage footprint in most places (by population). Mobile telephony isjust too expensive for most users, especially in Africa. ITU data from 2003-04 showaverage monthly costs more than double those in India, though the gap is shrinkingsomewhat. 9 More importantly, mobile telephony systems in Africa have primarily been designed to carry voice traffic, and they do it well (and profitably). Claims of the third-generation (3G) cellular systems providing a relevant data solution ignore the very high costs carriers hope to, or rather need to charge to recoup their investments. The totalbandwidth is also relatively limited, especially if we want broadband for users. The very success of mobile telephony poses a paradox for broadband – developed regions havehigh broadband usually because they piggyback on voice (for DSL broadband) or video(for cable broadband) services.

The total bandwidth that fibre has is greater than all competing technologies. It isalready the norm within the backbone and core of networks, and it is a matter of timebefore it becomes used for “last mile access” as well (in some cases copper can workwell, but only for very short distances, primarily in dense areas such as apartments; thecapital costs of copper and fibre are comparable). The primary reason fibre has not yet been used more is cost, especially labour. If developing countries have inexpensivelabour, they should harness it for installing fibre, especially in sub-ducted conduits. The US cites installation costs in the ten thousand dollar per km range (excluding rights ofway) for “easy” unpopulated areas, and an order of magnitude higher for urban areas.We know, from experience, that it can be done for much lower. In India, inter-city and rural fibre is now being drawn and installed for on the order of 1,000 USD/km or less. The capital cost of fibre itself is falling rapidly, and can be on the same order for dozens of strands of fibre.10

Governments can choose traditional, telecom development models. However, unlikeopen access models that aim for higher penetration, these cater more to “viablemarkets.” Universal service obligations have not met with equal success (exceptions like Chile and elsewhere notwithstanding), and this is even less well understood for dataconnectivity, let alone broadband. It is well accepted that a network designed for 1% of the population will be much more expensive per use and different from one designedfor 50% of the population.

In addition to issues of design and scale, if we examine the components of end-usercosts, other than profits, a significant fraction of costs relate to government charges orsurcharges, including license fees, spectrum fees, import duties, rights of way charges,

9 This is not to diminish the value of mobile telephones; complementary ICT can still have a significant

impact, e.g., use of SMS messages for agricultural price discovery. 10

These are actual numbers from Indian fibre manufacturers and network deployers; the figures quoted by leading global fibre suppliers will be higher.


etc. There are other regulatory costs that are not quite taxes such as restrictions ontechnologies, services (such as Voice over IP), etc. Ultimately, all of these simply raisecosts for the consumer. If the government wishes to waive such costs, it would beimportant that these are done for public users in a non-discriminatory manner (else,we’d simply enrich a private company). Open access networks (especially run asprofessional not-for-profits) help ensure any government support is passed on toconsumers. In addition, if governments are worried about losing revenue, they canremain revenue neutral by moving from up-front fees (e.g., from auctions or fixed fees) to revenue sharing models. This not only reduces barriers to entry, it can also raise total revenues as penetration increases. This was seen in the case of India’s cellularoperators, who today have possibly the lowest tariffs in the world and very high growth rates. In addition, there is economic analysis suggesting that lower prices are better forthe government and for the operator. Not only is the elasticity greater than one (solower prices more than compensate), greater uptake in services result in highereconomic activity overall.

One concept we propose is Fibre to the Village (FTTV). We don’t necessarily need fibreto all the homes, but getting it close enough to population centres (a village or cluster of villages) would allow the use of inexpensive, off-the-shelf wireless technologies toprovide high-speed data access.

If technology is improving, why do we need a new design? Unlike many otherinfrastructures in developing countries, telecom is largely profitable and growing rapidly.Conventional wisdom and orthodoxy in telecom has been that the benefits have comethrough private participation, if not competition, which also brings in new technologies such as mobile telephony.

One recognized issue is that telephony costs require reasonable penetration to achievepayback, otherwise the costs per user are high (similar to any niche technology). At the same time, competition means each participant would have lower market share. Optical fibres are a prime technology for connectivity, but can we expect three or moreindependent fibre networks being deployed across Africa in the near term? If we treatoptical fibres like a utility, built everywhere (or deeply enough) just once, then different players could compete to provide services on top of this infrastructure. For rural orunderserved areas, free or nearly free connectivity could be given for community access points. In urban areas, this could provide much cheaper up linking bandwidth forservice providers. Such an open access model would thus allow for public and privatecompetition.

The underpinnings for such an open access model are based on a layers approach tobusiness models and regulation (discussed, e.g., by Comstedt (2005) and by Pehrson(2005) at the 3rd Open Access Conference in Mozambique). Just like the Internet isdesigned with layers of technology, connectivity can operate at similar layers, with openaccess to layers above and below, instead of traditional vertical integration by a provider wanting to do it all.

One major reason why the layer unbundling of open access models is important,especially for greenfield deployments, is economics. Optical fibres last decades (like many


infrastructure). On the other hand, many of the electronics that lie on the network need to be replaced much sooner, not because they fail (some may), but because they arerapidly made obsolete by newer technologies. If a connectivity project has to recoup its investment in, say, 5 years, that is very difficult if one includes the optical fibre, whichwill last many times longer. Deeper discussions and analyses on Open Access modelsare presented in the InfoDev report “Open Access Models: Options for ImprovingBackbone Access in Developing Countries (with a Focus on Sub-Saharan Africa)(2005).11

Open access models mean we need to think of the physical infrastructure as a publicutility.12 Building basic ICT infrastructure similar to FiberAfrica can be achieved forone to two orders of magnitude lower cost than other infrastructure, such as roads.Roads are a good example of an analogous system that allows open access on top ofpublic infrastructure – we don’t want ten highways in parallel under the aim ofcompetition. But, there remains significant competition and private participation,ranging from hardware (e.g., cars) to maintenance (e.g., outsourcing or tendering fortoll-booth operations or even building the roads) to services riding on the infrastructure (e.g., courier and delivery companies).

3. FiberAfrica Model

Using geographical information systems (GIS) data on population, towns, and roads inAfrica, we calculated that approximately 70,000 km of core optical fibre would span all the major population centres in the continent, going along major roads. Such a fibrenetwork could provide tens of gigabits per second of connectivity from day one,hundreds if not thousands of times greater than today.

A backbone network is of limited value to end-users, and we propose a design thatincludes broadband access to the network using fixed broadband wireless technologies.Positioning such core transmission hubs along the backbone, every, say, 60 km, andadding an addition 30,000 km of spur fibres to reach additional areas, we find that such a system could provide nearby coverage to the majority of Africans. This assumesexisting or proven wireless technologies, including pre-standard variations of theupcoming Wi-Max standard (the long-range “cousin” of the wireless Wi-Fi standard).Wi-Max or equivalent receivers could be distributed up to tens of kilometres around the core hubs, and then inexpensive, off-the-shelf shorter-range wireless could be used forlocal redistribution.

Adding up all the equipment, installation, and electronics (as well as back-up powersystems, but not standalone electricity solutions) leads to only approximately 950 million USD of costs, or just over one dollar per person one time capital expenditure! Thisexcludes end-user receivers or computers, which would be a distributed cost, and thissolution doesn’t provide blanket coverage across the entire surface area of Africa.

11 Prepared by Spintrack AB, with core team of Anders Comstedt, Eric Osiakwan and Russell Southwood.

12 There are debates amongst professionals if such public networks should or shouldn’t include the retail

services themselves.


Rather, the model, dubbed “FiberAfrica,” optimizes coverage by population, and covers the majority of the population.

4. Why the Open Access FiberAfrica Model?

There is a philosophy held by many across various fields, and it is amplified in thenetworking world: incremental changes lead to incremental benefits. FiberAfrica is presented as a continental scale network based on optical fibre and (fixed) broadband wireless thatgives the majority of Africans virtually free basic broadband connectivity within walking or cycling distance. The network can be sustained by small-scale payments by lay users, and by commercial users who would pay slightly higher charges (but chargesdramatically lower than not only today but also based on alternative designs as proposed elsewhere).

The rationale for the FiberAfrica model as proposed is based on several realizations:

• Small “Internet Size” of most countries requires unique scaling and design. Most countries in Sub-Saharan Africa are very small in terms of “Internet Size,” and even the obvious exceptions are themselves modest in the scheme of the Internet overall. The number of users, hosts, content, and present interconnections are proportionately much lower than even their GDP when compared to most other nations. Today, most countries are attempting to “reinvent the wheel” with their individual international fibre connected gateways, data centres, security centres, etc. Instead, they could save significant costs by sharing many of these features (with appropriate security mechanisms and sovereign control, of course). A single large-scale core router could handle all the traffic going in or out of Africa today with ease. But, we have countries with a few million people, and less bandwidth than a small city in the US, building out their own networks without optimizing them for the size or scale possible under a trans-national network.

Mobile telephony appears to be synergistic instead of purely competitive, and the question becomes what other sustainable models there are for fibre networks. In the US and other countries, most fibre deployments came from a double if not triple play, but Africa lacks the widespread use of wire line voice or cable TV.

• Domestic content and connectivity are required. Without meaningful penetration within the country, building out international connectivity doesn’t achieve much. Meaningful penetration will only be driven by content that meets domestic (local) needs, and such content is unlikely to be made available from abroad, especially not in local languages. To that end, while international connectivity via optical fibre can be justified, it should be the cart that follows the horse (local needs), and not the other way around. Using international connectivity as the backbone for interconnections is poor and expensive design13 – domestic fibres will be much less expensive and easier to scale. We already have significant global fibre capacity (potentially, hundreds of Gbps) landing at multiple points in Africa (in multiple countries). This

13 Even India had segments until the late 1990s where e-mail from one city to the other would go through the

US!


has not, however, done much for bringing down connectivity/up-linking costs in those countries. 14

• Big bang approaches can sometimes be more acceptable than small interventions that keep the underlying system (and divides) in place. Envisioning any two neighbouring countriescooperating might be more difficult in some cases than creating a common playing field at the continental level. In addition, the vision of FiberAfrica ensures that a new divide between African countries is not created – market-driven solutions would otherwise only connect a subset of countries in a meaningful manner. This also makes it more likely for donors to consider investing in such infrastructure.

• There are certainly interim solutions and technologies that may be less expensive, with the trade-off that they are less scalable. FiberAfrica will not be built out at once, it will rather begin in certain regions first. However, the vision and end-goalshould be continental. Interim solutions will take enormous effort and cost to upgrade the solution down the road.

• There is no lower barrier to entry than free. In addition to innovations in technology, FiberAfrica has a unique business model, whereby public users (schools, hospitals, libraries, etc.) can get free or nearly free broadband access, and end-users can also get free basic connectivity in community access points, distributed throughout Africa. Such access points (kiosks for example) would themselves receive either free or virtually free connectivity, and could charge for value-added services or assistance with transactions and fulfilment. Affordability is a key aspect of the digital divide. Mobile telephony could, in theory, be availed by well over 2/3 of Africans, even a non-trivial fraction of rural populations. However, they choose not, based on the value proposition (or lack thereof).

• A suggested use of donors for the initial build-out is only one option, and private funds could also be used (leveraged through multilateral agencies, perhaps, whichcould help reduce risks). It is worth emphasizing that donors should only pay for the lowest level of open infrastructure; the actual retail services would be provided by other public or private providers, whose total investment would be much larger in the long run.

There are other design aspects to FiberAfrica that make it different from traditional data networks. Today users pay much higher costs for bandwidth, and applications aredriven at the edge (the network is a “dumb connection”). By enabling (virtually) freebasic access, this will allow much greater local content creation and utilization (again, at

14 There is a proposal by NEPAD to build a fibre-optic backbone for multiple countries in the Eastern

portion of Africa based on submarine cables – EASSy – Eastern Africa Submarine Cable System. This is a very positive step in bringing countries together, but FiberAfrica might be a better design for the same stakeholders to consider. In addition to the much higher cost for EASSy—estimated at about 10 times higher per km for the submarine fibre than the low-cost terrestrial figure of USD 3,000/km inclusive of initial equipment—“bringing the Internet to more countries,” even at high speeds, does very little for increasing access and penetration, especially in rural areas. Also, this network has not been envisaged with innovations in business plans or true open access, which would be required to make ICT available and affordable to more people.


the edge). Of course, people could buy bandwidth like traditional models, but theupfront costs would be much higher (not to mention they would face greater hardwarecosts).

If people wish to download digital libraries (books) from the US instead of localcontent, they would simply pay a little more for that service, either under the application value-add model, or the traditional bandwidth usage model, which in FiberAfricadifferentiates between users and types of usage. This approach makes it easier toprovide Quality of Service based applications, moving beyond the best-effort Internetof today. Based on this, we can make downloading educational programs free, whiledownloading the newest hit song or movie something users pay for. This also helpsreduce the need for greater international connectivity, which is an expensive link (along with optimized design, local data centres, etc.)15

Why should anyone build any such network in Africa, instead of first building suchnetworks in more well-to-do nations? This represents a leapfrog opportunity, with less legacy needs, and less regulatory hurdles (esp. compared to the US!) It might even beone of the best methods for developing ICT in the continent, combining innovations in technology and in regulation/business models.

In addition, the imperative for intervention in Africa is much higher given the starkdifferences in human development versus the rest of the world. Critics might believethat Africa’s limited development is due only to poor governance and corruption. Thepicture is somewhat more complex and Africa has been burdened with severaldebilitating challenges. The rains are seasonal and erratic, and the overwhelmingmajority of agriculture is rain-based, instead of based on irrigation. The soil is alsohighly depleted, reducing productivity dramatically. On top of this, Africa also bears the burden of a triad of endemic diseases – HIV/AIDS, malaria, and tuberculosis. WhileICT will not directly help with these, it can play a powerful supportive role in improving the efficiency and transparency of all development efforts.

5. Details on the FiberAfrica Model

The accompanying figure and write-up in Appendix II show a schematic for a newoptical fibre backbone proposed under this project, FiberAfrica. Combining newtechnologies including Dense Wavelength Division Multiplexing optical networking and broadband wireless, this project aims for near universal access at very low if not zerocost to most end-users, with small charges for value-added services.

Core network: The core of the network is an optical fibre backbone, connecting virtually all major population centres at multi-gigabit speeds. Even if these cities are not all “lit”in the first phase of the network, it becomes much easier to connect them over time as demand warrants. This design offers almost limitless capacity, scalability and “future-proofness.” A preliminary design indicates that the core network would be ofapproximately 70,000 route kilometres in size linking 400 cities in Africa with minimum

15 It is important to recognize FiberAfrica is not advocating a central hierarchy to connectivity; it maintains

the end-to-end principle of the Internet.


populations of 250,000 persons each. An additional 30,000 km of fibre spurs (notshown) would be laid to reach other areas and to provide for wireless hubs.

Access solutions: A backbone network is of limited value without users accessing thenetwork, and this is where new wireless technologies are expected to play a major role. Wireless technologies are exceptionally attractive not only because of their ability to be deployed with limited existing infrastructure, but also because they are based on ashared medium, which lends itself well to low densities of users. While 802.11 (“Wi-Fi”)has led to “hot spots” and entrepreneurial innovation for wireless access, there areemerging technologies better suited for wide-area networking (such as 802.16 –“Wi-Max” or specialized alternatives). The routes along the fibre backbone are an obviousstarting point for long reach wireless hubs, and regional optical fibre spurs would extendthe wireless footprint; satellite-based connectivity would be useful for remote locationswhere extending optical fibre connectivity is not cost effective. The aim is for mostusers to have access available within walking or easy cycling distance, and therevolutionary pricing model (free basic connectivity) eliminates barriers to entry forusers. A number of life cycle analyses have shown that hardware is not the majorexpense for end-users (and these are continuously becoming cheaper). Connectivity isthe major cost for many users, along with applications. With adequate bandwidth, many applications can be run remotely, and could even use voice-based interfaces in locallanguages.

Capital costs: The capital costs for the entire continent are expected to total under abillion dollars (based on the assumptions detailed below, and excluding certain end-userequipment like computers or modems). While this appears a very large sum, this is only roughly one dollar one-time cost per person. Amortizing the initial costs, these are only a few percent (about 2%) of the current annual telecom expenditure in Africa.

5. 1 Possible Business and Governance ModelFor this network, there can be several business and financing models, and a few of these are discussed here in brief (the models are not mutually exclusive). We propose toseparate the construction of FiberAfrica from its operation and from its ownership.This is important given the limited funds available from the nations themselves, and toensure that this network has a strong African stake and participation.

Capital Costs: Given the modest capital costs, which can be amortized under USD100million/year, the initial capital can come from several sources, including:

• Donor countries such as the G-8 and a few others. They would stand to gain in tangible manners, e.g., countries that donated funds could reap free (but secure) connectivity for their embassies, aid agencies and projects, etc.

• Private sources of funding would raise the costs to end-users, but probably not to a detrimental level. Even still, multilateral support would be beneficial for risk mitigation, and for providing “soft” support such as through establishing cooperative relationships with other countries and by supporting capacity building.


• Vendors and technology providers routinely provide vendor financing for large projects. They stand to gain as this initial network will seed numerous further contracts for regional, metro, and enterprise networks.

• Governments can also contribute through taxpayer revenues, just as they pay for infrastructure such as roads. In fact, the costs per government are relatively modest compared to roads, which can cost up to several million dollars per kilometredepending on the location and quality—making FiberAfrica two orders of magnitude less expensive.

Beneficiary countries are not required to pay any capital costs, but only provide the appropriateregulatory/policy environment to allow FiberAfrica to be built as their “buy in.”

Ownership: Ownership could lie with a consortium built of member-states, similar to the original Intelsat model. The system can be fair, for example, with voting rights separated from a GDP-based financial stake. We propose The New Partnership for Africa’sDevelopment (NEPAD) as the stakeholder entity to spearhead FiberAfrica, whichbrings together all the countries in Africa in a continental partnership. NEPAD’s e-Africa Commission has a mandate to bridge the digital divide in the continent andfacilitate development. Each country could choose one (or more) entity to act as a nodal entity or even local operator.

Complementary Participation: The model for the FiberAfrica Network envisionswidespread distributed access and complementary development, especially forapplications and reselling access. Kiosk and shared community access providers are akey component of helping spread the penetration into rural areas, as shared access isvital for bridging the digital, information and knowledge divide in an affordable andsustainable manner.

Operations: FiberAfrica itself would have operating costs that would be on the sameorder as the annualized capital costs, under 100 million USD per annum, a conservative estimate that includes multiple points of international connectivity, R&D, maintenance,emergency back-up power, rentals, etc. This excludes the costs (capital or operating) by resellers and entrepreneurs who help increase access to value added services such asvideo on demand, IP Videophones, e-learning and telemedicine, or the costs by thegovernment for content and e-governance. The network would be operated by a private operator(s) under the control or regulation of the ownership consortium; Intelsat itselfmight be an appropriate body given their technical expertise as well as relationships with all the countries in Africa.

Costs and Payments: The operating costs are estimated to be at least an order ofmagnitude lower than today’s costs for core network connectivity. We propose tostructure access charges in a manner where public end-users (like schools, libraries,hospitals, etc.) could receive free basic connectivity, while certain classes of commercial users, or end-users who use value-added applications (like voice-telephony,entertainment, etc.), would pay corresponding charges (perhaps using pre-paid cards),ensuring the business viability of the network. Kiosk operators could also be given veryinexpensive connectivity, under the condition that basic access by consumers would be free for limited and for educational use.


Appendix III has more details on the business model.

5.2 Alternative approaches consideredWithout a network designed similar to the one proposed, with great economies of scale, each country will need to replicate much of the network, especially when it comes tointernational connectivity, up-linking, and data centres. Alternate incremental networkdesigns and upgrades will only lead to incremental or marginal benefits. In fact, ICTdoes not always lend itself well to intermediate efforts but rather favours leapfroggingand widespread deployment, because of the so-called “network effect” – the value of a network can be proportional to the square of the number of users. This network isdesigned with rural areas in mind, and under alternative models, including trickle-downand private sector efforts, one remains unable to reach the “have-nots.”

Given that the goal is to have maximum penetration at lowest costs; FiberAfricaappears more attractive than traditional data solutions whereby internationalconnectivity (through submarine cables and/or satellite) is the design focus. Oceaniccables are much more expensive, have higher maintenance costs, and provide littleemployment to local labour. Because of the relatively small size (user base) of the Net inAfrica, a continental scale network (multi-region, in the initial stages) will provide thelowest cost up-linking.

5.3 Potential risks and mitigationThere are a number of issues and concerns with such a large project, not limited tofinancing. One of the main sources of potential failure would be fewer peopleconnecting than planned, but even then the cost-benefit calculus should remainpositive. The optical fibre infrastructure, a large fraction of the capital costs, will beusable well into the future, and ducted construction will even allow new fibres to beused in the future with much lower investment. In comparison, other grand schemes,such as Teledesic, were more capital intensive, had shorter timeframes for amortization, and had lower richness in terms of value-addition.

Cooperation: This vision requires cooperation between all the countries in Africa,potentially through NEPAD. While there are some regional cooperative networks, such as COMESA and WAEMU, etc., FiberAfrica, being Pan-African, reduces local andregional rivalry issues. A country may choose not to join, but it might lose outsignificantly (and face higher “entry costs” in the future).

Security: Physical and data security are paramount in this network, with extensiveredundancy and robustness in the design to counter issues such as inter- and intra-nation conflict. To ameliorate vandalism and theft concerns through greaterparticipation of local communities in the network, local participation is important. Forexample, at every site that requires equipment housing (every 60 to 80 kilometres, say), a local entrepreneur would be given concessionary connectivity for Value Added Services. He or she would help secure and physically maintain some of the equipment. Byproviding local connectivity, it is possible not only to spread access around the routingof the backbone, but also reduce local opposition and mischief. Experience from India also suggests several techniques for reducing theft of optical fibre and cables. While


copper is often dug out from access networks due to its resale value, optical fibre hasvery little resale value, something would-be-thieves quickly learned after superfluousbundles were purposely left behind at construction sites.

National Policies: Countries need to commit to investments to help spread penetration,and develop the applications for harnessing the power of the network, such as e-governance initiatives. Member countries must also enact enablinglegislation/regulations that allow FiberAfrica to be built, e.g., allowing appropriatespectrum availability and disruptive technologies such as Voice over IP. At the sametime, FiberAfrica must work within the bounds of sovereign decision-making. Countries must also allow appropriate cost reductions such as duty import waivers or free rights of way. Without these the costs to the consumers will increase. The network should bebuilt so that access is non-discriminatory and largely free. Otherwise, concerns remainthat incumbents and alternative players would object. Here, experience from ruraldevelopment initiatives in other countries have shown that urban utilities and serviceproviders often do not oppose networks that have a rural focus, as they consider suchareas unattractive commercially. ISPs and other providers today would also benefit from FiberAfrica, especially if regulators allow it to be used for complementary services (such as aggregated voice transport). However, such calculations are not part of the focus ofFiberAfrica, nor included in the business model.

6. Next Steps

Issues that need to be resolved:

• Working with the diverse countries within their sovereign regulations and policies; Inter-country collaboration (perhaps through NEPAD); issues in inter-countryconflicts and resolution

• Collaboration with donor and development agencies; optimizing interactions with existing and complementary projects

• Financing and business models that ensure independence and operational sustainability for FiberAfrica

• Synergizing sectoral linkages to maximize social and economic benefits

To begin, this project needs to build consensus amongst stakeholders. NEPAD and the World Bank can play complementary roles. The Bank (or perhaps UN/ITU?) can act as a facilitator, with relationships amongst all the African countries as well as donors andtechnology players.

The next steps would involve the development of a detailed project report/businessplan. This would incorporate synergies with other plans with similar objectives, as wellas governmental projects.

FiberAfrica will likely begin in one or a few countries in Africa. In addition to therequired planning, design, and financing, there must be strong synergies to existing


networks and infrastructure, e.g., unutilized or underutilized fibre.16 There are evenbroadband initiatives being started by select countries, which can serve as a startingpoint. One important step must be local partners and stakeholders (ideally, owners),and the identification of “champions” who can shepherd the concept through thepolitical system(s) in place. After agreements on such issues, the first portion of thenetwork can begin in a relatively short timeframe.

One suggestion has been to build a special Research and Non-Profit Network similar in design to FiberAfrica, but which would not threaten existing telecom infrastructures(and vested interests). While this would help prove the technology, cost structure, andperhaps some benefits, even building those networks may prove difficult. There aregood initiatives like SchoolNet that provide connectivity and other services, and thesemust be leveraged. We must, however, consider that if it takes extreme effort to build a network, why shouldn’t it be an open access network benefiting all the citizens of acountry or region?

We can see (Appendix III) that there are multiple models for revenues, and some ofthese may realize greater earnings than others. All of these are plausible, and somecombination of these can provide not only the operating costs, but even the capitalcosts (over time). The analysis shows that FiberAfrica is economically feasible, and the greater challenges are in regulatory mindset and political will. By embracing an openaccess network like FiberAfrica, we can help usher in an era of affordable connectivityacross Africa.

16 Often, optical fibre is laid along energy or other infrastructure. Such companies or entities might want to

use telecommunications for internal SCADA (Supervisory Control and Data Acquisition), but this only requires modest bandwidth. One win-win solution is for such fibres to be used for Open Access networks for the public (such as through FiberAfrica), and in-return the company can get free connectivity for internal and Internet needs, without the expense and headache of operating telecommunications facilities. They would also be reimbursed for the costs of the fibre.


Appendix I: Development Targets

WSIS Summary of Targets (Phase 1)

Target FiberAfrica Addresses?

1 To connect villages with ICTs and establish community access points.

Yes (community access points themselves will be set up by

entrepreneurs and others; FA enables these)

2To connect universities, colleges, secondary schools and primary

schools with ICTs.Yes

3 To connect scientific and research centres with ICTs. Yes

4To connect public libraries, cultural centres, museums, post offices and

archives with ICTs.Yes

5 To connect health centres and hospitals with ICTs. Yes

6To connect all local and central

government departments and establish websites and e-mail addresses.

Yes (FA enables the implementation of e-governance)

7

To adapt all primary and secondary school curricula to meet the challenges of the Information Society, taking into

account national circumstances.

Indirectly (FA enables knowledge-centric curricula)

8To ensure that all of the world's

population has access to television and radio services.

Indirectly (countries can leapfrog to digital information services,

carried over FA).

9

To encourage the development of content and to put in place technical conditions in order to facilitate the

presence and use of all world languages on the Internet.

Yes (FA encourages local content)

10To ensure that more than half the world's inhabitants have access to

ICTs within their reach.YES


Millennium Development Goals

Goal FiberAfrica Role

1 Eradicate extreme poverty andhunger Indirect

2 Achieve universal primary education Direct

3 Promote gender equality and empower women Indirect

4 Reduce child mortality Indirect

5 Improve maternal health Indirect

6 Combat HIV/AIDS, malaria, and other diseases Indirect

7 Ensure environmental sustainability Indirect

8 Develop a global partnership for development Direct


Appendix II: FiberAfrica Preliminary Design and Features

Salient FeaturesCore network estimate (shown):70,000 km

Regional Fibre Spurs (to reach other major areas not in yellow and to connect Wireless Hubs):30,000 km

Approximately 400 population centres are connected, including all capitals and largercities.

The links shown are largely along major roads, which thus cover many populationcentres. Preliminary GIS modelling indicates good penetration of this fibre/wirelesshybrid design.

Fibre (including the laying) is roughly half the capital costs. This implies mostexpenditure can have a long amortization horizon.


FiberAfrica Access Model – Wireless with Optical Fibre Interconnection

Generic Model for FiberAfrica Core

This figure shows a generic model for the fibre and wireless system. Major cities aretypically hundreds of kilometres apart, and connected using Dense Wavelength DivisionMultiplexing (DWDM) technology, which can carry terabits of data per second ifrequired. Cities include extensive fibre and wireless networking, equipment for which is not shown in the diagram. Along the route, optimized to cover the greatest populationat lowest cost using GIS modelling, networking equipment (optical amplifiers) isrequired every 60-80 km to amplify the signal. These locations become ideal sites forwireless transmission central hubs (potentially through 802.16 technology – “Wi-Max”),which can spread a very high speed signal over a 30-50 km radius (with line of sight).These can be picked up by receivers either for direct use (schools, libraries, government, select users, etc.) or can be resold by entrepreneurs, perhaps through complementarywireless technologies such as 802.11 (“Wi-Fi”) or simply community access centres.This design can reach a significant share of the population, since a sizeable fraction ofthe population lives in or relatively near bigger cities. This also bypasses many issues of up-linking that traditional designs (PTT-centric17) face, which rely on a third party tointerconnect sites, and it allows much higher speeds than typical links, which are oftenat most a few megabits per second.

While Wi-Fi is very inexpensive, it is not necessarily the best technology for a next-generation network. Wi-Max has been designed specifically for wide area networking(with features such as interference robustness, no requirement for line-of-sight, multiple frequency possibilities, etc.), and is approaching commercialization. With volume, we

17 PTT – Post, Telegraph, and Telephone (the governmental monopoly for such services, which has often

been transformed into many of the incumbents replacing government telecom service)

Additional optical amplifiers

. . .

Inline Optical Amplifier (with add/drop capabilities)

80 km 80 km 80 km 80 km

Major Cities (hundreds of km apart)

MajorCity

MajorCity

Wireless Transmission Central Hubs (10s of Mbps)

Upto 50 km

Wireless Receiving Hubs(can resell access nearby using 802.11 or other shorter-rangewireless)

Additional optical amplifiers

. . .

Inline Optical Amplifier (with add/drop capabilities)

80 km 80 km 80 km 80 km

Major Cities (hundreds of km apart)

MajorCity

MajorCity

Wireless Transmission Central Hubs (10s of Mbps)

Upto 50 km

Wireless Receiving Hubs(can resell access nearby using 802.11 or other shorter-rangewireless)


expect prices to fall dramatically (similar to 802.11’s price trends, shown below), eventhough our business model assumes 2004 prices, conservatively.18

Wireless Cost Trends – The example of 802.11

Year

Cost Per Node

(USD)*1997 8001999 4002000 2002001 1002002 502003 20

Compiled from various sources

* These costs are for the electronics including packaging and power supply, but exclude any external antennae or towers.

While Wi-Max based Central Hubs will cost significantly more, the receiver costs should fall significantly from the hundreds of dollars today. Declines in such costs areimportant since these are borne by the receiving entities. However, we are unlikely tosee similar volumes (or as low prices) as with Wi-Fi.

18 The design does not assume breakthroughs in wireless technologies, even though some might be realized

within a few years, including smart antennae (MIMO), mesh networks, and software defined or cognitive radios. We assume medium-range coverage for a central wireless hub, not the “best case” scenarios touted by proponents. We similarly assume 2004-05 technologies and costs for other network components.


Appendix III: Business Models for FiberAfrica

The most importance business design feature of FiberAfrica is the dramatically lowerprice of connectivity for consumers—there is no lower barrier to usage than freeconnectivity. Given the ample bandwidth (tens of gigabits in the core, operational from day one) and deep penetration, the population can enjoy free basic (limited) accessthrough kiosks, while paying categories of users also enjoy lower costs from theeconomies of scale and scope.

Capital costs: The capital costs of less than 1 billion USD19 can be amortized over 20years, with capital costs coming to about 80 million USD annually. Even this amountneed not be spent all at once, but over 2-5 years as the network expands with regionaland spur networks.20

Operating Costs: If we assume that donor agencies and countries themselves provide the capital expenses, or a large fraction of these, then only the operating costs must berecovered from users. Failing operating costs recovery, taxpayer and/or aid moneywould be required for the operating expenses, of about 90 million USD per year, which is not envisaged. This figure is conservative and includes all costs such as internationalconnectivity, maintenance, upgrades, salaries, extensive R&D, insurance, rent, electricity, etc.

Some fraction (perhaps 1/3) can be recovered through “power” users (such as nichesoftware companies, multinationals, etc.), who would receive bulk connectivity to thehigh-speed FiberAfrica network. In reality, such users could actually pay significantlymore, given the usage and price statistics of today. (Using FiberAfrica for aggregatingand transporting traditional voice traffic alone could justify a significant portion of theinvestment, but we exclude such calculations for the time being given regulatory andpolicy uncertainty). The remaining 2/3, or 60 million dollars, could be recovered fromthe users. 2,000 wireless transmission hubs could cover about 1,000 sq. km. each, andeach hub would reach, say, 10 public use users (schools, libraries, govt., etc.) and 10community access centres/resellers/kiosks (conservatively). The public users couldreceive free access, as the government pays a nominal charge for the connectivity, equal to the marginal costs, which would still be significantly lower than today’s costs paid by the government. Assuming 1,000 students/users per year (only!) per large school orlibrary, that implies only USD1 per user per year as costs towards FiberAfrica (paid bythe governments), raising 20 million dollars per year.

If we take the example of 20,000 community access centres throughout Africa, each of these could provide services to hundreds of users per year. One innovation forFiberAfrica would be that users could use basic services (limited email or Internet

19 The capital costs as calculated are conservative. Equipment is costed using 2004 prices, and much lower

labour costs for installing optical fibres have been seen in countries like India. 20

On an annualized basis, the investment would be only on the order of 1-2% of the telecom investment in Africa today (as per ITU data). Even when factoring in end-user equipment, etc., this would still be only on the order of a few percent.


access) for free. They would register to become users of FiberAfrica,21 paying a nominal amount (say, 5 USD) for a pre-paid card.22 Beyond limited free usage or if they usedvalue-added services (such as downloading a movie or making a phone call that ends at a traditional phone line), users would pay nominal charges for these. A fraction of thepre-paid charges would be given to the kiosk/community access centre to cover theircosts. Even when people use value-added services, the aim is for their total expenditure to be about 2% of their salary, which is affordable. (In contrast, we note that in the US, the average spending on ICT and media has remained constant for many decades atabout 6% of earnings, while the global fraction on ICT alone has grown to nearly 3%23).Using the African average of about 675 USD /capita (with great variance amongstregions), just 2% implies about 14 USD /year expenditure for such services.

Given that the entire population will not access FiberAfrica and assuming that just thesubscribers are counted, the remaining 40 million dollars of operating costs impliescharges of only 2 USD /year paid by about 1,000 users per access centre, which seemsrealistic (2 USD /user, 1,000 users/access centre, 10 kiosks/wireless transmission hub,2,000 hubs across Africa). In practice a higher number would be recovered from end-users (target 14 USD /capita), the bulk of which would go to content providers and tokiosk operators. Even this is modest, given that we are not targeting the average person in Africa, but a select group of about 20,000,000 regular users (2.5% of the population);the number of potential users is much higher, and usage can grow over time into thetens of percentage. This target expenditure by a few percent of the population is veryreasonable, when we consider that the roughly10% of the population in Africa usingtelecom today pays on the order of 300 USD per user annually (excluding expenditure on information).

21 This would also allow greater authentication and security for the network, a major issue for countries with

limited cyber security infrastructure or policies. Of course, traditional access similar to Internet Cafes of today would also be available, but would cost more than the proposed system (but still less than Internet Cafés of today).22

The use of prepaid cards has widespread acceptance in Africa, e.g., for mobile phones. Such a system might be linked to a universal cash card, usable for other economic transactions such as utility bills. The kiosk operator could sell such cards, retaining a percentage for his operating costs. 23

http://www.itu.int/ITU-D/ict/publications/wtdr_03/index.html


Potential breakdown of annual cash flows (FiberAfrica), at a continental scale:

Annualizedrequirement

(USDMillions)

Notes/Assumptions

Capital Costs 80 Paid by donors with long-term amortizationOperating Costs

(recovery) 90 All-inclusive

breakdown

Donors 0

Virtually free connectivity in return for their grants;

Future donations are for capacity building, content, and end-user equipment (e.g.,

computers for schools)

Governments 0

Virtually free connectivity in exchange for “buy-in” through appropriate

policies/regulations and other in-kindcontributions such as rights of way,

spectrum availability, etc.

Public Users (schools, libraries,

hospitals, etc.)20

Paid by governments; tens or hundreds of times lower costs than today; higher

recoveries from “power” users can reduce government contribution requirements, and donors could assist with such payments for

select countries

Shared end-users 40

If there are 20 million total users, they would pay only 2 USD/year as allocated access

charges for value-added services;Actual individual expenditure would be higher to cover cost of kiosk operator,

content, etc.

Commercial Users 30

If there is modest growth of such users, including retail (urban) ISPs, this sector’s

revenues can grow enormously;Excludes any potential revenues from voice

transport (subject to regulations)

114

Open Access: How Good is it for Africa?

Stelios Papadakis

ABSTRACTAccess to telecommunication infrastructure and services in Africa still remains limitedand strategic policy measures are needed to expand the continent’s limited and generally poor telecommunication infrastructure for greater access. Revolution intelecommunication technologies is opening a door for Africa to catch up with the restof the world in terms of provision and access to telecommunication services. It is nosecret that Africa’s current telecommunication environment is far behind the rest of the developed world and the gap gets bigger as time goes on. Among the technologiespresenting an opportunity for Africa, wireless and IP technologies could be seen as theones at the forefront. Wireless technologies allow rapid expansion of network coveragewhen compared to wired technologies, which for years have taken long to expand. IPtechnologies stimulate innovation, creating possibilities for anyone to create servicesregardless of their ownership of the transmission network. An IP network treats packets equally. On the other hand, technologies themselves will not help resolving accessproblems in Africa, better policies, regulations and stronger political will to developAfrican telecommunication infrastructures are required. Even though most AfricanGovernments have been considering telecommunication services as a priority sector for further liberalization, protectionist policies designed to shield inefficient incumbentoperators, high import duties on IT equipment, archaic licensing provisions, and afragmented approach to ICT development have been some of the most importantstumbling blocks to development. This has to change if the continent wants to getaboard the fast moving train of telecommunication development.

1. Introduction

Africa is the second-largest continent, after Asia, in size, and contributes with nearly14% of the world population, which makes it the second most populated continent inthe world. Despite that, less than 3% of the world’s main telephone lines are in Africa,Internet penetration in Africa has also been slow, thus further widening the access gapbetween Africa and the rest of the world. This gap is even worse when we take intoaccount the uniqueness of Africa in comparison with other regions of the world, as the majority of its population, about 80% lives in rural areas and have never seen atelephone, least to say use it.

The fact that most of Africa is still to be covered, positions the continent as a greenfield for big telecommunications investments, providing a wide range of opportunities toboth local and foreign investors. Therefore, the challenge for Africa is how to attractinvestments into their markets in order to ensure deployment of appropriate

Open Access: How Good is it for Africa? | 115

infrastructures, and adequate connectivity within and across the continent. If Africangovernments are to meet the increasing communication needs of the continent, then the issue of investment must be tackled head on. As such, appropriate policies must be put in place in order to gain investors’ confidence and enable growth of the market.

Recent reforms in the telecom sector have for the first time in many years led tooptimism across the whole of Africa. But despite some remarkable changes across thecontinent over the past few years overall teledensity remain extremely low, with therollout of fixed lines by incumbent operators barely inching forward in most countries, while the uptake of mobile telephony moves rapidly past it.

In this article, I intend to discuss how open access policies, supported by the explosion of wireless and IP technologies, can assist African policy makers and regulators toenable an investor friendly market in order for investors to become attracted to explore opportunities that are created and consequently contribute to the objectives of universal access to telecommunication services. This article is not intended to suggest whetheropen access to telecommunications networks is good or bad for the Africantelecommunications market. In fact, the choice between “open” and “closed” is not asstraight forward as one might think; rather it consists of different policy bases operating from different perspectives on the network.

The concept of open access to communication infrastructures has been a hot topic inmany discussions around the world, involving policy makers, regulators, academics andindustry players. These debates are now gaining wider attention from African policymakers in order to exploit the benefits it can bring to the development of Africantelecommunication infrastructure.

2. Current state of African telecommunications

2.1 InfrastructureAs mentioned before, Africa is a unique continent when compared to other regions inthe world, with a majority of its population living in areas that can be described as rural, and largely without basic infrastructure, including telecommunications.

Availability of telecommunication infrastructures in Africa is still far from ideal. Mostparts of the continent is still without access and even those with access haveinfrastructure that are unable to sustain today’s applications due to bandwidthconstraints, or the usage cost is usually prohibitive, if we consider the low income of the majority of Africans.

The lack of telecom infrastructure is one of the most important economic factorscurrently holding back Africa’s development. Aware of this, African leaders andoperators have embarked on a series of initiatives in order to stimulate the growth oftelecommunication infrastructure in the continent. These days, regional and sub-regional co-operation on regulatory and policy matters have been set up, and the NewPartnership for Africa’s Development (NEPAD) includes ICT as a strategic sector.


There are also other initiatives that are, for instance, aimed to stop routing intra-Africantelecoms traffic via European capitals.

As a result of these initiatives, there has been, in recent years, an increase in the rate of expansion and modernization of fixed telecommunication networks, associated to anexplosion of mobile cellular networks. Africa’s low population densities and dispersepopulation makes wireless cellular network systems more economical than a fixed-lineinfrastructure. African networks have been growing at the extraordinary rate of twicethe international average. This is primarily due to the small number and high cost offixed-line connections, which are seldom extended beyond the major urban centres, andthe liberalization of the mobile phone market.

There is a high level of variability between and within countries of the state of theirexisting telephone networks. Even if telecom infrastructure is beginning to spread, avery small proportion of the population can actually afford their own telephone. Some countries have made telecommunications a priority and are installing digital switcheswith fibre optic inter-city backbones and the newest cellular and mobile technology. But still, financing has been the main stumbling block in order to develop networks in theshort run. According to an International Telecommunication Union (ITU) 2004 report, the percentage of the African population within range of a mobile signal is estimated at only 60 percent, which is the lowest in the world.

All African countries are now connected to the Internet, but the Internet remains out of reach to the vast majority of Africans and is still mostly confined to the larger cities and towns. By early 2004, overall Internet penetration in Africa was below 1.2%. However,Internet access has increased significantly over the past decade, sometimes at rates four-to-five times higher than earlier estimates due to unreported users using free Webservices, Internet cafes or prepaid cards. Although Africa’s data traffic continues togrow strongly, most nations still lack sufficient international bandwidth to reliablydeliver either Web pages or the new digital services in any volume over the Internet.Limited penetration, particularly in rural areas can be attributed to unreliableconnections, and high telecommunication costs. The provision of infrastructure remains one of the key challenges facing Africa as it builds an Information society.

In fact, not all of these initiatives have been registering great success. For example, theAfrica One submarine cable project, which was intended to lay down a submarine cable around Africa, was technically fine, but lack of financing has caused it to fail. Theproject would have provided Africa with the bandwidth it needs in order to enablebroadband services across the continent and carry the majority of the telecom trafficacross the continent.

2.2 Policy trendsThe World Trade Organization (WTO), 1997 Agreement on Basic Telecommunications (ABT) laid the foundation for improved market access and the liberalization of foreigninvestment in the telecommunications sector. It was a necessary starting point forphasing out telecommunications carrier monopolies and providing the regulatoryprinciples and fundamental commitments needed to introduce competition in basictelecommunications services across WTO Member countries.


Since then, many African countries have embarked upon the process of liberalization of the telecommunication sector, coupled with privatization of state monopoly operators.There is no doubt that market liberalization has had a positive impact in thedevelopment of the sector in Africa. Liberalization has given an opportunity forcountries to attract foreign direct investment, strengthen basic and advancedcommunication infrastructure, and allow populations greater access to the worldthrough information technology, thus contributing to universal access objectives.

Still, while many countries have allowed competition in the mobile market, the samecannot be said about fixed-line. To date, few countries have allowed the introduction of new entrants and in some of them competition is already happening in both fixed-lineand mobile markets. Second National Operator (SNO) licences have been issued tocompanies in Ghana, Nigeria, Tanzania, Seychelles, South Africa, Zimbabwe, and South Africa, and other countries are following this lead.

As liberalization continues to forge ahead, more than one-third of all state telcos havealready privatized and several more are set to follow suit. Some countries have chosennot to privatize, thus adopting protectionist policies designed to shield inefficientincumbent operators. Protectionist policies do not benefit the country at all. In reality it only benefits a minority, leaving the majority without access to communication servicesand the few who have must pay costs well above the average when compared to othercountries.

Empirical evidence shows that privatization has contributed to the recent teledensitygrowth in Africa and in the world at large. The advantage of privatized companies isthat these companies incur the costs of installing and upgrading telecoms infrastructure, thus improving bandwidth with no expense to governments or their citizens. But thereis no doubt that privatization itself will not resolve the problem of teledensity in Africa.There must also be a strong regulator, as well as plenty of competition.

African policy-makers usually face the fact that their policies are not synchronized with technological developments and international market developments by themselves. Lack of African policy research institutions has been a problem, making it difficult forAfrican policy makers to develop policies that respond appropriately to the needs of the country. Policies are developed with major assistance from Western consultants, whousually are not familiar with African environments and can therefore not designpractical solutions efficiently suited to the markets of the continent.

2.3 RegulationsAs the wave of market liberalization swept across the continent, introducingcompetition in the market place, independent regulators had to be created in order tomonitor that market. Introduction of a transparent legal and regulatory framework wasseen as a must for telecommunication market development. To date, 33 out of 55countries have set up regulatory agencies, with varying degrees of independence.

Effective regulation of telecommunication environments by the governments of African countries encourage private companies to invest in telecommunication infrastructure,


providing for both sustainable access to telecommunication services and increasedbandwidth on the continent.

Regulators have been tasked with the duties of monitoring the quality of services being provided by licensed operators, spectrum management, and interconnection, amongothers. Interconnection, mostly fixed-mobile interconnection has been one of the most troubling issues that regulators have had to deal with. Current models being applied tocalculate interconnection rates have been contested by most of the fixed-line operators.

2.4 Market developmentFuelled by competition and the introduction of prepaid services, growth of mobilecellular and other wireless technologies in the past few years has been exponential. As a matter of fact, the latest report of the ITU has identified the African region as theworld's fastest growing mobile market with an annual growth rate of 62.4 percent.

According to the 2004 ITU African Telecommunications Indicators report, the speed of the mobile market growth can be attributed to demand, sector reform, the licensing ofnew operators, competition and the emergence of major strategic investors.

The report also states that mobile penetration in Africa reached 6.2 per 100 inhabitants by the end of 2003, compared to just 3 per 100 inhabitants for fixed line. Penetration is higher in sub-Saharan countries where three out of four telephone subscribers aremobile users. Nevertheless, Africa's overall mobile penetration is the lowest of anyregion at 6 percent in 2003 compared to the global average figure of 22 percent. Europe has a percent market penetration of 51.3, while North and South America has acombined 29.9 percent mobile market penetration.

Despite all these policy and market developments, the fixed-line business in Africa is anear monopoly, with each country having usually just one carrier. In stark contrast,more than 66% of African nations now allow competition in the mobile networks.

3. Open Access in the African context

The open access concept as applied to telecommunication is about network operatorssharing their facilities with other operators or service providers. Open access allowsmultiple service providers to deliver services, content or applications over sharednetworks.

In the past, technology obstacles imposed legitimate barriers to open access. All parties who wanted to share a network had to agree to a standard. New service providerswilling to provide telecommunication services, such as voice telephony, had no otherchoice than to build their own networks. Even after investing in new networkinfrastructure, they faced difficulties in getting customers as those who were alreadybeing served were nearly “locked” to their current service provider. On top of this,there were little or no incentives for investors to risk their funds in the market, alreadydominated by the incumbent.


Nowadays the telecom landscape is changing and technology barriers to open access are being naturally removed. Next-generation networks that allow multiple providers todeliver services and content over shared media networks such as cable and fixedwireless networks are already being used.

Open access policy stimulates competition, and competition stimulates growth,contributing to lower costs and providing a wider spectrum of choices to consumers.Innovative uses of new technologies (including wireless and IP technologies) at the“outer edges” of national backbone networks are revealing new opportunities to grownetwork connectivity organically and incrementally “from the edges”, a significantdeparture from the traditional model of growing networks “out from the centre”, as the intelligence was within the network.

3.1 IP Networks The Internet Protocol, IP, is probably the world's single most popular networkprotocol. Data travels over an IP-based network in the form of packets; each IP packet includes both a header (that specifies source, destination, and other information aboutthe data) and the data of the message itself.

The most well-known IP network is the Internet. The Internet is an open, publicnetwork that does not belong to any specific commercial interest. The open nature ofthe Internet has enabled its fast growth and the appearance of innovative Internetservices. Today, we can easily play games, watch movies and even do our shopping over the Internet. These services are being provided by different service providers, located in different parts of the world. This would probably not have been possible if the Internet was a closed network.

IP networks are not only facilitating the appearance of a number of services, but theyare also killing off the concept of distance. How much would it cost to send a 2 pagefax from Mozambique to the United States using traditional systems? And how muchwould it cost to send the same fax from Mozambique to South Africa? Certainly thecost will differ depending on many factors, such as duration of transmission and rates to each destination. But sending an e-mail to anywhere in the world will cost you the same, regardless of the distance between countries In an IP environment, network operatorscharge for the cost of transporting packets, and not the service that is being providedover the network.

IP networks are a practical proof of the benefits of open access concepts. We haveseen the growth of the Internet, and we all benefit today from instant access to various sources of information. It would be good to ask ourselves, how was the growth of theInternet possible in such a short time, if we compare with telephony? Maybe the answer lies within the fact that the openness of the Internet has allowed anyone to exploit theopportunities it has been presenting The transparency of the network and its reliance on distributed intelligence, foster innovation and empower users at the ends of thenetworks.


3.2 Should Africa adopt open access?This question can be put in another way: will open access bring any added value toAfrica? While the concept of open access is theoretically acceptable, much ground work needs to be done in order for it to be adopted by African markets. First and foremostAfrican leaders need to understand the current developments in telecommunicationmarkets and their trends. It is not that they should understand the technical ortechnological concepts regarding the development of telecommunication markets; allthey need to understand is that the older and closed models are no longer working, and new models such as open access are coming and should be adopted. Service providersno longer need to be the same as network operators. By adopting open access, Africanmarkets would benefit from:

• Increased market penetration• Avoidance of infrastructure duplication, as service providers would be sharing

infrastructure• Stimulation of innovation, as it takes place outside the network. In the case of IP

infrastructure, it allows anyone to create services• Consumers would have a wider pool of service providers to choose from, and thus

there would be more competition in the market and a consequent reduction in price

• Network operators would concentrate on network management and leave service provision and customer management to service providers

Keeping in mind the success of IP networks, specifically the Internet, it is difficult toimagine that Africa would lose anything by adopting open access policies to enablemarket development.

3.3 Barriers to Open Access in AfricaImplementing open access in Africa will not be a straight forward process. Most partsof Africa are still using equipment that was originally designed for closed environments, thus not supporting the requirements of open access. The other main constraint foropen access in Africa is the fact that in general, network operators are the same as theservice providers. The proliferation of vertically integrated companies in the continentmakes it difficult for regulators to enforce open access mandates. By separating network ownership from service provision, executives would no longer be focused on managing the network and at the same time worry about service development.

Lack of broadband infrastructure in Africa is also another constraint, as an increase inthe number of service providers sharing the same infrastructure would require morebandwidth. As said earlier, bandwidth is still a major problem in Africa.

Therefore, for open access to be a reality in Africa, bandwidth constraints andregulatory issues need to be resolved, at national and continental levels.

4. Conclusion

To meet the needs of growing demand for access to telecommunications infrastructureand more bandwidth in the new era of information society, the difficulties and problems


associated with policies regulating access to telecommunication infrastructures need tobe addressed. In addressing these problems, policy makers and regulators must findways to stimulate reduction in access costs, while allowing for network expansion.

In order for Africa to catch up with the rest of the world in terms of telecommunication developments, it is no secret that huge investments are required. In fact, currentoperators and service providers may not have the required capital to invest in themarket. Even so, current costs associated with deployment of telecommunicationservices in Africa are not that attractive to investors as the return on their investmentwill be slow, mostly due to the low purchasing power of the majority of Africanmarkets. As such, innovative policies are required, and open access should be seriouslyconsidered by Africans. The benefits that it brings could make it worth taking the riskof implementing it.

5. References

Jonathan Sallet, “Just how open must an open network be for an open network to belabelled "open"?” at http://www.firstmonday.org/issues/issue8_3/sallet/#s1

Mark Cooper, “The Public Interest in Open communication Networks”, ConsumerFederation Of America, July 2004

Osei Darwa and Fikile Mazibuko, “ Creating Virtual Learning Communities in Africa:Challenges and Prospects:, athttp://www.firstmonday.org/issues/issue5_5/darkwa/index.html#author

Muriuki Mureithi, “African Telecommunication Infrastructures for InformationAccess”, at http://www.american.edu/initeb/qj8944a/telecom2.htm

Jabulani Dhliwayo, “Developing a Fibre Optic Backbone for Africa”, NEPAD Council. See article in this publication.

122

Developing a Fibre Optic Backbone for Africa

Jabulani Dhliwayo

ABSTRACTThe high cost of telecommunications as a result of poor infrastructure in Africa, the fact that 90% of calls from African countries to other African countries are routed throughEurope at a cost of $400 million a year and the fact that current percentage populationpenetration in Internet usage is only 1.8% (with South Africa and North Africancountries dominating this statistic) give a compelling case for an all Africa optical based long-haul backbone network. For such an African backbone network to be successful, it has to be initiated and planned by Africans on the basis of potential long-term ability to help build wealth – not on the basis of short-term return on investment. In this paper, the status of African telecommunications is briefly reviewed and an argument for a fibreoptic backbone in support of planned and completed submarine cable systems arepresented. NEPAD Council’s guidelines for a successful long haul backbone networkare provided as a foundation for future network planning.

1. Introduction to NEPAD Council

NEPAD Council is a non-political, independent not for profit organization founded by young African experts and professionals and in which every African country isrepresented. The council is made up of 150 members – 3 from each of 50 Africancountries. Members of NEPAD Council include scientists, university professors,medical doctors and economists in Africa and abroad. The main objective of NEPADCouncil is to support the New Partnership for Africa’s Development (NEPAD) created by African heads of state and aimed at fighting poverty, consolidating democracy andgood governance, fostering trade, investment, economic growth and sustainability.

NEPAD Council will address several issues and problems affecting the Africancontinent through four commissions: Education; Science, Technology and ICT, SocioEconomic Development and Trade and Global Health.

One initiative of NEPAD Council’s Science, Technology and ICT commission is tosupport projects that address the connection of African countries to one another and to the rest of the world by high bandwidth fibre optic networks.

Developing a Fibre Optic Backbone for Africa | 123

2. Current status of African Telecommunications

Current telecommunications infrastructure in Africa consists of a combination of radiorelay links, open wire lines, radiotelephone stations, fixed local loop installations andsubstantial mobile cellular networks. In some African countries, mobile cellularnetworks have increased significantly over the past few years to match or even surpassthe number of fixed lines. For inter-state communications, satellite and microwave links are mostly used. Two African organizations administer these networks: Regional African Satellite Communication organization (RASCOM) and the Pan AfricanTelecommunications Union (PATU). Despite the existence of these authorities, about90% of African traffic is routed through Europe or North America. Consequently,African states pay about $400 million every year to have calls to other African countries routed through Europe or North America.

While for most African countries connections to overseas destinations are much betterthan connections to other African countries, international connections are still veryinadequate. There are less than 30 000 circuits interconnecting Africa with the rest ofthe world and most international traffic transits through Europe [1]. Thus a significantportion of the revenue from international calls is used to pay European networkoperators making it difficult for Africans to upgrade their infrastructure. In addition,due to the old infrastructure, there is an additional loss of revenue due to service theftand the inability to accurately bill communication services.

Internet usage has only just begun in most of Africa and is currently concentrated inlarge cities. Because of the large number of shared accounts and the high usage ofpublic services such as Internet cafés, it is difficult to accurately determine the numberof people with Internet access. However, Internet World Stats [2] estimates the averageAfrican Internet usage penetration at about 1.8% by 2005. A few African countries such as South Africa, Mauritius, Seychelles, Ile de Reunion and Egypt dominate this average– the situation is much worse for most sub-Saharan African countries. The low Internet usage is affected mainly by the inadequacy and poor quality of the telecommunicationsinfrastructure and the high cost of Internet service mostly through dial up charged perminute. Only a very small percentage of Internet access is Broadband or high speed,mostly through asymmetric digital subscriber lines (ADSL) and broadband wireless. Insome African countries, leaders and policy makers are not very supportive of theInternet for several reasons; lack of awareness of the long term benefits of the Internet, fear of the ultra freedom of communication and the “cultural invasion” associated with it and other competing priorities. Figure 1 shows selected global Internet penetrationwith Africa trailing every other region in the world.


The African telecommunication status - poor infrastructure, high cost of telephone calls to other African countries, loss of telecommunication revenue to Europe and lowinternet usage-can only change if a major effort is undertaken to develop a continentalfibre-optic backbone that interconnects Africa and the rest of the world via current and planned undersea fibre-optic cables. National and regional networks should therefore be planned so that they can come together to form a continental network.

3. Current status of fibre optic links in Africa

3.1. Submarine networksSome North African countries - Egypt, Tunisia and Algeria – are connected by one ofthe longest undersea fibre-optic cable: SEA/ME/WE-3 ("South East Asia -Middle East Western Europe-3"). SEA-ME-WE-3 includes 39 landing points in 33 countries and 4continents from Western Europe (including Germany, England and France) to the FarEast (including China, Japan and Singapore) and to Australia. Maximum capacity of this two fibre pair cable is 505 Gbps. The Mediterranean section of this submarine systemconnecting the North African countries is shown in the map of figure 4. An additionalcable, SE-WE-4 is currently being added to add capacity.

The first major attempt at an all Africa fibre optic cable was the ambitious projectdubbed “Africa One” by Africa One Ltd owned by Columbia Technologies of NewJersey, USA. Africa One Ltd was to develop and own a state-of-the-art undersea fibreoptic self-healing loop around Africa. Lucent Technologies were selected as thepreferred equipment and software supplier while Global Crossing Ltd. was to provideproject management and undersea construction. The network was to provide end-to-end connectivity for the African continent to 19 countries and 185 cities around the


world. The project collapsed because of the global downturn in the telecommunications industry and the subsequent collapse of some of the major partners. The failure ofAfrica One should provide Africans with valuable lessons on their plans for buildingfuture fibre-optic systems, especially on how these projects should be financed and thechoice of partners.

Perhaps the most successful attempt at an African fibre network so far is the twosegment submarine cable system; SAFE (South Africa - Far East) which links Malaysiaand India in the east to South Africa via Mauritius and Ile de Reunion and SAT-3/WASC (South Africa Trans-Atlantic - West Africa Submarine Cable) which continuesfrom South Africa to Portugal and Spain in Europe with landings at a number of West and Southern African countries. The funding agreement for the project was signed in1999 and President Wade of Senegal, one of the founding members of NEPAD,officially launched the networks in Dakar in May 2002. The original capacity was 20Gbps and is upgradeable to 120Gbps. The 20Gbps is reportedly fully subscribed and is in the process of being upgraded to 40 Gbps. The submarine cables span a total of28,000 km and connect the countries of Portugal, Spain (Canary Islands), Senegal,Ghana, Benin, Cote D’Ivoire, Nigeria, Cameroon, Gabon, Angola, South Africa, France (Ile de Reunion), Mauritius, India and Malaysia. Figure 4 also shows the 9 landing points in 10 African countries. However, the fruits of this submarine cable are yet to be fullyrealized; the bandwidth take-up on the system is very low and very expensive compared to other submarine cables – about $6000/Mbps/months (about the most expensive inthe world). The graph of figure 2 compares SAT-3/WASC bandwidth cost with othercable worldwide. The high cost is attributed to lack of competing infrastructure.

Another major submarine cable – East Africa Submarine Cable System (EASSy) –spanning the East African cost from Mtunzini (South Africa) to Port Sudan is now at an


advanced stage. This new submarine cable will connect SAT-3 to SE-WE-3, therebycompleting a ring of undersea cables around Africa. Other landing points will be inMaputo (Mozambique), Mahajanga (Madagascar), Dar es Salaam (Tanzania), Mombassa (Kenya) and Mogadishu (Somalia). Like the other cables already installed, EASSy will be a two pair fibre cable with a proposed capacity of 16 or 32 wavelengths at 10Gbps or a maximum capacity of 320Gbps. The NEPAD e-Africa commission has initiated abackhaul system for EASSy that would connect the landlocked countries of East andSouthern Africa.

Another important project in the planning stage is the West Africa Festoon System that will complement SAT-3/WASC that would connect Gabon, Equatorial Guinea,Cameroon, Nigeria, Sao Tome, Principe, Angola, DRC, Congo and Gabon.

3.2. Other National and regional networksAt the national and regional levels there are already significant efforts to provide fibre-optic and microwave backbones. Egypt, for example, has an SDH ring that connectsmost of its cities to undersea fibre cables such as SEA-ME-WE-3. In Zimbabwe,TeleAcess is constructing a fibre-optic link between Zimbabwe and South Africathrough Beit Bridge. In Nigeria MTN constructed a national microwave backbonenamed “Y'helloBahn” at a cost of $120 million and Globacom is about to construct asubmarine fibre cable from Nigeria to England. In Southern Africa, there is an ongoing program known as the SADC Regional Information Initiative (SRII) to Interconnectcountries of Southern Africa. The project involves upgrading the links between SouthAfrica and Zimbabwe, and between Zimbabwe and Botswana and Zimbabwe andMozambique. The COMESA – COMTEL project which has been in the planning stage for a couple of years seeks to connect all the countries of COMESA by a terrestrial fibreoptic network.

Some of the national/regional backbone projects have taken advantage of electricitygrids, railway lines and oil pipelines and their right of passage to install fibre-optic cable. In South Africa, the electricity parastatal ESCOM has been installing fibre along it’s grid in South Africa and in neighbouring countries. In Namibia, NAMPOWER is installingfibre-optic lines along all their new power lines while in Zimbabwe, PowerTel isconnecting major cities. In West Africa, the Volta River Authority, which distributeshydroelectric power from the Volta River, has been planning to install fibre cable along its grid. The Volta River Authority has already become a licensed telecom operator inGhana. A fibre link between Mombassa and Nairobi in Kenya has been planned to take advantage of the suggested EASSy landing points in Mombassa.

A number of railway operators in Southern Africa have joined forces to establish abroadband fibre optic link travelling from Tanzania through Malawi, Zambia, Namibia, Malawi and South Africa. The South African rail transport parastatal, TRANSTEL, isestablishing a national broadband network along its railway grid in preparation for entry to the market as part of a consortium to bid for the upcoming second nationaloperators' license.

4. Why fibre optic networks?


One of the most likely reactions one gets when discussing fibre-optic networks in Africa is “why not satellite technology?” Satellite communications has been around for a while and has provided telecommunications links between Africa and the rest of the world.However, a comparison between fibre optic and satellite technologies reveals thatalthough satellite systems are the most efficient solutions for TV broadcast, for accessto remote locations, and essentially, for wireless access to the local loop and thenetwork backbone, fibre optic networks are more suited for high bandwidthtransmission between countries and continents though core networks (or backbones)and submarine links respectively.

Fibre optic networks offer very high bandwidth necessary for African nations to catchup with the new global information technology. For example, fibre cables today canhave capacity up to 2 Tbps - an equivalent of millions of simultaneous voice channelsper cable. This is far from the reach of any anticipated satellite system, which is less than 1Gbps - lower than Africa’s own SAT-3/WASC/SAFE undersea cable system.

Real time transmission and very low bit error rate offered by fibre optic networks areamong the advantages of fibre over satellite. Satellite communications add a delay tocommunications making interactive data transmission difficult and subject the quality of transmission to external factors. A geostationary satellite link has a transmission delay of up to 600 milliseconds compared to 100ms for a combination of fibre and coaxial cable networks.

The open space nature of satellite (and any other wireless) communications makessatellite communication vulnerable to interception and corruption. Although severalschemes are available for data encryption for IP over satellite, the high bit error ratemay cause failures in the encryption systems. Fibre optic transmission offersundoubtedly the best confidentiality and security of transmission than any other meansby its mere nature.

In order to address increasing traffic demand, it is relatively easy to increase the capacity of fibre optic networks during their lifetime by means of wavelength divisionmultiplexing technology. For example, the SAT-3/WASC/SAFE system can beupgraded 12 fold from 10Gbps to 120Gbps. It is impossible to do a similar upgrade on satellite systems.

Perhaps the main disadvantage of satellite communication is their high cost relative tofibre optics communication. In the US, for example, the monthly rate for broadbandconnectivity through cable is about $35 for 3Mbps data rate compared to $200 for200Kbps by Satellite. While the initial cost of a continental fibre optic network forAfrica may appear too high, the long term cost savings over satellite transmission areoverwhelming.

Thus due to their high bandwidth, high reliability, high signal quality, long lifetime,better security and low service cost, fibre optic networks are suited for inter and intracontinental backbone network infrastructure. On the other hand, satellite systems aremore dedicated to video broadcasting and personal communication services such asmobile telephony satellite or to access remote areas.


5. Achieving NEPAD objectives with a successful network.

The availability (or of lack thereof) of a modern information technology infrastructurein Africa will have a profound impact on most of NEPAD’s principles of sustainabledevelopment. Education, trade and investment, African-ownership, digital opportunities and globalization are among NEPAD’s objectives that will be facilitated by theavailability of a high-speed fibre-optic network.

5.1. EducationWith the rise of the Internet, education has been completely transformed. Distancelearning, for example, used to be largely a lonely experience. In today’s information age, students do not only overcome difficulties interacting with the tutor, but can now easilyovercome the nightmare of waiting for tutorials in the mail for long periods of time. Inaddition, the Internet constitutes a virtual classroom in which interaction can take place between students anywhere in the world. The information age has seen the accelerationof research at educational and other institutions because of the abundance of scientificdata on the Internet and the advent of electronic journals. The cost of electronictechnical information is negligible compared to traditional research journals. Yet in most of Africa education has not taken advantage of the information explosion; distanceeducation is mostly what it was several decades ago, most libraries at universities andresearch institutions still rely on hard copy journals which are way too costly. In mostcases by the time the journals get to Africa, the information is obsolete and makes itimpossible for the quality of research in Africa to be comparable to that in developedcountries.

5.2. Investment and globalizationThe unprecedented exponential growth in telecommunications infrastructure andtransmission capacity has changed the way all businesses function in developedcountries. Corporations, institutions and governments in most of the world can nowinteract effectively with their clients, vendors, and their global branches without leaving the office. Teleconference technologies such as Net-Meeting and video conferencing are becoming very common in these countries, driving the demand for network bandwidth and the push towards fibre to the premise. Most Western corporations now base theirdecision to establish a business presence or manufacturing facilities abroad on theavailability of modern telecommunications infrastructure. Companies that are relocating or expanding want to be located where they can take advantage of the informationtechnology explosion. The lack of a viable broadband infrastructure in Africa will cause the continent to plunge deeper and deeper into global isolation.

According to Addison and Heshmati [4] the distribution of foreign direct investment(FDI) across developing countries is highly unequal (15 countries account for over 80per cent of FDI to developing countries) and very poor countries face major difficulties in attracting FDI. Africa’s share of FDI is less than 1% with most of it concentrated in only a few countries including Angola, Morocco and South Africa. One of the mainfactors preventing FDI into Africa is lack of infrastructure. Figure 3 from Booz, Allenand Hamilton shows infrastructure (including telecommunications infrastructure) as the main factor affecting FDI. Most Western companies are reluctant to invest in African


telecommunications infrastructure because they use Africa’s low purchasing powerparity (PPP) as an indicator of potentially poor return on investment. Yet a robusttelecommunications infrastructure is a pre-requisite to building wealth and to improving the PPP.

In North America and Europe, technology companies are cutting costs by movingsoftware and technical support jobs offshore and creating thousands of jobs for India'slow-cost engineers. The Indian software workers with two years of experience are paid$545 a month. Although this is about one sixth of what their U.S. counterparts earn, it is much higher than the average salary for an African software engineer. Sadly, Africashould not expect such fortunes until there is a reasonable digital infrastructure in thecontinent.

5.3. African Partnership and OwnershipAfricans do not own most of the limited network infrastructure available in Africa.Networks (land and terrestrial) were planned to facilitate communications betweenindividual African countries and Western countries making it difficult for Africancountries to communicate with each other. Traffic from one African country to another is oftentimes routed through Europe or North America making it very expensive asAfrican countries are required to pay Western network carriers. It is much easier andless expensive for somebody in Mozambique, for example, to communicate withsomebody in the US than it is for him or her to communicate with somebody inZimbabwe, 50 miles away! Unless a network infrastructure is initiated and owned byAfricans, easy communication between African states will remain a dream and themeans of communications will remain foreign owned.


6. NEPAD Council Guidelines for a successful African backbone

One of the main objectives of NEPAD Council is to address digital or informationtechnology in Africa. We present some guidelines for a robust continental high-speednetwork in support of NEPAD initiatives.

6.1. Continental view on network developmentMost of the work on fibre-optic links in Africa has been done without a wider view ofhow these links will come together to form an Africa wide network. We need to identify and develop a “big picture” of the African transport network to which national andregional networks will be connected. It is imperative that those who are in the processof planning fibre-optic links do so with this overall picture of how the different sub-networks will be interconnected. Figure 4 is an example of how an African backbonethat completely utilizes the current and planned submarine networks could look like.

6.2. Network partitioningAs optical communication technologies, photonic components and service networksevolve, networks become more and more complicated. The ITU has come up with anetwork-partitioning model to analyze complicated networks. The model can bepresented in a vertical or horizontal view [5, 6]. Our focus is on the horizontal viewwhich partitions networks as core, access, customer premises, service node interface and user-network interface. The advantages of such a model as listed in reference [7] are:• It is simpler to design and operate each layer separately than to design and operate

the entire network as a single entity.


• Each layer can have its own independent strategy for operation and maintenance, such as for protection and restoration. The layered architecture minimizes the influences of operation and maintenance among different layers.

• It is possible to add or change a layer and even introduce new topologies without affecting other layers from an architectural point of view.

• Each network layer can be defined and specified independently.• It is simpler to define management boundaries when network operators jointly

provide end-to-end paths within a single layer.• It is convenient to define domain boundaries within the layered network of a single

operator. This enables apportioning of performance objectives to subsystems that compose the network.

• It is possible to define independent routing domain boundaries in relation to the operation of path management processes.

The core networks can be further divided into inter-state long-haul networks, intra-statelong-haul networks and junction networks. The proposed landlocked countries projectshould conform to the model of an inter-state long-haul backbone. Smaller Africancountries can be bundled into regions so that the network nodes do not necessarily have to be in every country. In the example network of figure 4 countries such as Uganda,Rwanda and Burundi or Ethiopia, Eritrea and Djibouti could share network nodes. It is recommended that the east Africa submarine cable be planned with the African corenetwork in mind. In particular, the landing points for the East-Africa submarine cableshould coincide with planned nodes for the African long-haul network. NEPADCouncil plans to work with the e-Africa commission to come up with a model for theAfrican backbone based on long-term traffic projections.

6.3. Choice of optical fibreIn comparison to network equipment, the selection of fibre for use in a long haul fibrebackbone is very critical because fibre deployment is very time consuming andexpensive. In the case of underground networks, fibre should be laid once for a verylong time. The decision on the type of fibre for Africa’s long haul network backboneshould not be based on short time cost savings but on longer term cost effectiveness.Careful study of projected traffic patterns and network performance analysis shouldgive guidance in the selection of fibre to be used in this long haul continental backbone. ITU G.655 compatible non-zero dispersion shifted fibre tends to be the fibre of choice for long haul networks. Other more effective solutions are currently being developedand should be investigated.

6.4. Which network serviceWireless cellular is becoming one of the most prominent telecommunication means for rural Africa. For high bandwidth cellular communications, Africa will have to upgradeto the third generation (3G) wireless technology, which provides up to 2Mbpscompared with 64Kbps for 2G systems. 3G wireless technologies will require abroadband transport technology solution based on a cell- or packet-switched accessnetwork such as ATM. A fibre optic backbone for Africa should be able to support this protocol.


On the other hand, the most common underlying global transport mechanism isSONET/SDH time division multiplexing. Most digital networks in Africa (such asSAII) are being upgraded from the older PDH technology to SDH. Some Africancountries are already planning for gigabit Ethernet in their fibre networks – takingadvantage of the low cost of the service.

Like in every other region of the world, the Internet will ultimately drive bandwidthdemand in Africa. It is important that as Africa plans its network backbone, it plans tofully support the Internet protocol, IP. With the advent of high-density WDM networks plans must be made for optical multiplexed telecommunication services where different wavelengths can support different "electrical" multiplexed services. For example, anumber of wavelengths should be dedicated to high bandwidth optical IP network,others should be dedicated to optical ATM networks and other wavelengths should bededicated to traditional SONET/SDH services while others can be dedicated to gigabit Ethernet.

7. Conclusion

The current African telecommunication infrastructure is currently in a sorry state thatcan only be improved significantly by the installation of optical transport networks.There is a strong commitment to building submarine and national and regionalbackbones in Africa. To connect all African networks to the global “superhighway”,NEPAD has proposed a landlocked countries project. NEPAD Council offers someguidelines to the development of this network and future technical support in theplanning and network modelling of such a network.

8. Acknowledgements

The author would like to thank Dr. Jim Grochocinski, the Rev. Dr. Mark D. Vaughn,Rich Wagner, Paul Robertson, James Trice and Linda Gaskill of Corning Incorporatedfor helpful discussions and support. The author would also like to thank members ofNEPAD Council for their support.

9. References

1. William C. Marra, and Joel Schesser, “Africa ONE: The Africa Optical Network”IEEE Communications Magazine, February 1996

2. http://www.internetworldstats.com/stats.htm

3. Jean-Marie Beaufils , “How Do Submarine Networks Web the World”, Optical FibreTechnology vol . 6, no. 15, p32 2000.

4. Tony Addison and Almas Heshmati, “Democratization and New CommunicationTechnologies as Determinants of Foreign Direct Investment in Developing Countries”World Institute for Development Economics Research of the United NationsUniversity (UNU/WIDER) http://www.wider.unu.edu/


5. International Telecommunications Union, “Generic Functional architecture oftransport networks”, ITU-T Rec. G8.805, version 11/1995.

6. International Telecommunications Union, “Architecture of transport networks based on synchronous digital hierarchy (SDH),” ITU-T Rec. G.803, Version 6/1997

7. Leping Wei, Yi Chen and Gerald G. Wong, “The evolution of China’s opticalnetworks”, Bell Labs Technical Journal, January-March 1999.

134

Innovation to Improve Access to ICT

Martin Curley

1. Introduction

We are living in an era of dramatic change. Conventional thinking and technologyhowever, will only lead to conventional solutions which will improve but not transform access to ICT. If we continue to follow the conventional approaches that exist for ICT, the digital divide will in fact grow larger, not smaller. Even developed countrygovernments are worldwide experiencing difficulties in keeping pace with technologychange and the acceleration of the information society. Despite this there is increasingevidence that societies24 that invest more heavily in ICT are those that will reap thedigital dividend whilst laggards will be left behind on the other side of the digital divide. To drive a more concerted adoption of ICT worldwide, different and more lateralthinking is required.

Whilst governments in developed countries face economic and political pressure todemonstrate that their investment in ICT increases the effectiveness and efficiency ofservices, governments in developing countries face the real challenges of leveraging ICT to improve basic national services such as education and healthcare. ICT, if properlyapplied to a particular area such as healthcare, can help lower the cost of care but canalso help significantly improve quality of life. It is increasingly clear that while ICT is an unconventional resource, there are a few models or role-models to help governmentsmanage their ICT portfolio in a holistic fashion.

2. Integrated Digital Strategy and Common Infrastructure and Services

Governments should not, nor are they able to afford to address ICT in a vacuum butneed to address it as part of an integrated digital strategy for their country and for thedifferent services that need to be provided. There is increasing logic to suggest thatgovernments should manage their ICT infrastructure and the applications of ICT in the same way that enterprises run their ICT infrastructure and applications. A key tenet ofthis approach is the concept of shared infrastructure and services where the sharingresources provide a foundation for better service at lower cost. The idea of having acommon approach to supporting solutions (covering the range of solutions thatgovernments need to operate) on a standardized architecture and a converged platform may sound like nirvana, but this makes sense from an economic, integration andfunctional perspective. Where multiple government departments and operations share

24 Van Ark, Bart et al, Research Memorandum GD-60, University of Groningen, The Netherlands, 2002

Innovation to Improve Access to ICT | 135

the same back-end infrastructure, the same support approaches and the same commonnetwork, these costs can be shared across all the different departments, giving an overall lower cost of ownership for each type of operation and improving the end result.

The appointment of a national CIO to work on developing a national infrastructure and a common services and support approach to best meet the evolving needs of thecountry, whilst working within a limited budget would seem to make much sense. Fromone perspective, developing countries may be in a better position than the moredeveloped countries since they typically will have fewer legacy systems in place whenadopting a common services based approach. Indeed, the emergence of ServicesOriented Architecture is a key enabler to allow this common services and infrastructuretype of approach. One example of the common infrastructure approach at work is inWestminster City Council (WCC) in the United Kingdom. The WCC is deploying awireless infrastructure to support a closed circuit television network intended formultiple usage models including surveillance, noise monitoring and for viewing thefunctional operation of street lights. Additionally, WCC plan to deploy other servicessuch as e-Learning and e-Health solutions over the same network, increasing theutilization of the deployed assets and significantly reducing the unit cost per solutionand user.

3. Lowering ICT unit costs through improved ICT Management Practices

Governments can significantly increase access to ICTs if they can lower ICT unit costs, thereby allowing much more to be done with the same amount of money. Thus a keystrategy for governments should be to attempt to lower ICT unit costs on an ongoingbasis. This can be achieved through a combination of measures including takingadvantage of Moore’s Law 25 and improving the management practices for ICT in aparticular country26.

Lowering ICT unit costs also requires new approaches to cost effective deploymentmodels, drawing from the experiences that many global corporations have had indeploying technology. Central procurement, standardized platforms and builds, buyingPCs with headroom to maximize useful life, central support centres and down the wire management are all practices which are used effectively today by many corporations and these practices must be implemented by governments and regions to drive down theTotal Cost of Ownership (TCO).

Standardizing an infrastructure for both front-end and back-end computing savesmoney from both a capital and an operating expenditure standpoint. Governments that can aggregate demand and negotiate volume discounts from vendors for a standardplatform will typically receive significant cost reductions. The benefits ofstandardization in terms of ICT unit cost reduction are often more strongly felt in

25Moore’s law was forecast by Gordon Moore in 1965 and today means that raw computing performance

effectively doubles every eighteen months or so at less or equal cost than the previous capability.26

More details can be found in Curley, Martin; “Managing Information Technology for Business Value” Intel Press, 2004.


operating expenditure savings since standardization means that typically less staff andoverall spare parts may be required to support systems, thereby lowering overall cost.(Organizations typically spend three to five times the original capital investment in asystem to maintain it over its lifetime).

In terms of choosing front-end computers for education, governments may very well be tempted to buy the lowest specification of PC available to maximize the number of PCs which can be purchased with a constrained budget. This might not be the wiseststrategy. Intel made the decision in the late 1990s to purchase value PCs instead ofPerformance PCs for Intel employees – this turned out to be a poor decision as we had to write off the twenty thousand PC’s purchased after only one and a half years ofoperation, when we needed to do an operating system upgrade and install some newapplications that needed more computing power. Specifying the right PC is a keydecision in better enabling access to ICTs.

One method of calculating what PC specification to buy and what the appropriaterefresh cycle should be, is to calculate the “equivalent annual cost” (EAC) of differentspecifications and different refresh cycles. (EAC is a fairly common financial method to compare capital investments with different lifetimes and is an annuity that has the same present value and life as the underlying cost flow). In determining Total Cost ofOwnership for PCs, governments should take into account all the cost components of a PC’s lifecycle including PC deployment, PC usage, PC support and PC retirement costs. Calculating the lowest equivalent annual cost of different configurations helps inchoosing the right system and refresh cycle for PCs.

Consolidating back-end systems to reduce the number of servers required can be asignificant tactic in reducing the unit cost of ICT. Again this is a practice that manycorporations have used to reduce ICT unit costs where for instance email serverconsolidation has been a frequent method for quick cost reduction. The ever improving price/performance ratio of communications, particularly the role of wireless as adisruptive technology, is also an enabler to consolidate multiple data centres into fewer larger data centres. Also as blade technology becomes more mainstream, computerdensity in computer rooms can be significantly increased, again lowering cost. 27

4. Disruptive Technology and Win-Win Business Cases

Countries also require a new kind of “ICT for development” business case, one whichtakes advantage of Moore’s Law and delivers both new business value and alsoimproves ICT efficiency. Collectively we need to understand how ICT can be used tocatalyze and transform society, not merely enhance it. As ICT becomes more pervasive we need to find solutions which actually lower aggregate and unit IT operating costswhilst at the same time increase access and coverage.

27Blade servers are the next step in the evolution of dense rack mounted computers. Blade computing

introduces a new data centre paradigm where many ultra-thin computer blades share centralized resources in a single chassis. The dimensions of blade servers can allow more than 300 blades to be housed on a single rack.


Disruptive technologies and breakthrough thinking is required. Emerging technologiessuch as Wi-Max and peer to peer computing hold great promise. Wi-Fi networks already enable fast and low cost establishment of local area networks. This phenomenon isalready changing the speed at which communities can be networked. It is alsosignificantly improving the economics of getting access since provisioning a wirelessLAN network is significantly cheaper than having to install a cable. Wi-MAX will likely have a more dramatic impact as it will enable broadband connections at high speed atdistances of over 20 km from a single base station which will enable cost effectivebroadband for rural areas as well as metro areas.

And Wi-MAX is not just hype but it is real. As the people of Banda Aceh, Indonesiahave been rebuilding their lives after December 2004’s catastrophic Indian Oceantsunami, Intel, together with others, Intel has quietly been helping reconnect that part of south Asia with the rest of the world. This, in the form of a very large wirelessbroadband ‘umbrella’ that lets humanitarian and disaster relief groups in hard-hit Banda Aceh communicate with each other and the rest of the world.

In June 2005, engineers flipped on a pre-Wi-MAX network that today covers some 600 square miles (1,500 square kilometres) of the Aceh province, where the tsunamiwreaked the greatest destruction. The network consists of three pre-Wi-MAX basestations providing high-speed Internet data connection at speeds up to 6 megabits persecond within the coverage area, and 28 megabit per second backhaul connectionsbetween base stations and connections to multiple small satellite terminals, calledVSATs. This relief project demonstrates that the technology works and can operate inthe harshest of environments. The economics are compelling and Wi-Max will enablenew connectivity in the coming years that until recently was inconceivable due to bothcost and distribution factors.

Meanwhile, peer-to-peer computing can enable pseudo broadband performance overnarrowband pipes and drive network efficiencies by moving a significant proportion oflarge file transfers off expensive wide are networks onto local area networks. There aretwo immediate benefits from this: firstly, large files can be downloaded much faster and secondly, cost savings can be achieved through minimizing wide area network transferof files.

In parallel, new virtualization technologies will allow multiple users to run differentapplications simultaneously on the same PC with multiple screen interfaces.New hardware such as the HP 441 desktop solution allows four simultaneous users tointeract with a PC, driving down both capital requirements and overall TCO –(estimated 50% reduction in capital acquisition costs and 65% in systems managementcosts28).

28HP 441 Product Specification http://h20247.www2.hp.com/PublicSector/cache/119709-0-0-225-

121.html


Open source can also be a key tool in particular circumstances to lower overall TCO.One example is the region of Extremadura in Spain where the local authorities haveachieved a 2:1 student: PC ratio through using a Linux based approach. All in all,disruptive technologies continue to emerge and they will continue to dramaticallyimprove access to ICTs.

Because of the remarkable price/performance of Information Technology (primarilydriven by Moore’s Law) governments should, in some cases, over-invest in ICT withrespect to other resources to drive lower overall total cost of operations of the complete intended system (such as health, education or other government services). Aninteresting example is the Higher Colleges of Technology in Abu Dhabi, UAE thatinvested in a wireless notebook approach for students. By taking this approach theyavoided the building of fifty computer labs, lowering the overall total cost of operationfor the college. Moving from a lab based computer approach to an integrated wirelessnotebook approach can deliver a significant improvement in education performance aswell as lowering overall total cost of operation of the education system.

In developing countries government policy changes will likely need to change to drivebetter adoption of ICT. For example only 41% of developing countries allow use of Wi-Fi in unlicensed spectrum compared to more than 95% of developed countries.

5. Conclusion

Many people acknowledge that rich content is the biggest gap or issue in increasedadoption of ICT. If we take the example of education, teachers, armed with theappropriate tools and the appropriate standards, can and should be encouraged todevelop curriculum focused content which can be re-used by other teachers. The Intel® Teach to the Future program has trained more than two million teachers worldwideand these kinds of programs rapidly improve teacher competency in using laptops in the classroom. A disruptive innovation that could transform the development andproliferation of content is the establishment of a Napster-like solution based on Peer-to-Peer networks which could enable easy sharing of teacher prepared content betweenschools and across borders.

Another tactic to improve access to ICTs is the increasingly popular concept of publicprivate partnerships. This occurs when both public and private sector organizationscome together to fund, develop and operate solutions or services. Intel’s ® Skooolproject is an example of this where public and private partners can come together todevelop and host curriculum based rich media to enable e-learning to be accelerated ona national scale. As we move into an era of third generation corporate socialresponsibility, corporations and governments can step up their collaboration to drivelarge scale changes that produce win-win outcomes. While developing overarchingstrategies that address this, we also need to continue to drive pilots and leveraginglearning from rapid solution prototyping.


Finally, governments should try to create virtuous circles of innovation throughcoordinated strategies on broadband deployment, PC purchase programs, digital literacy programs and online e-services provisioning. While each of these components havevalue in isolation, a network effect can only be achieved through the co-managementand co-evolution of strategies which co-evolve the 4C’s of ICT – Computing,Connectivity, Content and (Human) Capacity29.

294C’s explained by Mr. Rahul Tongia, School of Computer Science from Carnegie Mellon.

http://www.contrib.andrew.cmu.edu/~tongia/FiberAfrica--ending_a_digital_divide.pdf.

140

Open and Closed Skies: Satellite Access in AfricaPolicy Reform and Regulatory Issues in Bridging the Digital Divide through Satellite Technologies

© IDRC 2004An abridgement of the original IDRC funded report – 2005

Martin Jarrold

Acknowledgements

This original report on which this article is based was made possible through supportfrom the International Development Research Centre of Canada (IDRC). The printversion of the report was published in September 2004 in both English and Frenchlanguage editions, and has since been widely circulated throughout the regulatorycommunity in Africa and throughout the world. A soft copy, organised for bothnarrowband and broadband download is available fromhttp://web.idrc.ca/ev_en.php?ID=89311_201&ID2=DO_TOPIC.

1. Introduction

It is now widely recognised that access to information and knowledge throughaffordable communications represents a significant opportunity for social and economic development, for regional cooperation and integration, and for increasing theparticipation of people in the emerging global information society. Addressingdeficiencies in access to low-cost communication services is therefore now regarded asan urgent imperative for improving the quality of life in African communities, especially in remote and rural areas where the bulk of the population still resides.

But Africa is fragmented into many small national markets, and limited economies ofscale have combined with low-income levels to reduce the ability of telecommunication operators to provide services. Compounded by lack of competition in the sector, thishas resulted in low levels of investment in infrastructure. As a result, even where accessis available, costs often remain extremely high, especially outside urban areas. Althoughthere are a growing number of initiatives to expand terrestrial infrastructure, these areusually confined to the major cities and along trunk routes. As a result, the cost ofbandwidth for Internet and other services is generally 10-100 times higher than inNorth America or Europe.

Fortunately, satellite technology presents an immediate solution to this bottleneck, even in the vast terrain of rural Africa. The IDRC Pan-Africa Satellite Survey that providesthe basis for Open and Closed Skies: Satellite Access in Africa confirms that systems

Open and Closed Skies: Satellite Access in Africa | 141

using the new high-power satellites over Africa make it possible to obtain bandwidthanywhere in the region about 10 times more inexpensively than in the past.

2. African VSAT Regulation Today

A growing number of African Administrations have begun to implement policies andregulations that seek to open telecommunication markets to varying degrees ofcompetition. These policies are being applied to telecommunication structures that, onone level, have traditionally been remarkably uniform.30 Without exception, the sectorof each African country has been organised on the principle of national operatingentities having responsibility for providing telephone service. In some cases,international links were - and in some countries still are - the responsibility of a separate entity. Government ownership of operating entities has been the norm.

In some African countries that have adopted a liberalised regulatory framework, private VSAT networks are allowed to function under the authority of the incumbent operator, while the latter still retain a formal monopoly. There is also usually a limitation on theprovision of voice services.

Another common restriction in Africa involves limiting private VSAT networks only todomestic use. VSAT network operators may be required to route their private networktransmissions through the national hub of the incumbent operator, regardless of thefinancial or even the technical disadvantages this may have for private VSAT networkoperators. In some cases, obtaining a VSAT licence may require a bilateral arrangement with the incumbent operator with a “landing-rights fee” or tariff to be paid to theoperator, even if the incumbent does not participate in the service chain. In othermonopoly jurisdictions, the incumbent is the only entity that may install and serviceVSATs or the only entity that may own, operate and maintain satellite earth stations.

A commercial/legal presence is typically required in Africa as a pre-condition for licence issuance. This can be an obstacle to the effective roll-out of VSAT services in thecountries concerned, because it increases overhead costs to the private VSAT operatorsand inflates prices to the end-users.31

And finally, in a number of African countries, rules are often not transparent and areinaccessible to the general public. The licence-application process can be extremelycomplicated, including processing periods that require up to two years, payment of awide variety of fees - including additional taxes, annual operator fees, landing rights, etc. Added to licensing fees are customs duties, which are often so high as to prevent cost-effective access to VSAT equipment.

30 DeTeCon International “Study on Low Cost VSAT Technologies and Licensing Regimes” for the World Bank and African Virtual University. 200331 See http://www.gvf.org, “Strengthening Access to Communications”, Policy & Regulatory Guidelines for Satellite Services, 30 May 2003, GVF Regulatory Working Group, pg. 23.


3. National Experiences in Satellite Regulation and Policy

Three in-depth case studies were conducted for this Report. Algeria, Nigeria andTanzania, were selected because they serve as representative examples of AfricanAdministrations where satellite market liberalisation has been – and continues to be –applied in order to promote universal access. In addition, the countries were drawnfrom the western, eastern and northern sub-regions of the Continent. This does notonly demonstrate that the liberalisation trend is not confined to a single sub-region, but also provides an opportunity to compare and contrast the satellite regulatory approaches being implemented across the Continent.

For full details of the three national case studies, please consult the full report which can be downloaded from www.gvf.org.

4. Implications of the Three-Country Analysis

The case studies found that the three countries are on different points of the ICT-development curve and that the varying levels of progress – particularly with regard toaccess to satellite-based telecommunication services – are largely attributable to theeffectiveness of each country’s policies and regulations.

In trying to compare the license fee burden on VSAT networks in the three countries a hypothetical 100-terminal network was costed according to each country’s license feestructure, assuming an arbitrary monthly revenue or turnover of US$200 per terminalover five years. The table below shows that Tanzania’s license fees place almost 2.5times as much burden on the network than does Nigeria. In Tanzania the fees over the five-year license period amount to over US$260,000, or about 22% of the five-yearoperating cost vs. US$106,000, or about 9% in Nigeria. Algeria’s cost is considerablylower, at about US$33,000, but this is likely to increase when licensed VSAT operatorsare introduced.

Licence Fee Cost Comparison for Nigeria, Algeria and Tanzania (US$)

Nigeria License Fee Cost/5years1 ISP License 3,846.00 3,846.00 1 Satellite Network License 65,000.00 65,000.00 1 Network Switching

Equipment 7,692.00 7,692.00

1 2.5% Turnover (0.025 x 240000) x 5 30,000.00 Total 106,538.00

Algeria100 VSAT Admin Fee 43.00 4,300.00 100 Annual user fee 286.00 28,600.00

Total 32,900.00 Tanzania


In a ranking of telecommunications development in the three countries surveyed,Nigeria would be at the top of the curve, followed by Tanzania and Algeria. Nigeria’ssuccess is largely attributable to how much further it has progressed in liberalising andderegulating its market. But the underlying explanation for Nigeria’s progress is theeffectiveness of the regulator.

Algeria has begun restructuring its telecommunications sector, but growth of theAlgerian ICT market in general has been stalled by an inconsistent regulatoryframework.

By contrast, Nigeria has seen dramatic growth in ICT investment since 2001, coinciding with liberalisation and deregulation of the sector. The regulatory framework is alreadyopen, relatively consultative and enabling and commercial users consider the NigerianCommunications Commission (NCC) to have transformed from a highly bureaucraticorganisation to one run efficiently along business lines.

Tanzania also has a progressive approach to liberalisation and deregulation, but theextent to which the regulator has used licensing to generate revenues has limited localinvestment, and consequently, development of the sector. The current approach tolicensing in Tanzania incentivises operators to focus on high-margin corporate-enterprise business to pay their licence fees. Operators do not perceive Tanzania tohave an environment that will provide them with a return on their investment and thisalso explains their reluctance to invest in a local hub. Added to licensing fees arecustoms duties, which are often so high as to prevent cost-effective access to VSATequipment.

The case studies also revealed that access to satellite-based services is generally beinghindered by lack of knowledge. Broadly, the information requirements suggested byeach of the country case studies can be summarised as follows: • Algeria: Support is needed relating to technical considerations (e.g. local VSAT

hubs), economic factors (e.g. satellite bandwidth costs), and effective regulatory approaches (case studies of countries that have liberalised the VSAT sector).

• Nigeria: Dissemination of VSAT technical literature and marketing of Ku-bandVSAT services are needed to promote the technology’s ability to serve as a cost-effective alternative to C-band systems for some applications.

• Tanzania: Dialogue amongst the regulator, ministries and other government offices needs to be strengthened with the aim of developing the local ICT sector.

Finally, these three Administrations stand in stark contrast to African countries whereduopolies and monopolies are still in place. As was revealed by the IDRC Pan-Africa

1 Public data network 100,000 100,000.00 1 3% Turnover (0.03 x 240000) x 5 36,000.00 1 ISP License 1,000.00 26,000.00

100 VSAT licenses 1,000.00 100,000.00 Total 262,000.00


Satellite Survey, when an Administration is focused on protecting state investments in amonopoly or duopoly, the inherent potential of market forces to more rapidly increaseaccess and decrease cost of service is greatly inhibited… or prevented outright.

5. Challenges & Solutions: Satellite Regulatory Guidelines for Africa

The following draws upon successful satellite regulatory and policy practises currentlybeing applied in Africa and aims to provide African Administrations with practicalinformation that constructively informs their decisions relating to the formulation ofeffective satellite communications regulations and policies.

5.1 Optimising the Regulatory FrameworkAfrican Administrations are, like their counterparts in other regions, discovering that itis beneficial to establish a legislative and regulatory environment in order to promotecompetition and attract private investment. Without an appropriate legal framework for sustained telecom infrastructure development, other efforts aimed at bridging the“Digital Divide” may have little long-term impact.32 The introduction of competitionand privatisation has made most governments fully aware of the importance ofeffective, well-financed, and professionally-staffed regulatory authorities.

The International Telecommunication Union (ITU)’s Regulatory Database reflects that, as of 2004, there were 124 separate regulatory authorities worldwide. Of these, Africanregulators represent 27% of the total – second only to Europe as the largest regionalnumber. African Administrations’ trend toward establishing separate regulatory agencies was echoed by the IDRC Pan-Africa Satellite Survey, which asked Administrationswhether there was an “independent regulator” established in their country. Of 24responses to the question, seven replied that there was no independent regulator and 17 answered in the affirmative.

What was even more telling was that of the 17 who said there was an independentregulator in their country, 12 added comments relating to the fact that, while there was a regulator, its “independence” had yet to be affirmed. The overwhelming trend toward the establishment of Telecommunications RegulatoryAuthorities (TRAs) gives credence to the assertion33 that the standard institutionalstructure for the telecommunications sector around the world today includes a separateand autonomous regulator.

However, it should be noted that while TRAs may influence policy formulation,typically TRAs are only able to implement government-approved policies. Thus, thegovernment has primary responsibility for developing policies that promote expandedaccess to telecommunications through increased competition and improved regulations. Positive reasons for independence and separation of TRA activities include:• The perceived neutrality and insulation of TRAs from political or operational

pressures

32See http://www.itu.int, ITU-D Question 17-1: “Satellite regulation in developing countries”.

33 InfoDev: Telecommunications Regulation Handbook, Module 1, pg. 1-6.


• Operators and investors will generally have greater confidence that an independent TRA will regulate a market objectively and transparently

• This leads to increased investment in the sector and to related benefits that satellite services can provide to any economy.

5.2 Strategic Liberalisation in the VSAT SectorAfrican Administrations, more than many of their counterparts in developing countriesof other world regions, have begun to implement strategic liberalisation of VSATservices.

The African trend toward strategic liberalisation of VSAT was underscored by the latest regulatory developments in Algeria, Nigeria and Tanzania. In each case, theAdministrations had identified VSAT-based services as one of several keytelecommunication tools to be liberalised and for which regulations would be optimised.

Why have some African Administrations not yet liberalised the satellite sector? Thereare at least three reasons, according to Perminus Karungu, who (at time of writing)serves as Senior Officer, Licensing and Compliances, for the CommunicationsCommission of Kenya.34 Mr. Karungu outlined three key reasons as follows: desire toprotect the incumbent operator(s), fear of the unknown – Myths regarding satellitecommunications and lack of appreciation of the additional benefits of deregulation.

Malawi’s experience35 reinforces the ITU’s conclusion that “market opening works”. A VSAT license costs US$5,000/site initially and $2,500 per year subsequently. About 20licenses have been issued. According to the Malawi Communications RegulatoryAuthority (MACRA), the uniform licensing regime may in future be reviewed with “adownward adjustment [of fees] likely to take into account the emergence of low-costKu-band two-way VSAT-based Internet services aimed at small businesses andresidential users.”

The trend by African Administrations to apply satellite regulations that echo Malawi’sapproach underscores a commitment to opening markets to the provision of satelliteservices in a manner wholly consistent not only with national policy objectives – which are being achieved through strengthened access to affordable business, consumer andgovernment communications - but also with the goals of the World Summit for theInformation Society (WSIS) and the World Trade Organisation.

5.3 Liberalisation and Universal AccessAs the trend toward fully liberalised VSAT markets continues to develop, severalAfrican Administrations interviewed for the IDRC Pan-Africa Satellite Survey expressed reservations about permitting an unlimited number of market participants to provideservices. Much of the Administrations’ concern relates to reports that unlicensed service

34Mr. Karungu’s views were provided during a presentation given at Satcom Africa 2004 on 17 February in

Johannesburg, South Africa.35

Derived from a presentation given by a representative of Malawi’s regulatory agency, MACRA, during an ITU Roundtable on Least Developed Countries held in Arusha, Tanzania, on 3-4 April 2003.


providers are eroding incumbents’ revenues – and thereby undermining their ability toprovide, among other services, universal access.

However, it has been observed that, when a country limits competition through arestriction on the number of market participants, it may inadvertently encourage a form of “black market” to develop in which non-mainstream businesses attempt to provideservices and meet consumer demand in violation of the government’s licensingrequirements. The prevalence of non-mainstream service providers often makes itmore difficult for governments to ensure compliance and enforcement with theirregulations and licensing conditions.

Many countries have traditionally restricted the number of authorised terrestrial andsatellite-based telecommunications service providers that are permitted to serve acountry in order to support an implicit program of universal service for consumers.Specifically, many countries require that their dominant telecommunications providersubsidise the cost of local telephone services primarily by charging higher rates for long distance and international telephone services. Such a system of cross subsidies between different services is inefficient from an economic perspective and is difficult orimpossible to sustain following a conversion to a competitive market.36

5.4 Creating TransparencyThe IDRC Pan-Africa Satellite Survey found that the African satellite regulatory process is severely lacking in “transparency”. The difficulty of obtaining information aboutVSAT regulation in Africa is so acute – and the demand for such satellite solutions is so great – that it has given rise to a lucrative business, attracting international consultantswho sell the information to would-be satellite service providers.

Addressing the obscurity of accurate information about African satellite regulation is aprimary objective of IDRC, the Catalysing Access to ICTs in Africa programme(CATIA) of the U.K. government DFID, the Global VSAT Forum (GVF) and, to asignificant extent, the ITU Satellite Regulatory Survey. Transparent practices are criticalto the success of satellite regulation, enabling parties to benefit in a variety of ways.37

Recognition of this fact has resulted in significant moves by Administrations worldwide to post their regulations and/or policies online.

With two exceptions, all respondents to the ITU-D Question 17/1 Satellite RegulatorySurvey indicated that their laws, decrees and legal instruments were publicly availableand in many cases are posted on the web. Sixty-eight percent of the respondentsindicated that their license application forms were available, and in the majority of cases can be found on the web. But considerations of transparency are difficult to address.What is posted on websites, such as Kenya does for its tariffs, is often misleading orincomplete – which amount to the same thing.

36See http://www.gvf.org, “Strengthening Access to Communications”, Policy & Regulatory Guidelines for

Satellite Services, 30 May 2003, GVF Regulatory Working Group, pg. 7.37 ITU: Trends in Telecommunication Reform 2002.


Applicants are not the only beneficiaries of transparency; administrations also havemuch to gain. Online publishing of regulatory requirements is inexpensive, reduces theburden on Administrations (by reducing the need to respond to numerous individualinquiries), enables industry to more effectively provide services, and serves as aneffective platform from which to promote regulatory harmonisation.

In addition, regulators rely upon transparency to safeguard their legitimacy andefficiency.38 Regulators also obtain information from the regulated industry and otherinterested parties that they need in order to base their decisions on all relevant facts and diverse views. Operators and service suppliers depend on transparency to ensure thattheir concerns are heard and that they play a role in shaping important decisions.39

To facilitate this process, CATIA is co-ordinating with regional inter-governmentalgroups throughout Africa, such as Telecommunications Regulators Association ofSouthern Africa (TRASA) in southern Africa and West Africa TelecommunicationsRegulators association (WATRA) in West Africa, to establish an online One-Stop-ShopVSAT license-application framework that also includes public access to the VSATregulatory requirements applied by each African Administration. All Africangovernments have been invited to participate in the programme.

5.5 Streamlining LicensingThe ITU has called attention to the impact of the licensing process on the largerregulatory environment and the market as a whole, noting: “The licensing process canbe one of the most important regulatory processes related to reform of thetelecommunication sector. Licensing policy and its implementation determine thestructure of markets, the number and types of operators, the degree of competitionamong them, the revenues earned by governments in opening markets, and, ultimately,the efficiency of the supply of the services to the market.”40

Despite significant liberalisation gains made in recent years by African Administrations, VSATs are still among the most heavily-regulated technologies in the region, a fact that is most apparent in the realm of licensing. In combination with the sheer number ofAdministrations, Africa has become one of the most difficult regions in which to rollout a VSAT network.41

Several types of licensing requirements have been employed effectively byAdministrations, both in Africa and elsewhere in the world. These licensing rules tendto focus either on: A) The “space segment” of a satellite network or B) The terrestrial or “ground segment” of the network. In both situations, care is increasingly being taken to ensure that licensing requirements do not become barriers to free trade, but instead areused sparingly in order to accomplish legitimate regulatory requirements.


39“Feedback to Regulators from the Private Sector”, a study presented to the ITU Global Symposium for Regulators

held in Hong Kong in December 2002.40 ITU 2002 Trends in Telecommunication Reform Report, pg. 55.41

IDRC Pan-Africa Satellite Regulatory Survey.


Space SegmentLanding Rights - The Case for ‘Open Skies’ Policies.Governments are realising that tremendous demand for Internet, data, voice, video and other services is best addressed by policies that permit open and direct access to allsatellite resources assuming that they have been properly co-ordinated through theITU.42

Ground SegmentService Provider and Network Operator Licensing.Many countries require that public-network operators hold licenses so that there issome quality assurance of the service being provided to their public. A few countrieshave adopted this rule also for private VSAT services. As the nature of private satellite services is being understood better, the application of this type of license is declining.43

As it is not a public service and not usually connected to the PSTN, and can be privately owned, it is increasingly the view of Administrations that this is a redundant licensingprocess that causes time delays and confusion. These types of licenses can also bereferred to as Service Provider Licenses, Value Added Service Licenses and sometimescertain types of Class Licenses.

Blanket Licensing and General Authorisations.Traditionally, most governments have required each VSAT or mobile satellite terminalto be licensed individually; this was in addition to requiring a network operator’s license. But more than 10 years ago, a new approach to regulating VSATs - “blanket licensing” – began to be implemented.44

With this regulation, VSATs are configured based upon technical criteria - involvingpower level, frequency, etc. - that eliminate the risk of harmful interference. Thus, asingle blanket license can be issued covering a very large number of VSAT terminals.

Another finding of the ITU and IDRC surveys was that the majority of AfricanAdministrations either do not apply any licensing to receive-only systems – whetherthey are used for video or data – or they apply blanket licensing. The rationale behindthis fact is that, in theory, the verifiable purpose of licenses is public safety andpreventing harmful frequency interference; receive-only systems, because they do nottransmit, are incapable of creating interference or of posing a radiation hazard, solicensing need not be applied.


43Ibid.

44For example, the U.S. Federal Communications Commission implemented VSAT blanket licensing more

than 10 years ago. During a GVF Satellite Regulatory Workshop held during ITU WTDC in Istanbul, Turkey in 2002, an FCC speaker was asked what they would have done differently with regard to blanket licensing. Their response: “We would have done it sooner.”


5.6 Licensing FeesImplementation of general authorisations or blanket licensing results not only in fasterimplementation of service, but also lower costs of implementation.45 This derives from the fact that with individual licensing of terminals or services, licensing fees are oftenimposed on the use of individual terminals – which is likely to make the serviceunaffordable to potential end users – or on each of the service providers and requiremore administrative work on behalf of the regulator or responsible national body.

As shown in the examples for 100 site networks in the country comparison for Nigeria, Algeria and Tanzania above, the licence fees can be 10-25% of the operating cost of asmall network. These fees must be considered in the context of still other cost factors faced by VSAT network operators; there may be customs duties to be paid (up to 47% of equipmentvalue), surtaxes (up to 20%), extended surtaxes (up to 17.5%), value-added tax (up to15%), fees to be paid to the incumbent Post and Trademark Office, equipment-inspection fees, spectrum fees (up to $2000/18MHz), percentages of share capital,percentages of revenues for PSTN access (up to 0.5%), development levies (up to0.5%), administration fees, regional levies (up to 0.5%), surcharge for use of non-Intelsat bandwidth (up to 50%), and more.

The fundamental rationale for licensing fees is that they should compensateadministrative costs to the Regulator but should not be used as a source of real profitfor the government.46

5.7 Addressing Commercial or Local PresenceThe Satellite Survey revealed that numerous African Administrations maintaincommercial or local-presence requirements, the costs of which have been reported byAfrican VSAT network operators to be far-reaching.

Foreign ownership rules are capable of complicating the entire process of incorporating a company within a jurisdiction. In addition, even after a local presence is established,local partners in such arrangements may gain inequitable benefits. It has been assertedthat foreign ownership restrictions are generally contrary to the spirit if not the letter of foreign trade agreements including the General Agreement in Trade in Services(GATS).

5.8 Technology Neutrality and ConvergenceAs elsewhere, in Africa, technology neutrality is the new trend as regards the provisionof satellite services.47 Administration representatives have long called for a “leap-frogging” of technology from those building out African networks; the interest in thelatest equipment and technologies is therefore a long-established policy.

45Presentations given by the Danish, Hungarian and Norwegian Administrations during the CEPT WGRA

Meeting, Vilnius, Lithuania, February 2004.46

Ibid.47

IDRC Pan-Africa Satellite Survey.


The Botswana, Mauritius and Ugandan Administrations’ current technology-neutralapproaches to satellite regulation provide an example of governments’ recognition thatmodern telecommunications services are being provided to consumers using a numberof different technologies, such as wire line, satellite and terrestrial wireless networks.48

5.9 Managing SpectrumRegulation of satellite and other radio communications services is necessary to manage scarce spectrum resources. This is particularly true in those limited cases in whichsatellite services share a co-primary allocation with other radio communications services in the same frequency bands.49 In many frequency bands, however, satellite services donot share the same spectrum with other radio communications services andadministrations throughout the world increasingly see no reason for regulators to placeany restrictions on satellite networks that have been licensed by other countries andhave completed spectrum coordination through the ITU.50 Instead, regulators in eachcountry have begun to only impose licensing and spectrum coordination requirementson satellite networks that are based in that country. Such an approach ensures thatspectrum resources are used efficiently, by requiring each and every satellite network to secure a license from its country of origin and coordinate spectrum through the ITU.

There are a number of trends occurring around the world of considerable relevance toAfrican administrations. One discernible trend in the spectrum regulatory reformmovement in many countries is increased interest in or consideration of the use of so-called market-based allocation methodologies (i.e., auctions) as a preferred means forspectrum allocation decisions, as well as expanded opportunities for use of spectrum-related fees in connection with licensing activities. Irrespective of how this may (or may not) work in other areas, given the regional/global character of the provision of satellite services, reliance on auction-based allocation mechanisms can be fraught withconsiderable difficulties. At a minimum, it can subject a global or regional satelliteoperator to considerable uncertainty and vulnerability in its ability to provide service.

5.10 Optimising Equipment CertificationApproximately US$135 billion in telecommunication and information equipment isaffected by type-approvals processes throughout the world each year51, a significantpercentage of which is satellite-based systems. These type-approvals costs are passedon to consumers in the form of higher equipment prices, and an additional layer ofexpense is often added when Administrations require type-approval testing andcertification for satellite terminal equipment already tested and certified by otherAdministrations.

48Ibid.

49See http://www.gvf.org, “Strengthening Access to Communications”, Policy & Regulatory Guidelines for

Satellite Services, 30 May 2003, GVF Regulatory Working Group, pg. 12.50

GVF Satellite Regulatory Survey.51 USTR announcement made on the occasion of the conclusion of a Mutual Recognition Arrangement (MRA) for telecommunications equipment among members of the Asia-Pacific Economic Cooperation (APEC) forum, 15 September 1998.


In Africa, the current state of type approvals and equipment-registration requirementsfor satellite earth stations suggests a strong interest in streamlining these traditionalprocesses in order to lower consumer prices and enable more cost-effective access tosatellite services. The IDRC Pan-Africa Satellite Survey shows that most AfricanAdministrations are content to recognise the type-approval marks that apply elsewherein Africa (such as ICASA in South Africa52), as well as in Europe, the United States,Japan, Korea or China.

The ITU-D Question 17/1 Survey made a similar finding: the majority of nationssurveyed in every region recognise Mutual Recognition Agreements (MRAs), with atotal of 38 out of 50 countries indicating acceptance. Africa’s proportion is higher,where 11 countries recognise MRAs, versus only one that does not.

In order to help facilitate use of the MRA process for satellite-based systems, the private sector has also offered a solution. A technical framework that enables Administrationsto mutually recognise test results generated during the satellite operator type-approvalsprocess. This framework is encapsulated in a document entitled “GVF 101: MutualRecognition of Performance Measurement Guidelines and Procedures for Satellite System OperatorType Approvals”53. It defines a set of standardised measurements that can be used tocheck compliance of an earth station antenna model with applicable performancerequirements. The procedure also provides for independent auditing of the accuracyand completeness of the data by authorised test entities, which are elected by satellite-operator members of the GVF.

Other administrations are bypassing mutual recognition to go a step further. A goodexample is Ghana, which reported that it has begun to apply self-declaration ofconformity by manufacturers. This approach, which shifts responsibility for type-approval testing and certification from the Administration to the manufacturer, is alsoin line with global trends: Overall, 42 of 56 nations that responded to the ITU SatelliteRegulatory Survey question on type approvals allowed self-declaration.

5.11 Achieving Content NeutralityContent is not addressed in the applications to provide VSAT service in sub-SaharanAfrica. This is not to say that the Administrations are indifferent to content, however,the nature of the content is not addressed during the application process except as it isused to define the service (e.g., ‘corporate Intranets’ or ‘data transmissions’).

5.12 Enforcing ComplianceAll operators face the risk of fines, suspension or annulment of licences, andconfiscation of their equipment if they are discovered to be operating without a licence, whether they are in an African country or elsewhere in the world. Operators areparticularly at risk using C-band, which continues to be heavily used for terrestrialservices.

52Based on an ICASA presentation given on 8-9 March in Lesotho at a TRASA/CATIA Low-Cost VSAT

Workshop.53

www.gvf.org


Countries increasingly are developing laws and regulations for the telecommunicationssector that are objective (non-discriminatory), easily understood (transparent) and highly predictable. Such laws and regulations also prohibit government actions that arearbitrary or discriminatory.

Mainstream businesses tend to avoid investing in countries that lack objective,transparent and predictable regulatory structures. Furthermore, a government-imposedrestriction on the number of participants that can exist in a particular market segment(i.e., a mandatory monopoly, duopoly, or other numerical restriction) also serves toprevent many mainstream businesses from providing services in the country.

6. ConclusionsOf the world’s 49 Least Developed Countries (LDCs), 31 are in Africa.54 The ITU has calculated that, until a country has passed the threshold of 1 main line per 100inhabitants, it takes, on average, 50 years to reach a teledensity of 50 main lines, a level reflecting high telecommunication development. But until a country reaches 1 main line per 100 inhabitants, it is “virtually impossible” to predict how long it will take to reachhigher levels. Thirty-four of the 49 LDCs have a teledensity of less than one.55

Providing increased access to ICTs in Africa is a complex problem. Access to theInternet and other telecom services has been held back not only by restrictive regulatory frameworks, but also antiquated infrastructure, high fixed costs, low economic andinvestment activity, diverse geography, language and culture, and much more.Accordingly, VSAT is not proposed to be the tool for Africa’s challenges; it is one ofseveral tools, each of which plays to its respective strengths – fibre for point-to-pointservices, mobile for voice and narrow-band data, satellite for point-to-multipoint narrow and broadband solutions.

The Pan-Africa Satellite Survey and case studies conducted for Open and ClosedSkies: Satellite Access in Africa have also demonstrated is that VSATs have emerged as an important tool that is capable of leveraging accelerated access to ICTs, providedthat African Administrations are prepared to actively facilitate its use. In the satellitearea, frequency use, network operations, service provision and the use of radio terminals can be considered as the main elements which have been the target of a number ofregulatory measures (e.g. licensing conditions and procedures) normally meant to helpthe development of satellite telecommunications and facilitate market access to satellite providers, but which may also act as market barriers.56

At the same time, the industry’s competitive structure has also changed at the level ofnational and international markets: Many Post, Telegraph and Telephone organisations (PTTs) have been privatised as well as intergovernmental satellite operators. Thisconcurrent evolution of satellite operators, service providers, and applications - as wellas their corresponding regulatory treatment - highlights the importance of ensuring

54From the ITU pamphlet on the Special Programme for the Least Developed Countries, 2000-2003.

55Ibid.

56 Comprehensive Satellite Initiative Report, 11 July 2001, CEPT ECTRA ERC JPT SAT doc. (01) 265.


transparent and non-discriminatory market access conditions as the best means ofpromoting an individual country’s development. Like never before, and as stated in the definition of ITU-D Question 17/1, “Administrations must ensure that their regulatorytreatment provides a level playing field for both existing and emerging satelliteoperators, service providers and satellite-based applications.”

Such treatment may be facilitated at various levels: Globally through organisations likethe WTO and ITU; regionally through groups like the Asia Pacific Telecommunity(APT), African Telecommunications Union (ATU), European Conference of Postal and Telecommunications Administrations (CEPT), European Commission (EC), Inter-American Telecommunication Commission (CITEL); and sub-regionally throughgroups like EARPTO, TRASA, WATRA and others.

However, while these organisations are one of the ideal types of forum through whichto pursue harmonisation, implementation of regulatory reform is largely being driven byinitiatives taken at the national level.

Of those African Administrations surveyed that had implemented satellite reforms,most were in relatively early stages and none had fully implemented all – or even a few –of the above-noted approaches. While this poses a major obstacle to addressing Africa’s “digital divide”, it also represents what may be one of the largest regional ICTopportunities in the world.

As African Administrations continue to develop satellite regulatory reforms – and theIDRC Pan-Africa Satellite Survey confirms this is underway right now – they are doing so in co-ordination with neighbouring Administrations, with whom they share not only a common border but also common policy objectives, including enforcement of radiofrequency allocations and ensuring that licensees are protected against harmfulinterference. Increasingly, those shared policy goals have also come to include theprovision of ICT solutions applied domestically and internationally or, more to thepoint, the promotion of public interest, social welfare and trade - to, from and withinthe region.

The regional regulatory authorities - East African Regulatory PostalTelecommunications Organisation (EARPTO), TRASA and WATRA have eachrecognised the potential to develop an international and harmonised operatingenvironment that embraces satellites’ inherent ability to provide end users with aninexpensive, single communications platform capable of serving almost any location,regardless of population density or proximity to urban areas.

Active participation by Africa’s inter-governmental groups is essential, because one ofthe greatest strengths of satellite networks also, nonetheless, creates the greatestregulatory challenge. Satellite operators depend on harmonised regulatory approaches,because of their wide coverage. The services that the operator provides, often on a pan-regional scale, depend on authorisations and spectrum allocations in each country in its coverage zone. If those authorisations are difficult to obtain or allocations are notuniformly recognised across a region, then the advantages of satellite services can beblocked.


Liberalisation, transparency and a commitment to satellite regulatory harmonisation arewithin Africa’s reach. If Africa seizes this opportunity, it will transform the lives of itspeople forever.

155

Economic Development in Africa Powered byMobile Telephony

Olof Hesselmark and Anders Engvall

ABSTRACTAfrica's ICT landscape has changed completely in five years. Many countries, amongthem Nigeria with Africa's largest population, now have ten times as many telephonesas a decade ago. Large direct foreign investments have flowed into the mobile telecomsector and created a modern, efficient and geographically widespread infrastructure,already reaching into areas where previously no services were available. The mobilebusiness is private and profitable, and Africa's people are enjoying telephone servicesthat they never had access to before.

Historically, economic growth has gone together with telecom growth. Now however,the wireless technology has cut the costs of building infrastructure and of providingaccess to a fraction of what it used to be. Africa's economies are suddenly endowed with large-scale communication facilities, which will give them a growth engine to powereconomic development. Whether this will actually transform into greater economicgrowth will naturally depend on many things, but at least one important impediment to trade and development is rapidly being removed.

This article takes a closer look at Africa's telecom development and its potential forpowering future economic growth.

1. The Mobile Success Story in Africa57

There can be little doubt that good telecommunications is an enabling factor foreconomic growth58. The following chart shows the development of the number oftelephones and GDP for the OECD countries between 1960 and 1999. This period was characterised by rapid growth – real GDP grew by a factor three, but the number oftelephones increased five times in the same period.

57 Africa south of the Sahara, (SSA) excluding South Africa, is a region with great variation. There are small

and large nations, a wide variety of climate, natural resources and culture. However, almost all countries have low average incomes and generally poor infrastructure, especially regarding the fixed telecom networks. When we refer to Africa in this article, we will be dealing with SSA. 58

The research presented in “Africa: The Impact of Mobile Phones”, the Vodafone Policy Paper Series, March 2005 has made an excellent attempt to address this aspect of mobile telephony. The report is availableat www.vodafone.co.uk


OECD countries telephones and GDP

0

50,000,000

100,000,000

150,000,000

200,000,000

250,000,000

300,000,000

350,000,000

400,000,000

450,000,000

500,000,000

1960

1963

1966

1969

1972

1975

1978

1981

1984

1987

1990

1993

1996

No

of

ph

on

es&

GD

P/7

5000

TelephonesConstant GDP

Fig 1. Source: ITU Telecommunications database.

The comparison does not tell us that telecommunication causes growth, only thattelecommunications and economic growth go together in developed economies59. In the poorest countries of the world, economic growth has been low and at the same timethere were few telephones before the arrival of mobile networks. Low per capitaincomes in Africa and other poor countries has thus been associated with the absenceof telecommunications. The arrival of mobile phones has in a very short time changedthis relation, and most poor countries have now ten times as many telephone users ascompared to only a decade ago. It is an intriguing question to ask if this telecomexplosion will bring about a more rapid economic growth in Africa’s poor countries.The following diagram provides a clue to the answer.

Africa's least developed countries: Shares of world GDP and telephones

0

0.1

0.2

0.3

0.4

0.5

0.6

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

Per

cen

t

Share of GDP Share of telephones

Fig 2. Africa’s GDP and telephones. Source ITU world telecom database.

59 Research by Roeller and Waverman (2001), referenced in the Vodafone study, suggest that the spread of

fixed line telecom services in the OECD was directly responsible for one third of output growth between 1970 and 1990. The Vodafone study suggests very high growth effects on developing economies as a result of increased use of mobile phones.

Economic Development in Africa Powered by Mobile Telephony | 157

The diagram above shows the development over 15 years of GDP and telephonepenetration. Using data from the World Bank and ITU we have calculated the share ofthe world’s GDP and telephones for all low-income countries in Africa. These countries have about 0.5% share of the world’s GDP, with a slowly declining trend over the lastdecade. Between 1986 and the end of the 1990s their share of the world’s telephoneswas constant at about 0.3%. After 1998 this share has risen sharply as a result of theintroduction of mobile phones. By 2002 – the last year in the series – there clearly wasno indication of any growth in the share of incomes, so it seems too early to sayanything about the macro effects of mobile phones on GDP. But by 2005, one canestimate that the share has doubled – to 0.6% of the world’s phones – and when reliable income data become available it will be interesting to see what has happened.

1.1 Mobile growthMobile telephones are now a normal thing in almost all countries in Africa. Initiallycovering cities and large towns, network cover is rapidly reaching the rural areas wherethe majority of Africa’s people live. In Sub-Saharan Africa, a few countries (Uganda,Kenya, Senegal) have already reached 10% penetration and full population coverage inspite of low per capita incomes. The relatively wealthier countries of South Africa,Gabon, and Botswana have reached 20% penetration. In the rest of the Sub-SaharanAfrican countries, average penetration is between three and four percent, adding up to a total of about 30 million users by 2005.

Networks in Africa are still in their initial growth, and large investments in new capacity continue to be made by over one hundred individual operators. An average of 10%penetration for all countries in SSA would mean 65 million mobile users and there aregood prospects that this level will be reached in the next few years. Foreign owners and financiers are behind most operators, which are run on purely commercial principles.Cash from revenues is in many cases financing investments in new capacity, an unusual situation in Africa considering that the operators are just a few years old. The following two maps show the mobile cover in 2004, and Africa’s population density.

Fig 3. Source: GSM map from GSM Association.


The maps show that mobile networks are only beginning to cover the vast rural areaswhere the majority of the people live. Full or almost full population cover has beenreached in South Africa, Namibia and Botswana and in the Mediterranean states inNorth Africa. In other countries, for example Angola and Ethiopia, the GSM networks are only in their initial stages of expansion.

There are a number of reasons why the rate of adoption is as varied. One is political, as in the case of Angola. During the civil war in that country, the risk for investors were at a maximum level, and very little foreign investment would find its way there. Thecreation of a national GSM network requires a great deal of money and manytechnology specialists, which in turn demands a certain level of stability and security.Angola is therefore a late starter as far as GSM infrastructure goes, but there is nowlittle doubt that a rapid deployment is underway. As a contrast, South Africa has had cell phones for over ten years now, enjoys political stability, maintains a welcominginvestment policy and its population has one of the highest purchasing powers inAfrica.

Another important reason is a country’s investment climate and its telecom policy.Ethiopian authorities are reluctant to allow foreign investment, and all telecom business in the country is reserved for the state owned telecom monopoly. The telecom company has been unable (or unwilling) to raise the large sums of money needed for a nationalrollout of a GSM network.

A third reason is the general business climate in the country. Nigeria is often describedas one of the worst countries in the world to do business in, with extremely highbusiness risks. In the case of GSM roll-out, a long battle over licences delayed theprocess for several years, and it was only in 2002 that the networks were initiated. Assoon as the licence issues were resolved, however, large amounts of venture capital were made available. The growth has been phenomenal, and Nigeria is likely to reach tenmillion users during 2006.

1.2 Investments and Telecom InfrastructureThe demand for telecom services is an irresistible force throughout Africa. Behindalmost all GSM investments there are entrepreneurs and foreign investors who earlysaw the opportunities and started negotiations for licences with governments andexisting telecom operators. Expectations were low ten years ago – in many cases theactual number of users in 2005 are five times greater than shown in the forecasts made in 1995. As a result we can see a financial success story like never before – a newbusiness type multiplied over 55 countries in Africa with very few failures in ten years.GSM is one of the most profitable business sectors on the continent.

Not only are the investments in mobile telephony very large, but they are also affecting large geographical areas. This is demonstrated by the number of radio base stationsinstalled, which can be estimated from data reported by some of the large international operators. For example, MTN report60 6,234 installed base stations for 9.5 million users. On average, each cell covers about 1500 users. With 30 million users, about 20,000 base

60 In their semi-annual report for Sept. 2004. www.mtn.co.za


stations have been installed in all of Sub-Saharan Africa. Each base station is part of acomplex telecom infrastructure consisting of large scale switching units, control unitsand transmission links. MTN added about 1200 base stations in 2003-2004, andinvested a total of 7.6 billion Rand61.

Since the start of GSM networks in the 1990s, many parameters have changed in thebusiness case for a new network. The technology has matured, and a wide range oftechnical options for switching, radio control systems, transmission and handsets arenow available. The cost of building a network has been dramatically reduced62, reflecting better technical modularity, lower costs for computers and electronics, and improvedcompetition among network suppliers. The prepaid account has been invented,changing the entire business model of operators. The uncertainty about usage has been eliminated – almost all operators in the world have surpassed their original userestimates by wide margins, so the ex ante risk conditions are now much lower thanbefore. All this means that the anticipated costs of running a GSM network is nowmuch lower than ten years ago, and many of the business risks associated with aninvestment are also much lower. New investors can be much more certain of gettingtheir money back, and this should lower their profitability requirements.

In addition to the network investments there are the handsets, which are directly paidfor by the users themselves. The average retail price of a handset is about US$ 70-100,so the total African investment value for that part alone is at least two billion dollars. It is worth mentioning that US$100 may be above the annual income of the poorest half ofthe population in most African countries. Prices for handsets are coming down thoughand recent efforts by the GSM Association show that there is a market for handsetsspecifically designed for low-income countries.

1.3 Competition and business performanceThe price competition between existing GSM operators in Africa is generally weak.They compete mainly on coverage, while rapidly building out their networks and there is little price pressure for getting more customers. Tariffs are high in comparison withfixed networks, but not high enough to prevent growth. As a result, profitability issoaring as the networks expand. An additional investment in network capacity is filledup quickly, also in rural areas. In an interview in July 2005 by the BBC, Safaricom’sChief Executive in Kenya expressed surprise at how quickly new base stations in ruralareas become fully utilized after commissioning.

The existing mobile operators enjoy protection from competition through the fact that only few national licences have been issued. When these licences were issued a few years back, nobody could foresee the spectacular growth in demand. What was then regarded as credible business cases underestimated the actual growth over five years, and it wasgenerally thought that the mobile market could only sustain services in high-income

61This amount is MTN’s investments in the 2003-2004 period. It includes all network related equipment,

switches, control units, base stations, transmission, offices, transport, power units etc. Based on these data, the total investment per base station can be estimated to about one million US$. Multiplied with the number of BTS in Africa, the total investment volume is simply amazing. 62

The technical director of a large European operator told us that the 2004 prices of radio network components have dropped to 15% of the levels 10 years ago.


urban areas. In reality, many operators enjoy spectacular results and a very convincingcase can be made for dramatically reduced tariffs. GSM operators have an interestingcost structure, with more than 50% of their total costs being fixed capital costs. Newcustomers are added at virtually no cost to the operator. Not only do they generaterevenues immediately but also pay in advance for the services. This will continue untilcoverage has reached all prospective population centres and until that point in time,price competition is very unlikely. This explains why gross profit from operations inmany cases is higher than 50% of revenues and why many operators are already able tofinance a large part of their expansion from internal cash generation. This is aremarkable achievement in view of the rapid growth and the market not yet havingreached a mature state. Consequently, the market capitalisation value of the companiesis very high in the order of US$ 400 to 750 per customer63.

Some countries in Africa have just one GSM operator, while most have two or more.Economic experience tells us that competition exercises a downward pressure on prices and profits. This downward pressure has not yet occurred in the African GSM marketwhere prices remain high and profits still increase. This situation does, however, havesome very positive effects. In countries where there are two or three internationaloperators (such as MTN, Vodacom, Celtel or Orange) there seems to be an almostfrantic investment activity, where each player tries to out-build the others with regard to coverage. This race takes place in a complete telecom vacuum – new coverage isprovided in areas where there have never been any telephones before, and largenumbers of urban as well as rural people can get connected. This process is in itself anextremely beneficial one. It creates value, as witnessed by the rapid adoption of theservices as soon they become available. People are willing to pay for connectivity,wherever they are.

1.4 Users and pricesIt is remarkable that prices can remain as high as they are, without affecting growth. The explanation lies partly in the prepaid accounts that allow a user to stay connected at just a few dollars per month. The other explanation lies in the customers’ creativity ingetting the maximum utility from a very low usage. Calls are of very short duration –state your business, get an answer and hang up. SMS, “beeping” and other tricks help to reduce the unit cost per message. Social chatting for hours at 25 US¢ per minute is notpossible, since the prepaid account balance will run out long before the hour is up. The high prices in relation to incomes create a very high efficiency in using the services.

From a social perspective the situation with high call charges is not as bad as it looks.To begin with, mobile services will cover a majority of the population with potentialtelecom services in country after country. The result is that the Universal Access goalhas been fulfilled in areas with GSM coverage, far earlier than anybody would havethought just five years ago. The high prices will induce consumers to adopt innovative

63 The Swedish owners of MTC Namibia sold their 49% share for 500 million Namibian dollars (US$ 71

million) in May 2004. MTC had at the time about 300,000 users, and the value equals US$ 450 per user. MTN’s market capitalisation value in March 2004 was 53 billion Rand. With 10 million users, the value per user is about US$ 750. Celtel International was sold in 2005 at a price of $660 per subscriber.


and efficient modes of use to get access to a much desired telephone service64.Secondly, the high profit generation has enabled a rapid service provision to rural areas. Generated cash has been reinvested in the network, instead of distributed to theowners. Since the customers are willing to pay the high call charges, the presentsituation is probably economically more favourable than if price caps or other tariffrestricting measures had been imposed by the regulatory authorities. Such measureswould have protected a limited number of existing customers at the expense of thosewho have no access to the service due to lack of coverage.

1.5 Taxation and public financeIn Uganda, there is presently a dispute going on between the three mobile operators and the Government as to the level of taxes that should be imposed on the mobile sector.With a combination of VAT and excise taxes, the mobile revenues are subject to ataxation of 30%. The mobile operators have started to state their rates in pre- and post-tax figures to better inform the public of their tax burden.

In Kenya, recent press reports say that Safaricom is now the most profitable companyas well as the largest taxpayer in the country. Without too detailed examinations, it could safely be assumed that the mobile industry is by far the highest contributor of any sector to the treasuries of Sub-Saharan Africa. In the following paragraph, Tanzania will beused as a typical example of the level of tax revenues being generated by the mobileindustry.

In Tanzania there is 20% VAT charge on the sales of handsets and prepaid mobilecards. There is also a 20% duty on the CIF (Cost, Insurance and Freight) value of allimported goods, including handsets and network equipment. In addition, the operatorspay profit taxes, and employees pay income taxes. With mobile telephone turnover for1.5 million users and ARPU (Average Revenue per Unit) of US$ 20 per month, the total tax value can be estimated. Add import duty and VAT on handsets and pre-paid mobile cards, profit tax and income tax to estimate the total tax on mobile telephone services.If all the taxes were actually collected, the Tanzanian tax authority should be collectingabout US$ 100 million this year from the mobile business, not an insignificant amount.

The potential tax should be of similar magnitude for most other countries in Africa.Tanzania has about 5% of the phones in Sub-Saharan Africa, so we are looking at atotal tax potential of about $2 billion per year. This would in many cases be "windfall"revenue for the tax authorities, since because of the rapid growth of the telecom sector the taxes will not have been anticipated in the budget process. Since all mobile revenues pass through the mobile operators it is also likely that money that previously was spent in the “informal” part of the economy now has been transferred to the “formal“ sector. For this reason, the growth of the telecom sector will have an important indirect effecton the public finances of many countries.

It is beyond the scope of this document to discuss the “proper” level of taxation formobile services. We would like to point out, however, that the tax revenues generated

64 In Ghana, for example, a practice of street vendors has developed providing a service similar to phone

booths.


from regular operations are substantial and far in excess of any license fees or monopoly profits collected by the authorities. This has considerable economic impact and it isstrongest in the countries with well-developed mobile telecommunications. Countrieslike Ethiopia with a state owned, monopoly operator are largely passing up such taxbenefits.

1.6 EconomicsWhen telephones are introduced in low-income economies, two main types of effectsoccur65. The first is that the cost and time of collecting information to supporteconomic decisions are reduced. The result is quantitative and qualitative – an individual can make more transactions in a given time, and each decision is based on betterinformation. In both cases, value is added. The other effect is the network effect. As the number of users in a communication network increases, the number of possibleconnections increases with the square of the number of users. This effect accelerates the creation of value added.

Since a poor person by definition has very limited economic resources for gatheringinformation, the value of information from a successful phone call is relatively greaterthan for a rich person. A “successful phone call” can be pragmatically defined as asituation where a person receives information that enables her to avoid the time andexpense of a long bus trip or one that gives her knowledge about the availability andlocation of cheap goods or services. It is in this indirect way that cheap and accessibletelecommunication services can benefit the poor66. This is true for all sorts of telecomservices where connections between people can be made fast and at dramaticallyreduced costs, be it telephones, SMS or email. The Vodafone report suggests that countries with an early growth of mobile phones will have an advantage of as much as one percent of GNP growth compared with lateadopters. Transferred to African conditions, this could mean that Ethiopia’s slowacceptance partly explains the difference in growth compared to, for example, Uganda,Mozambique or Tanzania.

2. What happens next?

2.1 Network growthIn Sub-Saharan Africa, telecom networks will continue to expand for many years andnew subscribers will be connected following the expansion of coverage. At current tariff levels, the saturation point will soon be reached and the interesting issue is what actionthe mobile operators will then take.

One likely direction could be that tariffs would be considerably lowered by competition, thereby bringing in new, low intensity users. A strong argument for this scenario is that once the networks get close to universal coverage, additional traffic is carried at lowmarginal costs. Profitability will therefore be high even if traffic charges wereconsiderably reduced. Intelligent price differentiation schemes could be introduced to

65 See “Africa: The Impact of Mobile Phones”, the Vodafone Policy Paper Series, March 2005.

66 The field studies carried out for the Vodafone report in Tanzania, South Africa and Egypt confirm that this

type of effects indeed are the norm when mobile phones become widespread in rural communities.


maintain high profitability of the intensive business users, and new tariff schemes will be designed to attract the rural population.

Another, opposing scenario would be that once the main road arteries and mainpopulation centres are covered, the operators would more or less cease to expandcoverage. Most of them would already have exceeded the license requirements forpopulation coverage and thus have no legal obligation for continuing to build networksfor low-income, rural areas. This may be an attractive option for some owners who wish to take their capital gains and invest them elsewhere. The Millicom group of companies seems to follow this strategy.

The authors have shown67 that low-income rural areas would best be served by applying special solutions like limited mobility (home zone billing) provided by franchised smallcommunity operators. This would meet the Universal Access requirements atmaintained profitability, albeit at a lower level than presently experienced by thedominant mobile players in Africa. For this to come about, regulatory interventionwould be required.

2.2 IP Based ServicesThe mobile operators are building an entirely new telecom infrastructure all over Africa. Due to the inefficiency of the incumbent operators, they have been granted the right by the regulators to build their own microwave routes, which are beginning to constitutecomplete national backbone transmission networks in many countries. Thisinfrastructure of transmission and wireless distribution is capable of carrying not onlyvoice traffic, but also data traffic. Since voice is the primary communications driver inAfrica, voice revenues will pay for the entire network infrastructure. Data services canthus be offered based on marginal costs and at relatively low rates. Once voice servicesbecome saturated, we believe that many mobile operators will turn their attention to IP based data services. Most GSM networks can today easily be upgraded to carry dataservices and the speed of GPRS/EDGE networks would in many cases be sufficient for meeting the requirements of many low- intensity users of e-mail and Internet services.Since the mobile operators control the transmission networks, they can also offeralternative wireless data access at higher bit rates when required.

2.3 Non-Communication Based Services Mobile operators are significant business organizations not only in the telecom sectorbut also in the entire economy. They have countrywide operations and an efficientdistribution mechanism for selling their pre-paid cards. The high security on mobilephones makes them an ideal vehicle for payment and money transfer services. A logical expansion of the pre-paid systems is thus to use the same system for money transfers.In addition, SMS messages are increasingly being used for confirming transfer ofprepaid balances also between individuals. Safaricom in Kenya has announced that itwill be offering money transfer services via their pre-paid system. It is likely that manymore operators will follow. Banking services are poorly developed in rural Africa andmobile operators might well be the first organizations to bring this new important

67 “Profitable Universal Access Providers” by Engvall and Hesselmark, October 2004, www.scanbi-

invest.com


service to rural areas. This will stimulate economic activities and open up new businessopportunities.

2.4 Liberalization and LicencesThe African mobile story shows convincingly that competition and private enterprisehave been able to deliver world-class telephone services. Liberalization of telecommarkets along WTO requirements is underway in most countries, which will provideadditional customer choice particularly for international services.

With the success of the mobile communications industry there follows a risk that a new breed of “dominant players” has been born. The largest mobile operators now controlconsiderable network resources in several countries. Because of their profitability, their financial positions are much stronger than that of the incumbent telecom operators. All telecom regulatory practice emphasise that special attention should be given to thedominant players. The incumbent fixed line operators that have so far received suchattention are no longer dominant. So far, however, the mobile operators have beenvirtually unregulated, except for a few coverage requirements in their licenseagreements.

The communications regulators need especially to pay attention to the vast transmissioninfrastructure now controlled by the mobile operators. These network need to beopened up to other sector players, be it ISPs, community operators or even theincumbent fixed line operator. If no mobile price competition comes about when themarket becomes saturated, the regulators need to follow the European example andfoster the introduction of Mobile Virtual Network Operators (MVNO). Considering the already prevalent existence of duplication of infrastructure facilities, we considerMVNO licensing to be preferable to issuing new countrywide mobile licenses.

A MVNO typically does not own its own network infrastructure, but has businessrelations with existing mobile operators, buying traffic minutes on their network at bulk prices. MVNOs are emerging in the industrialized countries for providing services tospecific market segments. A possible African application is discussed below.

2.5 Regional operatorsThe following map shows parts of Northern Nigeria and Southern Niger. It is an areawith very high population density. The established Lingua Franca of the region isHausa, spoken by an estimated 40-50 million people from Northern Ghana to Chad.


Fig 4. The border area between Nigeria and Niger is one example of a densely populatedarea where mobile services are expanding rapidly on both sides of the border.

A great deal of travel and trade takes place across the borders in this area, but untilrecently any cross-border business had to be carried out without the help of telephoneservices. The national boundaries made services very expensive and in practiceunavailable until the arrival of mobile services in the last few years. The towns ofKatsina, Maradi and Zinder in the map have had mobile cover only since 2003-2004.The Hausa-speaking area stretches for 800 km from Northern Ghana to Chad, andstraddles the national boundaries of six countries. Assuming a potential mobilepenetration of 10%, there will be two to three million future subscribers living in thisarea and speaking the same language. As common language is very important for cross-border trade, this region would greatly benefit from a single tariff policy withoutroaming charges.

As long as only national licenses are issued, it becomes difficult to build traffic-optimalnetworks in areas like this. With six countries and 15 operators involved to cover thearea, networks become fragmented, and several different interconnection regimes willbe involved. International boundaries complicate things in a mobile environment.Ideally, this area should be served by a number of competing operators with regionallicenses, without roaming charges.

With a flexible approach from the regulators in the six countries, the above situationcould be solved with mobile technology. To cater for the language communality, theregulators in the six countries could license several MVNOs for serving the Hausa area. The licensing should include tariff conditions in addition to coverage objectives toensure that it is the mass market that would be addressed. The interregional mobile


market is large and grossly untapped, and there are good reasons to stop looking attelephone services as strictly national. For the individual states, low cost telephoneservices in the border areas create development, trade and business opportunities.

Fig 5. A regional example

3. Conclusion

This article has put forward a number of arguments that empirically would indicate that mobile telephony is powering economic development in Africa. Mobile telephony is atits infancy in this region with many more spectacular growth years yet to come. There is therefore every reason to be optimistic as regards the possibilities of mobile telephonybeing a major driving force for transforming African economies and for reducing thedigital divide.

167

List of Abbreviations

AAU African Association of UniversitiesAERC African Economic Research ConsortiumAfDB African Development BankAFNOG African Network Operators GroupAfrISPA African ISP AssociationAGORA Access to Global On-line Research in AgricultureAIF Agence Intergouvernementale de la FrancophonieAIMS African Institute for Mathematical SciencesATICS African Tertiary Institution Connectivity SurveyATM Asynchrounous Transfer ModeAU African UnionAUF Agence Universitaire de la FrancophonieAVU African Virtual UniversityCATIA Catalysing Access to Information in Africa – DFID programCERN European Organisation for Nuclear ResearchCIDA Canadian International Development AgencyCODESRIA Council for the Development of Social Science Research in

AfricaCOMESA Common Market for Eastern and Southern AfricaCOMTEL planned communications network in Eastern and Southern

AfricaEUMEDConnect

EU program to network the Mediterranean countries

DANIDA Danish International Development AgencyDANTE Delivery of Advanced Network Technology to Europe –

management agency for GEANTDFID UK Department for International DevelopmentDot-EDU USAID program to promote equitable access to digital

technologiesEASSy East African Submarine Cable SystemECOWAS Economic Community of West African StateseIFL Electronic Information for LibrariesESCOM Electricity Supply CommissionFAO Food and Agricultural Organisation of the United NationsGbps Giga Bits per SecondGDLN Global Development Learning Network (World Bank)


GEANT Collaborative project and network among 26 National Research and Education Networks in Europe (EU program)

GeSCI Global e-Schools and Communities InitiativeGTZ German Organisation for Development CooperationGVF Global VSAT ForumHF High FrequencyHINARI Health Internetwork Access to Research InitiativeICT Information and Communication TechnologiesIDRC International Development Research CentreIEEAF Internet Educational Equal Access FoundationIGAD Inter-Governmental Authority on Development (Horn of

Africa)IICD International Institute for Communication and

DevelopmentIMF International Monetary FundINASP International Network for the Availability of Scientific

PublicationsINTIF Francophone Institute for New Technologies and Training

– AIF InWent Capacity Building International – German Govt cooperation IP Internet ProtocolISOC Internet SocietyISP Internet Service ProviderITU International Telecommunication UnionIXP Internet Exchange PointJSTOR not for profit journal archiveKbps Kilo Bits per SecondKENET Kenya Education NetworkKTH Swedish Royal Institute of TechnologyMbps Mega Bits per SecondMDG Millennium Development GoalsMICTI The Mozambique Information and Communication

Technology InstituteMIMCom Multilateral Initiative on Malaria Communications NetworkMSI-SIG Millennium Science Initiative – Science Initiative GroupMTN Maritime Telecommunication NetworkMVNO Mobile Virtual Network OperatorsNAMPOWER Namibia PowerNectarNet proposal for high speed network linking Africa with US via

Europe – Georgia TechNEPAD New Partnership for Africa’s DevelopmentNetTel Network for Capacity Building and Knowledge Exchange in

ICTNGN Next Generation Network (ITU intiative)NGO Non Governmental Organization

List of Abbreviations | 169

NMI Nelson Mandela InstitutionNORAD Norwegian Agency for Development CooperationNREN National Education and Research NetworkNSF National Science FoundationNSRC Network Start-Up Resource\CenterNUFFIC Netherlands Organisation for International Cooperation in

Higher EducationPAREN Promotion of African Research and Networking (IDRC

Program)PATU Pan African Telecommunications UnionPDH Plesiochronous Digital HierarchyPHE Partnership for Higher EducationPoP Point of Presence – service provider’s location for

connecting to PPP Purchasing Power ParityPSTN Public Switched Telephone NetworkRASCOM Regional African Satellite CommissionRIA Research ICT AfricaSADC Southern Africa Development CommunitySADC Southern Africa Development CommunitySAFE South Africa Far EastSAREC SIDA Department for Research CoordinationSARUA Southern Africa Regional University AssociationSAT South Africa Trans AtlanticSDH Synchronous Digital HierarchySEA South East AsiaSIDA Swedish International Development AgencySIST Systeme d’Information Scientifique et Technique – French SITA Studies in Information Technology ApplicationsSME Small and Medium (sized) Enterprise(s)SPIDER Swedish Program for Information and Communication

Technology in Developing RegionsSRII SADC Regional Information InitiativeTbps Tera Bits per SecondTENET Tertiary Education Network – South AfricaTISPA Tanzania Internet Service Provider AssociationTRASA Telecommunications Regulators Association of Southern

AfricaUEM Eduardo Mondlane University, MozambiqueUHF/VHF Ultra High Frequency/ Very High FrequencyUNDP United Nations Development ProgrammeUNESCO United Nations Organisation for Education, Science and

CultureUNU United Nations University


USAID United States Agency for International DevelopmentVoIP Voice over Internet ProtocolVSAT Very Small Aperture Terminal for satellite communicationsWASC West Africa Submarine CableWB World BankWATRA West Africa Telecommunications Regulators AssociationWCE World Computer ExchangeWDM Wavelength Division MultiplexingWE West EuropeWHO World Health OrganisationWiderNet University of Iowa, International programsWi-Fi Wireless FidelityWi-MAX Worldwide Interoperability for Microwave AccessWLAN Wireless LANWSIS World Summit on the Information SocietyWTO World Trade Organization

171

About the Authors

Anders Comstedt has been involved in several European, African and US projectsrelated to new competition creating telecom infrastructure. Being an advisor in industry transformation and related policy issues, he is a senior consultant in several businessimplementations of alternative telecommunication networks and studies. Mr. Comstedt is also a senior advisor at the Swedish Royal Institute of Technology in Stockholm. He is an advisor and consultant to Spintrack AB, Sweden. In 2004-2005 he was a boardmember of Labs2 AB, a telecom software developer. In his previous positions, he isrecently most recognised for being the CEO of Stokab, taking that company from anearly 1994 idea to a model for how a sustainable position could be created for anindependent, infrastructure-only telecommunications company. Additionally, he hastaken part in the development of related telecom policy issues in Sweden for more than a decade. Prior to his work with Stokab, he has had several executive positions, mainly within the telecommunications industry in companies like Ericsson and Telia. Mr.Comstedt holds a master degree in electrical engineering.

Martin Curley is Senior Principal Engineer and Global Director of IT Innovation atIntel Corporation managing a network of Intel IT Innovation centres developingadvanced IT solutions. Previously Martin held a number of IT Management positionsfor Intel including Director of IT Strategy and Technology based in Sacramento,California and Fab14 Automation Manager based in Dublin, Ireland. Martin has alsoheld IT management positions at General Electric in Ireland and Philips in theNetherlands. Martin has a degree in Electronic Engineering and a Masters in BusinessStudies from University College Dublin, Ireland. Martin is author of “ManagingInformation Technology for Business Value” published by Intel Press Jan 04 and nowin its third re-print.

Samuel Danofsky has five years of experience working with international development issues. He joined the United Nations ICT Task Force secretariat in 2003 where he is afocal point for Africa and for the United Nations ICT Task Force Working Group onthe Enabling Environment. Prior to joining the Task Force secretariat Mr. Danofskyspent two years at the Swedish Ministry for Foreign Affairs where he worked withSwedish policy on poverty reduction and global development with particularresponsibility for issues related to trade and development, private sector developmentand ICT and development. Mr. Danofsky holds a BA in Economics and PoliticalScience form Uppsala University in Sweden.

Jabulani Dhliwayo is NEPAD Council’s Senior Vice President and Science,Technology and ICT commission chairman. His main interest in Africa is thedevelopment of modern fibre optic networks that will interconnect all Africancountries, facilitate broadband access and help improve the African economy thoughthe use of modern ICTs. Since 1999, Dr. Dhliwayo has been working for CorningIncorporated, the discoverer and world’s leading supplier of optical fibre and cable for


telecommunications. There he has held positions as a Senior Research Scientist, project engineer and Senior Applications Engineer. Dr. Dhliwayo is a 2004 US National BlackEngineer of the year special recognition award winner. Dr. Dhliwayo holds a PhD inPhysics from the University of Kent, UK, an M.sc in Applied Physics from LaurentianUniversity in Canada and a B.sc degree in Physics from Sierra Leone.

Astrid Dufborg, is Ambassador and ICT Adviser at the Swedish mission to the United Nations in Geneva, leading the Swedish government’s work in relation to the WSIS. Ms. Dufborg is also a member of the United Nations ICT Task Force where she amongother things convenes the Working Group on the Enabling Environment. She is alsocurrently the Vice-Chair for the Global e-Schools and Communities Initiative (GeSCI).Ms. Dufborg has a vast experience working with development issues, having beenemployed by the Swedish International Development cooperation Agency (Sida) formore than 30 years of which 10 years have been spent in various African countries. Her last position was as Assistant Director General. Ms. Dufborg has a political scienceeducational background.

Anders Engvall holds Master Degrees in Business Administration and ElectricalEngineering. He has more than 30 years experience in telecommunications includingsenior management positions in Ericsson and Hewlett Packard. Mr. Engvall is aspecialist in financing and structuring complex industry projects, and in assessingcorporate performance and identifying corrective measures. He has extensiveexperience of business operations in developing countries as well as of managingorganisations in highly competitive environments. He also has a long experience ofissues related to privatisation of the communications sector.

Karanja Gakio is a highly respected Internet specialist in Africa. He has developed, andled teams developing Internet technology, services and applications professionally forover 10 years. As the original founder of Africa Online and its technical director he wasinstrumental in establishing Internet infrastructure in six African countries. Mostrecently he was director of international engineering for iBasis in Boston, the largestInternet telephony company in the world. In this position, he led engineering anddeployment of the network, growing coverage from 2 to over 60 countries and wasinstrumental in successful project management throughout the organization. Karanja iscurrently CEO of Cyberplex Africa, based in Botswana.

Olof Hesselmark holds a Masters Degree in Business Administration and a degree inInformatics. He has more than 30 years experience in IT, business administration anddevelopment economics. Mr. Hesselmark spent 11 years as a senior IT consultant in alarge Swedish IT consulting firm. He has also been a manager of agriculture business in Kenya and Zambia. Mr. Hesselmark has over 12 years resided in four countries inAfrica. Since 1991 he has worked as an independent consultant in IT for development.

Martin Jarrold joined the Global VSAT Forum (GVF) and was appointed to theposition of Chief of International Programme Development in June 2001. His particular responsibilities include outreach to the member organisations of the GVF and for thefurther development of the profile of the Forum within the satellite communicationsindustry, and across the global telecommunications policy and regulatory community.

About the Authors | 173

Through this role Mr. Jarrold has become a frequent contributor to varioustelecommunications publications. Prior to joining the GVF, Mr. Jarrold wasCommissioning Editor and Head of Research for Space Business Internationalmagazine. His earlier career was predominantly in teaching and writing. Mr. Jarroldholds an honours degree in History and Politics from the University of Keele in theUnited Kingdom.

Mike Jensen is an independent consultant with experience in more than 30 countries inAfrica, assisting in the establishment of information and communications systems overthe last 15 years. He subsequently returned to South Africa to work as a journalist onthe national Rand Daily Mail newspaper in Johannesburg in 1983. When the paperclosed he moved back to Canada and in 1986 he co-founded the country's nationalInternet service provider for NGOs, called coincidentally, The Web. After helping toset up a similar ISP in Australia in 1989, he returned to South Africa where he workswith international development agencies, the private sector, NGOs and governmentsassisting them in the formulation, management and evaluation of their Internet projects and Telecom projects, focusing on public access, wireless technologies, VSAT and ITinfrastructure. Jensen is a trustee of the African IT Education Trust, a board member of the South African Internet service provider for NGOs - SangoNet - and was a member of the African Conference of Ministers' High Level Working Group which developedthe African Information Society Initiative (AISI) in 1996.

Jamo Macanze is currently working as Manager of the Technology and BusinessIncubator and as Scientific Officer for MICTI. He is member of the MICTIimplementation team which developed the MICTI business plan. He is also a memberof the African Research for Information Society Emergence (ARISE) and has published various papers in the Mozambican Engineer Congress and on the IDRC AcaciaProgramme publications. Mr. Macanze obtained his licentiate in electrical engineeringfrom UEM, where he started to work in 1999 as a research assistant in the Mozambique Acacia Advisory Committee Secretariat, an organization developing, coordinating andsupporting ICT projects for development and research.

Daniel Mannestig has eight years of professional experience as manager, businessdeveloper, project manager, coordinator, systems evaluator, developer and systemsimplementer in various companies, having worked in Brazil, Sweden and Mozambique.He came to Mozambique in January 2002, employed by Sida (Swedish InternationalDevelopment Agency), working for CIUEM in various projects. Later, he planned,implemented and managed the MICTI Technology and Business Incubator, and thenimplemented the MICTI E-learning centre. He is currently working as a consultant atMICTI. Mr. Mannestig has an MS in Computer Engineering and MS in BusinessStudies.

Stelios Papadakis was born in Mozambique where he did his primary and secondarystudies. In 1992, he won a scholarship to attend a pre-university program at Waterford Kamhlaba United World College Of Southern Africa, Swaziland. He graduated in 1996, and was awarded the International Baccalaureate (IB) Diploma. In 1997, he wasawarded a second scholarship to study in Australia. In 2002, Stelios graduated fromCurtin University Technology - Australia, being awarded a bachelors degree in


Electronics and Communication Engineering. After graduating, Stelios has worked asproduct manager for a computer company in Mozambique, as well as being a part-timelecturer in some private universities. In 2003 he joined the Ministry Of Transport andCommunications, taking the position of senior telecommunication adviser. Currently,apart from working at the Ministry, he is also a part-time lecturer at the High Institutefor Science and Technology of Mozambique (ISCTEM), lecturing ontelecommunication subjects.

Björn Pehrson is a full professor and head of the Telecommunication SystemsLaboratory at KTH, the Royal University of Technology in Stockholm, Sweden, since1992. He has his BSc in 1966 from the University of Stockholm, MSc in 1969 and PhD in 1975, both from Uppsala University, where he also worked as a senior lecturer up to 1985. During the period 1985-1992, he pioneered the establishment of the Swedish Institute of Computer Science, 1992-97 he served as department head. 1997-1999 hespent as visiting scholar at Stanford leading the Sweden - Silicon Valley Link program supported by Knut and Alice Wallenberg foundation, resulting in the WallenbergGlobal Learning Network and the Wallenberg Hall at Stanford. Since 2000, he isinvolved in development cooperation in several countries in Africa, Asia and LatinAmerica supported by Sida and other development cooperation agencies, currently as director of ICT development projects in Laos and technical coordinator of theSarua-Fibre project supported by IDRC with the objective to establish an academicnetwork serving universities in Southern and East Africa. He has also served as vicedean, director for a graduate school and program director of master programs inInternetworking and ICT Entrepreneurship.

Constantino Sotomane, since 2003, is the project developer of MICTI with generalresponsibility for development of the project. From 1997-2000, he served as head ofcomputer maintenance department at CIUEM, the Informatics Centre of UEM. Hejoined CIUEM in 1989, participating in various projects, including implementation of a computer network in university departments and participation in design,implementation and management of Telecentre project and Schoolnet project inMozambique. He also participated in several other projects related to ICT fordevelopment in the country and is now studying for a diploma in Business management from Cambridge University. Mr. Sotomane completed his MS in Computer Science atUniversidad de San Nicolas de Hidalgo, Mexico, and degree in electrical engineering at the Eduardo Mondlane University (UEM), Mozambique.

Roy Steiner is the former CEO of Cyberplex Africa, a regional web solution andknowledge management company with offices in Gaborone, Harare and Pretoria. HisUndergraduate education was completed at the Massachusetts Institute of Technologyand Harvard University after which he went to Cornell University for a Ph.D inengineering. Dr Steiner worked at the Rockefeller Foundation in New York, McKinsey&Company in Toronto and helped found two Internet companies, Africa Online, which is now the most widespread Internet Service Provider in Africa and Cyberplex Africa.He has a long standing interest in African Universities and has played critical roles in the development of TEEAL - The Essential Electronic Agricultural Library among otherprojects.

About the Authors | 175

Nyasha Tirivayi completed her M.Sc. in Agricultural Economics at the University ofZimbabwe. She currently works as an evaluation and knowledge managementconsultant with Cyberplex Africa. Her research interests include ICT for development,environmental economics and management gender economics and rural development.She intends to return to graduate school in pursuit of a Ph.D in about a year or two.

Rahul Tongia is a research faculty member at Carnegie Mellon University (CMU), inthe School of Computer Science and the Dept. of Engineering & Public Policy, wherehe focuses on interdisciplinary issues of technology policy. He is also Senior Fellow inthe Centre for Study of Science, Technology and Policy (CSTEP), a non-profit ThinkTank under incorporation in India. His core interests are in infrastructure development, with an emphasis on technology, policy, economics, and security. He is active in thetelecom sectors (including digital divide and access technology issues) and in the power sector (including reform, regulation, and IT/smart metering). He is presently Vice-Chair of the UN ICT Task Force Working Group on Enabling Environment (formerly, Low-cost Connectivity Access), and has organized global conferences on ICT andSustainable Development for the UN, World Bank, and National Science Foundation.His undergraduate education was in Electrical Engineering from Brown University, and he holds a doctorate from CMU.

Kate Wild is an Independent consultant on ICTs and development based in Toronto.She has more than thirty years of experience in information and development and hascombined management at the international level with a wide variety of project andprogram responsibilities in all developing regions. She has developed a realisticappreciation of both the potential and the limits of technology and a keen awareness of the importance of grounding external initiatives firmly in realities of the countries inwhich they will be implemented. Before starting work as an independent consultant Ms. Wild has held senior positions within IDRC and the ILO as well as having served asAdvisor to the Mozambique Information Policy Commission and Mozambique AcaciaAdvisory Committee on ICT policy, strategy and project development. Ms. Wild holds a BA (hons) in Philosophy and History from the University of Toronto and a MLSInformation Studies from Syracuse University.