KNOWLEDGE ORGANIZATION KO - Ergon-Verlag

Knowl. Org. 29(2002)No.2

KNOWLEDGE ORGANIZATION KOOfficial Quarterly Journal of the International Society for Knowledge Organization ISSN 0943 – 7444

International Journal devoted to Concept Theory, Classification, Indexing and Knowledge Representation

Contents

Call for papers for special issue of KnowledgeOrganization: Domain Analytic Contributionsto Knowledge Organization.Guest Editor: Birger Hjørland. Royal School ofLibrary and Information Science. Copenhagen................ 61

Call for papers and conference announcement forISKO-France ‘4: Conceptual Approaches inKnowledge Organization. Grenoble, France.July 3-4, 2003...................................................................... 62

Conference announcement for CAL ’03 21stCentury Learning Conference. Queen’s UniversityBelfast, Northern Ireland. April 8-10, 2003....................... 63

Conference announcement for the 6th Conferenceof the ISKO Chapter: 4th International Colloquiumon Library and Information Science. Salamanca,Spain. May 5-7, 2003 .......................................................... 64

Erratum............................................................................ 64

Articles

Devadason, F. J.; Intaraksa, Neelawat;Patamawongjariya, Ponprapa; and Desai, Kavita.Faceted Indexing Based System for Organizingand Accessing Internet Resources...................................... 65

Žumer, Maja and Reisthuis, Gerhard, J. A.Consequences of Implementing FRBR:Are We Ready to Open Pandora’s Box? ........................... 78

Miller, Uri and Teitelbaum, Ruth.Pre-Coordination and Post-Coordination:Past and Future. ................................................................. 87

KO Reports

Williamson, Nancy. Challenges in KnowledgeRepresentation and Organization for the21st Century: Integration of KnowledgeAcross Boundaries; Seventh ISKO InternationalConference, Granada, Spain, 10-13 July 2002. .................. 94

Book Reviews

SCHMITZ-Esser, Winfried. Expo-Info 2000:visuelles Besucherinformationssystem für Weltaus-stellungen [Expo-Info 2000 : a visual informationsystem for visitors of World Exhibitions]. Berlin,Germany : Springer, 2000. xii, 119 p.ISBN 3-540-67307-5. ......................................................... 103

WRIGHT, Sue Ellen, and BUDIN, Gerhard, eds.Handbook of Terminology Management : Vol.2Applications-oriented Terminology Management.Amsterdam, The Netherlands : John Benjamins,2001. 549 p (pp. 371-920). ISBN 9-0272-2155-3................ 104

SMIRAGLIA, Richard P. The nature of “a work”:Implications for the organization of knowledge.Lanham, MD : Scarecrow Press, 2001.ISBN 0-8103-4037-5. ......................................................... 107

TAYLOR, Arlene G. Wynar’s Introduction toCataloging and Classification. 9th ed. Assisted byDavid P. Miller. Englewood, CO : LibrariesUnlimited, 2000. xx, 552 p. ISBN 1-56308-857-6(pbk).... 109

Knowledge Organization Literature

29(2002)No.2......................................................................... 110

Personal Author Index

29(2002) ................................................................................. 121


KO KNOWLEDGE ORGANIZATIONOfficial Quarterly Journal of the International Society for Knowledge Organization ISSN 0943 – 7444


Contents page

Francis J. Devadason, Neelawat Intaraksa, Ponprapa Pa-tamawongjariya, Kavita Desai. (2002). Faceted IndexingBased System for Organizing and Accessing Internet Re-sources. Knowledge Organization, 29(2). 61-77. 20 refs.

ABSTRACT: Organizing and providing access to the re-sources on the Internet has been a problem area in spite of theavailability of sophisticated search engines and other softwaretools. There have been several attempts to organize the re-sources on the World Wide Web. Some of them have tried touse traditional library classification schemes such as the Li-brary of Congress Classification, the Dewey Decimal Classi-fication and others. However there is a need to assign propersubject headings to them and present them in a logical or hi-erarchical sequence to cater to the need for browsing. Thispaper attempts to describe an experimental system designedto organize and provide access to web documents using a fac-eted pre-coordinate indexing system based on the Deep Struc-ture Indexing System (DSIS) derived from POPSI (Postulatebased Permuted Subject Indexing) of Bhattacharyya, and thefacet analysis and chain indexing system of Ranganathan. Aprototype software system has been designed to create a data-base of records specifying Web documents according to theDublin Core and to input a faceted subject heading accordingto DSIS. Synonymous terms are added to the standard termsin the heading using appropriate symbols. Once the data areentered along with a description and the URL of the webdocument, the record is stored in the system. More than onefaceted subject heading can be assigned to a record dependingon the content of the original document. The system storesthe surrogates and keeps the faceted subject headings sepa-rately after establishing a link. The search is carried out onindex entries derived from the faceted subject heading usingthe chain indexing technique. If a single term is input, the sys-tem searches for its presence in the faceted subject headingsand displays the subject headings in a sorted sequence reflect-ing an organizing sequence. If the number of retrieved head-ings is too large (running into more than a page) the user hasthe option of entering another search term to be searched incombination. The system searches subject headings alreadyretrieved and looks for those containing the second term. Theretrieved faceted subject headings can be displayed andbrowsed. When the relevant subject heading is selected thesystem displays the records with their URLs. Using the URL,the original document on the web can be accessed. The proto-type system developed in a Windows NT environment using

ASP and a web server is under rigorous testing. The databaseand index management routines need further development.

Maja Žumer, Gerhard J.A. Riesthuis. (2002). Consequencesof Implementing FRBR: Are We Ready to Open Pan-dora’s Box? Knowledge Organization, 29(2). 78-86. 24 refs.

ABSTRACT: The study Functional Requirements for Biblio-graphic Records (FRBR) was commissioned by IFLA and pub-lished in 1998. It defines the core functions of a catalogue (andbibliographic records) as a gateway to information. For thatpurpose an abstract entity-relationship model of a catalogue isproposed. The FRBR model is revolutionary. The (computer)catalogue is not seen as a sequence of bibliographic recordsand a replica of the traditional card catalogue, but rather as anetwork of connected entities, enabling the user to performseamlessly all the necessary functions. So far there has beensome theoretical discussion of the model and some limitedexperiments, but there is a lack of research in how to imple-ment this theoretical model in practice, in new-generationcatalogues. In this paper some reactions to the model are ana-lysed. The main focus is on consequences of the model for theOPAC interface design, particularly the searching functional-ity and display of results.

Uri Miller, Ruth Teitelbaum. (2002). Pre-Coordination andPost-Coordination: Past and Future Knowledge Organiza-tion, 29(2). 87-93. 31 refs.

ABSTRACT: This article deals with the meaningful process-ing of information in relation to two systems of informationprocessing: pre-coordination and post-coordination. The dif-ferent approaches are discussed, with emphasis on the needfor a controlled vocabulary in information retrieval. Assignedindexing, which employs a controlled vocabulary, is de-scribed in detail. Types of indexing language can be dividedinto two broad groups – those using pre-coordinated termsand those depending on post-coordination. They representtwo different basic approaches in processing and informationretrieval. The historical development of these two approachesis described, as well as the two tools that apply to these ap-proaches: thesauri and subject headings.

This contents page may be reproduced without charge.




KNOWLEDGE ORGANIZATION

This journal is the organ of the INTERNATIONAL SOCIETYFOR KNOWLEDGE ORGANIZATION (General Secretariat:Dr. Gerhard RIESTHUIS (Literature Editor), Liendenhof 60, NL-1108 HB Amsterdam, Netherlands).

Editors

Dr. Hope A. OLSON (Editor-in-Chief), School of Library and In-formation Studies, 3-20 Rutherford South, University of Alberta,Edmonton, Alberta T6G 2J4, Canada. Email: [email protected]

Dr. Michèle HUDON (Book Review Editor), School of Libraryand Information Science, University of Montréal, Québec, Canada

Prof. Dr. Jens-Erik MAI (Reports Editor), Information School,University of Washington, Mary Gates Hall, Box 352840, Seattle,WA 98195-2840, USA. Email: [email protected]

Dr. Gerhard RIESTHUIS (Literature Editor), Liendenhof 60, NL-1108 HB Amsterdam, Netherlands

Dr. Nancy WILLIAMSON (Classification Research News Editor),Faculty of Information Studies, University of Toronto, 140 St.George Street, Toronto, Ontario M5S 3G6 Canada

Hanne ALBRECHTSEN (Communications Editor), Risø Na-tional Laboratory, Roskilde, Denmark.E-mail: [email protected]

Georgina R. Zaharia SARANCHUK (Editorial Assistant), Schoolof Library and Information Studies, 3-20 Rutherford South, Uni-versity of Alberta, Edmonton, Alberta T6G 2J4, Canada. Email:[email protected]

Consulting Editors

Prof. Dr. Ulf G. BARANOW, Rua Ubaldino do Amaral 580,Apto. 51, BR-80060-190, Curitiba - PR, Brazil

Prof. Clare BEGHTOL, Faculty of Information Studies, Univer-sity of Toronto, 140 St. George Street, Toronto, Ontario M5S3G6, Canada

Dr. Gerhard BUDIN, Dept.of Philosophy of Science, Universityof Vienna, Sensengasse 8, A-1090 Wien, Austria

Mr. Christian GALINSKI, Director: Infoterm, SimmeringerHauptstr. 24, A-1110 Wien, Austria

Dr. Francisco Javier GARCIA MARCO, Area de Biblioteconomiay Documentacion, Facultad de Filosofia y Letras, Universidad deZaragoza, E-50.009 Zaragoza, Spain

Dr. M. A. GOPINATH, DRTC, Indian Statistical Inst., 8th Mile,Mysore Road, R.V. College P.O., Bangalore-560059, India

Dr. Rebecca GREEN, College of Information Studies, HornbakeBldg. (So. Wing), Room 4105, University of Maryland, CollegePark, MD 20742-4345

Dr. Norbert HENRICHS, Im Luftfeld 80, D-40489 Düsseldorf,Germany

Dr. Roland HJERPPE, Dept. of Computer and Information Sci-ence, Linköping University, S-58183 Linköping, Sweden

Dr. Barbara H. KWASNIK, School of Information Studies, Syra-cuse University, 4-206 Ctr. for Science & Tech., Syracuse, NY13244-4100

Dr. Jacques MANIEZ, 61 Rue de la Corve, F-21000 Dijon, France

Dr. Ia C. MCILWAINE, Director: School of Library, Archive &Information Studies, University College London, Gower Street,London, WC1E 6BT, U.K.

Prof. Dr. Francis MIKSA, Graduate School of Library and Infor-mation Science, University of Texas at Austin, SZB 564, Austin,TX 78712-1276

Ms. Joan S. MITCHELL, Editor: Dewey Decimal Classification,Library of Congress, Washington, D.C. 20540-4330

Dr. Widad MUSTAFA el HADI, URF IDIST, Université Charlesde Gaulle Lille 3, BP 149, 59653 Villeneuve D’Ascq, France

Dr. Giliola NEGRINI, Consiglio Nazionale della Ricerche, Ist. diStudi sulla Ricerca e Doc. Scientifica, Via Cesare De Lollis, 12, I-00185 Rome, Italy

Ms. Annelise Mark PEJTERSEN, Systems Analysis Dept., RisoeNational Laboratory, P.O. Box 49, DK-4000 Roskilde, Denmark

Dr. Stephen A. POLLITT, CeDAR Centre for Database AccessResearch, School of Computing & Mathematics, The Universityof Huddersfield, Queensgate, Huddersfield HDI 3DH, U.K.

Dr. M. P. SATIJA, Guru Nanak Dev University, School of Li-brary and Information Science, Amritsar-143 005, India

Dr. Jos SCHREINEMAKERS, Erasmus University, Rotterdam,PO Box 1738, 3000 DR Rotterdam, Netherlands

Dr. Otto SECHSER, In der Ey 37, CH-8047 Zürich, Switzerland

Dr. Winfried SCHMITZ-ESSER, Rothenbaumchaussee 3, D –20148 Hamburg, Germany

Dr. Dagobert SOERGEL, College of Information Studies, Horn-bake Bldg. (So. Wing), Room 4105, University of Maryland, Col-lege Park, MD 20742

Dr. Eduard R. SUKIASYAN, Chief Editor of the Library-Bibliographical Classification, Russian State Library, 3 Vozdviz-henka Str., Moscow, 101000 Russia

Dr. Elaine SVENONIUS, Graduate School of Education and In-formation Studies, UCLA, 5380 ½ Village Green, Los Angeles,CA 90016

Dr. M.S. van der WALT, Department of Information Science,University of Stellenbosch, Private Bag X1, Stellenbosch 7602,South Africa

Prof. Dr. Harald ZIMMERMANN, Softex, Schmollerstrasse 31,D-66111 Saarbrücken, Germany


KO KNOWLEDGE ORGANIZATIONOfficial Quarterly Journal of the International Society for Knowledge Organization ISSN 0943 – 7444


PublisherERGON-Verlag, Grombühlstr. 7, GER-97080 WürzburgPhone: +49 (931) 280084; FAX +49 (931) 282872E-mail: [email protected]; http://www.ergon-verlag.de

Editor-in-chief (Editorial office)Dr. Hope A. OLSON (Editor-in-Chief), Georgina R. ZahariaSARANCHUK (Editorial Assistant), School of Library and Infor-mation Studies, 3-20 Rutherford South, University of Alberta,Edmonton, Alberta T6G 2J4, Canada.Fax +1 (780) 492-2430; Email: [email protected]

Instructions for Authors

Manuscripts should be submitted electronically (in Word,Wordperfect, or RTF format) in English only to the editor-in-chiefand should be accompanied by an indicative abstract of 100 or 200words. Submissions via email are preferred; submissions will also beaccepted via post provided that four copies are submitted or thatsubmissions are accompanied by a 3.5” diskette encoded in Word,Wordperfect, or RTF format.

A separate title page should include the article title and theauthor’s name, postal address, and E-mail address, if available. Onlythe title of the article should appear on the first page of the text. Toprotect anonymity, the author’s name should not appear on themanuscript, and all references in the body of the text and in foot-notes that might identify the author to the reviewer should be re-moved and cited on a separate page. Articles that do not conform tothese specifications will be returned to authors.

Criteria for acceptance will be appropriateness to the field of thejournal (see Scope and Aims), taking into account the merit of thecontents and presentation. The manuscript should be concise andshould conform as much as possible to professional standards ofEnglish usage and grammar. Manuscripts are received with the un-derstanding that they have not been previously published, are notbeing submitted for publication elsewhere, and that if the work re-ceived official sponsorship, it has been duly released for publication.Submissions are refereed, and authors will usually be notifiedwithin 6 to 10 weeks. Unless specifically requested, manuscripts andillustrations will not be returned.

The text should be structured by numbered subheadings. Itshould contain an Introduction, giving an overview and stating thepurpose, a main body, describing in sufficient detail the materials ormethods used and the results or systems developed, and a conclusionor summary.

Reference citations within the text should have the followingform: (author, year). For example, (Jones, 1990). Specific page num-bers are optional, but preferred when applicable, e.g. (Jones, 1990,p.100). A citation with two authors would read (Jones & Smith,1990); three or more authors would be: (Jones et al., 1990). Whenthe author is mentioned in the text, only the date and optional page

number should appear in parenthesis – e.g. According to Jones(1990), …

References should be listed alphabetically by author at the end ofthe article. Journal names should not be abbreviated. Multiple cita-tions by the same author should be listed chronologically andshould each spell out the author’s name. Articles appearing in thesame year should have the following format: Jones, T. (1990a) ... ,Jones, T. (1990b) …Examples:Dahlberg, I. (1978). A referent-oriented, analytical concept theory

for INTERCONCEPT. International Classification, 5(3). 142-151.

Graesser, A., Person, N. & Huber, J. (1992). Mechanisms that gen-erate questions. In T. W. Lauer, E. Peacock, & A. C. Graesser(Eds.). Questions and Information Systems. Hillsdale, NJ: Law-rence Erlbaum Associates. 167-187.

Sager, J.C. (1990). A Practical Course in Terminology Processing. Am-sterdam: John Benjamins.

Sukiasyan, E. R. (1996). Change as a problem of classification sys-tem development. In R. Green (Ed.). Knowledge Organizationand Change: Proceedings of the 4th International ISKO Conference.Frankfurt: Indeks Verlag. 119-122.Footnotes should be kept to a minimum. They should be indi-

cated in the text with numbered superscripts, and the correspond-ing notes should be collected at the end of the article, before the ref-erences, under the heading Notes.

Illustrations should be kept to a necessary minimum and shouldbe submitted electronically when possible. Photographs (includingcolor and half-tone) should be scanned with a minimum resolutionof 600 dpi and saved as tif files (Tagged Image File Format pre-ferred). Tables and figures should be embedded within the docu-ment or, alternatively, saved as separate files with clear instructionsindicating their placement in the text. Tables should contain anumber and title at the top, and all columns and rows should haveheadings. All illustrations should be cited in the text as Figure 1,Figure 2, etc. or Table 1, Table 2, etc. Illustrations submitted in hardcopy only should be marked to indicate their placement in the text.

Upon acceptance of a manuscript for publication, authors mustprovide a wallet-size photo and a one-paragraph biographicalsketch. The photograph should be scanned with a minimum resolu-tion of 600 dpi and saved as a tif file (Tagged Image File Format).AdvertisingResponsible for advertising: Dr. H.-J. Dietrich, ERGON-Verlag,Grombühlstr. 7, 97080 Würzburg (Germany).© 2002 by ERGON-Verlag Dr. H.-J. Dietrich.All Rights reserved.Printed in the Federal Republic of Germanyby Offizin Hildburghausen GmbH Druckhaus.

KO is published quarterly by ERGON-Verlag.The price is € 89,50/ann. including airmail delivery.




Scope

The more scientific data are generated in the impetuous pre-sent times, the more ordering energy needs to be expended tocontrol these data in a retrievable fashion. With the abundanceof knowledge now available the questions of new solutions tothe ordering problem and thus of improved classification sys-tems, methods and procedures have acquired unforeseen sig-nificance. For many years now they have been in the focus ofinterest of information scientists the world over.

Until recently, the special literature relevant to classifica-tion was published in piecemeal fashion, scattered over thenumerous technical journals serving the experts of the variousfields such as

philosophy and science of sciencescience policy and science organizationmathematics, statistics and computer sciencelibrary and information sciencearchivistics and museologyjournalism and communication scienceindustrial products and commodity scienceterminology, lexicography and linguistics

Beginning 1974, KNOWLEDGE ORGANIZATION(formerly INTERNATIONAL CLASSIFICATION) has beenserving as a common platform for the discussion of both theo-retical background questions and practical application prob-lems in many areas of concern. In each issue experts from manycountries comment on questions of an adequate structuring andconstruction of ordering systems and on the problems of theiruse in opening the information contents of new literature, ofdata collections and survey, of tabular works and of other ob-jects of scientific interest. Their contributions have been con-cerned with

(1) clarifying the theoretical foundations (general orderingtheory/science, theoretical bases of classification, dataanalysis and reduction)

(2) describing practical operations connected with indexing/classification, as well as applications of classification sys-tems and thesauri, manual and machine indexing

(3) tracing the history of classification knowledge and meth-odology

(4) discussing questions of education and training in classifi-cation

(5) concerning themselves with the problems of terminologyin general and with respect to special fields.

Aims

Thus, KNOWLEDGE ORGANIZATION is meant to be aprogramme for the improvement of classification methods andprocesses, a forum for discussion for all those interested in theorganization of knowledge on a universal or a subject-field scale,using concept-analytical and/or concept-synthetical approaches aswell as numerical procedures and comprising also the intellectualand automatic compilation and use of classification systems andthesauri in all fields of knowledge, with special attention beinggiven to the problems of terminology.

KNOWLEDGE ORGANIZATION publishes original arti-cles, reports on conferences and similar communications, theNewsletters of the International Society for Knowledge Organi-zation (ISKO News) and the Committee on Classification Re-search of the International Federation for Information andDocumentation (FID/CR News) as well as book reviews, lettersto the editor, and an extensive annotated bibliography of recentclassification and indexing literature, covering some 500 items ineach issue.

KNOWLEDGE ORGANIZATION should therefore beavailable at every university and research library of every coun-try, at every information center, at colleges and schools of libraryand information science, in the hands of everybody interested inthe fields mentioned above and thus also at every office for updat-ing information on any topic related to the problems of order inour information-flooded times.

KNOWLEDGE ORGANIZATION was founded in 1973 byan international group of scholars with a consulting board of edi-tors representing the world’s regions, the special classificationfields, and the subject areas involved. From 1974-1980 it was pub-lished by K.G. Saur Verlag, München. Back issues of 1978-1992are available from ERGON-Verlag, too.

As of 1989, KNOWLEDGE ORGANIZATION has becomethe official organ of the INTERNATIONAL SOCIETY FORKNOWLEDGE ORGANIZATION (ISKO) and is included forevery ISKO-member, personal or institutional in the member-ship fee (US $ 55/US $ 110).

Rates: From 2001 on for 4 issues/ann. (including indexes)€ 98,- (forwarding costs included). Membership rates see above.ERGON-Verlag, Grombühlstr. 7, GER-97080 Würzburg;Phone: +49 (931) 280084; FAX +49 (931) 282872; E-mail:[email protected]; http://www.ergon-verlag.de

The contents of this journal are indexed and abstracted in ReferativnyiZhurnal Informatika and in the following online databases: InformationScience Abstracts, INSPEC, Library and Information Science Abstracts(LISA), Library Literature, PASCAL and Sociological Abstracts.

Knowl. Org. 29(2002)No.2 61

Domain Analytic Contributions to Knowledge Organization.

A Special Issue of Knowledge Organization on Domain Analytic Methods and Contributions.

Guest Editor: Birger Hjørland, Royal School of Library and Information Science, Copenhagen.Co-editor: Jenna Hartel, Department of Information Studies, University of California-Los Angeles

Call for papers

The last few years have seen a growing interest in morecontextual, social and cultural approaches to Knowledge Or-ganization. Concepts such as discourse analysis, epistemo-logical and historical perspectives, and hermeneutics havegained ground in knowledge organization as in library andinformation science in general. Such new approaches sup-plement more traditional approaches.

In the forthcoming special issue of Knowledge Organiza-tion, DOMAIN ANALYSIS will be understood in an open wayas encompassing different approaches to studying the or-ganization of knowledge fields. Many different approachesmay be used and may contribute to the construction andevaluation of systems for knowledge organization includingfacet analysis, and bibliometrical, historical, sociological,linguistic, epistemological and pragmatic approaches. Anygiven method may be applied to different domains, andcontributions to the study and construction of knowledgefields broad or narrow, scientific, scholarly, applied, com-mercial or private are welcome.

Papers that are able to interpret concrete systems forknowledge organization (such as traditional thesauri, classi-fication systems and indexing) in the light of more generalperspectives are particularly welcome.

This Special Issue of Knowledge Organization (scheduledas volume 30, number 3/4) will aim at bringing together thecommunity of researchers working with special approachesto domain analysis as well as scholars interested in specificsubject fields. It will aim at providing an overview of differ-ent approaches as well as of substantial knowledge aboutknowledge organization in specific fields. Further, it willprovide initiatives for strengthening international coopera-tion in these areas.

We invite original and unpublished submissions to thespecial issue covering any of, but not limited to, the follow-ing topics. Contributions previously published in a lan-guage other than English may be considered.

Possible Topics

A: Approaches to domain analysis– Empirical approaches (e.g., bibliometric approaches)– Epistemological, pragmatic or critical approaches– Historical approaches– Sociological approaches– Discourse analytic and linguistic approaches (languages

for special purposes, composition studies, genre analysis)– Rational approaches (e.g., facet-analytic approaches)

B: Investigations of specific domains– Humanities

e.g., arts, history– Social sciences

e.g., sociology, law, economics– Sciences

e.g., chemistry, mathematics– Technologies

e.g., information technology, building research– Applied fields

e.g., nursing

Also welcome are:Descriptions of ongoing projectsBook reviewsBrief communications

Submission Information

Deadline for submissions is: 1 May 2003

Submissions will be refereed by a special committee of ex-perts in the field. Manuscripts should be submitted elec-tronically in Word, WordPerfect or RTF format in Englishonly and should be accompanied by an indicative abstractof 100-200 words.

Submissions via e-mail are preferred; submissions will beaccepted via post provided that four copies are submitted orthat submissions are accompanied by a 3.5" diskette en-coded in Word, WordPerfect or RTF format.

Please refer to the instructions for authors published ineach issue of Knowledge Organization.

Please use the subject line: KO DOMAINSubmissions should be directed to:Birger HjørlandE-mail: [email protected]

Birger HjørlandRoyal School of Library and Information Science6 Birketinget,DK-2300 Copenhagen S, DenmarkTel: +45 32 58 60 66Fax: + 45 32 84 02 01

Knowl. Org. 29(2002)No.262

ISKO-France ‘4

Conference Announcement and Call for Papers

ISKO-France, in cooperation with RI3 Research Group and the IUT2, University Pierre Mendès France – Grenoble 2 will con-duct its fourth conference in Grenoble, ISKO-France ‘4, July 3-4, 2003. The theme of the conference is:

Conceptual Approaches in Knowledge Organization

All papers must be in the French language for publicationpurposes. English speakers wishing to present their papersin English can do so; although we cannot guarantee thatsimultaneous translation facilities will be available.

The aim of the conference is to bring together the com-munity of researchers in the field of Knowledge Organiza-tion (researcher, academics, practitioners and experts) fromvarious specialties to exchange theoretical approaches toknowledge organization.

The Conference objectives are to:– promote scientific thought on knowledge organization,

go beyond documentation approaches, and look at ob-jects, organizations and taxonomies, and

– extend the scope of the domain and open it to practicalapproaches.

The theme of the conference is: Conceptual Approaches inKnowledge Organization.This field had been restricted fora long time to researchers and practitioners in InformationScience and its main focus was classification and indexing(International Classification had been, for a long time, thename of the ISKO periodical).

Since its creation the French Chapter has organized bi-ennial conferences: Lille, 1997; Lyon, 1999; and Paris, 2001.The themes of the three conferences were, respectively:Knowledge as an Element in Representational and Informa-tion Retrieval Systems; Indexing in the Era of the Internet;and Information Filtering and Automatic Summarizationin Networks. The next conference in Grenoble will ad-dress, specifically, conceptual approaches to knowledge or-ganization.

Knowledge organization should be understood as:– all types of schemas from simple alphabetic lists superfi-

cially structured (e.g., authority lists, glossaries, diction-aries, nomenclatures) to hierarchical classification sche-mas (e.g., general and special classifications, taxonomies,nomenclatures) and to classificatory schemas (e.g., classi-fication display, subject headings) or other types of or-ganizing that give precedence to non-exclusively hierar-chical relationships;

– dealing with all kinds of objects from documents in theclassical meaning of the term (e.g., texts, stationary ormoving images, and voice recordings) to whole concrete

and abstract phenomena or discrete ones that can becounted, organized and dealt with (e.g., objects, events,and processes); and

– dealing with different aims and objects to retrieve in-formation, teach, produce new knowledge, communi-cate, and apply appropriate processed treatments.

In addition to information specialists the conference willinvolve researchers and practitioners dealing with knowl-edge organization in all its forms, whatever their field of in-terest might be: epistemology, philosophy, pedagogy, natu-ral sciences, publishing, product design, and multimediaproducts.

The conference is structured around five different topics,but is not limited to them:1. Theories, paradigms and history of knowledge organiza-

tion. This theme aims at bringing about proposals onparadigms and principles underlying the origins of natu-ral and medical classifications, together with encyclope-dic classifications concerning the interactions betweenknowledge development and science as they are struc-tured in institutional circles.

2. Knowledge organization and language: concepts, terms,terminologies, ontologies. This will deal with basic unitsin terminologies and documentation, concept relation-ships, words and terms, key-words (e.g., identificationcriteria, properties) from ontologies to terminologytools and design considered as structuring elements of adomain of knowledge.

3. Knowledge organization and formalization. This topiccalls for studies dealing with categorization, technicalrepresentation, and models design.

4. Socioeconomic and political stakes in knowledge organi-zation. This topic aims at revealing the invisible aspectsof knowledge organization schemes on social organiza-tion such as the effect of statistical categories, adminis-trative categories (e.g., medical acts, customs), politicaldecisions and the social order.

5. Interoperability of knowledge organization systems.This topic can be dealt with from the point of view of:*Systems that be can turned into compatibles ones (e.g.,general systems, universal systems, domain-specific sys-tems, systems on different linguistic and cultural bases,systems with different organizational characteristics);*Conceptual methods used for guaranteeing systems’ in-terpretability (e.g., translation,adaptation, derivation, es-

Knowl. Org. 29(2002)No.2 63

tablishing equivalences); and *Computational methodsused to manage interoperability (e.g., pivotal language,semantic networks, classes of equivalence).

The Program Committee is being organized and its compo-sition will be announced shortly.

Submission Information

Deadline for submissions is: 2 December 2002.First notification to the authors of the provisional accep-tance is: 17 January 2003.The deadline for papers submission for final acceptance is:March 2003.Second notification deadline is: 5 May 2003.The deadline for submission of papers for the printed con-ference proceedings will be 15 June 2003.

Authors are invited to first submit an electronic text ofthree pages (10 000 signs maximum) which sums up theirproposals, and mentions the relationship between theirproposals and at least one of the topics of the conference.Abstracts should be concise, clear, complete, specific, andclearly indicate the underlying methodology.

Authors should not include any personal identification intheir papers. Evaluation will be anonymous and papers willbe refereed by a special committee of experts in the field.

Submission should be exclusively via e-mail and in anRTF format ONLY, before December 2 to the followingaddress: Comite.Org@ isko-france-2003.org

For further details regarding manuscripts and publica-tions, please refer to the instructions for authors availableat: http://isko-france-2003.orgNotification for acceptance or refusal date is: 17 January2003.

Papers Selection

Each proposal will be anonymously examined by at leasttwo referees. Upon acceptance, proposals should be ex-tended to 8 pages maximum (25 000 signs) and sent before15 March 2003. A second notification will be sent on 5 May2003. This notification will indicate modifications necessaryfor final article acceptance.

Call for demonstrations

A specific session will be planned for demonstrations.Authors are invited to send their proposals, maximum fourpages, to the Organizing Committee at: [email protected] . Demonstrations will not be accepted un-less they are relevant to the general theme of the confer-ence.

Registration

For all information regarding registration, and call for pa-pers, please refer to the ISKO-France website: http://isko-france-2003.orgRegistration fees will not exceed 150 €.

Local Organizing Committee:

Conference Chair:Gérard HENNERON, IUT2, UniversitéPierre Mendès France GrenobleMembers:Sylvie DALBIN, société ATD; Céline PAGAN-ELLI, IUT2, Université Pierre Mendès France Grenoble;Rosalba PALERMITI, IUT2, Université Pierre MendèsFrance Grenoble; Yolla POLITY, IUT2, Université PierreMendès France Grenoble

CAL ’03 ‘21st Century Learning

Conference Announcement

Queen’s University Belfast, Northern IrelandApril 8-10, 2003

The CAL '03 '21st Century Learning Conference will takeplace at Queen's University Belfast, Northern Ireland, 8-10April 2003.The conference will provide a forum for the sharing of ex-perience, knowledge and research for those working at theforefront of learning and teaching with technology.Papers have been invited for oral and poster presentation atthe conference on the following themes:– Emerging Contexts for Learning (e.g., non-institutional,

informal, work-based, community-based, VLE-based);– Emerging Pedagogies (e.g., visions, designs and examples);– Assessment and Technology (e.g., innovations in assess-

ment of, with and for learning);

– Addressing the Digital Divide (e.g., national and interna-tional differences, globalisation, culture change); and

– Emerging Technologies for Educational Contexts (e.g.,educational uses and implications of emerging technolo-gies).

The deadline for submission of abstracts was 27 September2002.For further information visit http://www.cal2003.com orcontact April Williams at the CAL 2003 Conference Secre-tariat: Tel: +44 (0) 1865 843089 Fax: +44 (0) 1865 843958

Email: mailto:[email protected]

Knowl. Org. 29(2002)No.264

4th International Colloquium on Library and Information Science

6th Conference of the ISKO Spanish Chapter

Salamanca, May 5-7, 2003

The 4th International Colloquium on Library and Informa-tion Science and the 6th Conference of the ISKO (Interna-tional Society for Knowledge Organization) Spanish Chapterwill consider the research trends and methodologies inknowledge organization (KO); the paradigmatic and episte-mological aspects of this research, the ethical and social con-siderations and the dissemination of the results.

The motto of the conference is: “Trends of KnowledgeOrganization Research”. The conference program will in-clude speeches, papers, and round tables; a meeting of KOuniversity professors; and a homage to Dra. Emilia Currás.

The topics of the conference are the following:1. Trends in knowledge representation research.2. Trends in user interface improvement research.3. Trends in information retrieval systems research.4. Epistemological foundations of KO research.5. The linguistic paradigm in KO research.6. The cognitive paradigm in KO research.7. The physical paradigm in KO research.8. Methodologies, methods and techniques of data collection

and analysis in KO research.9. KO decision making research.10. The KO research environment.11. KO in the digital environment.12. Ethical, social and sociological aspects of KO research.13. Scientific communication and dissemination of KO re-

search.

The following titles and speakers have already been con-firmed:1st speech: “The research in Knowledge Organization in their

historical evolution: from the arrangement of the books inthe shelves to the search of information in the web”(JENNIFER ROWLEY, Edge Hill, Great Britain)

2nd speech: “Current trends in Knowledge Organization re-search” (IA MCILWAINE, School of Library, Archiveand Information Studies, University College London,Great Britain)

3rd speech: “Some fundamental problems in Knowledge Or-ganization” (BIRGER HJØRLAND, Royal School of Li-brary and Information Science, Denmark)

4th speech: “Applied ethics and Knowledge Organization”(CLARE BEGHTOL, Faculty of Information Studies,University of Toronto, Canada)

5th speech: “Transgressive deconstructions: feminist/post-colonial methodology for research in Knowledge Organi-zation” (HOPE A. OLSON, School of Library and In-formation Studies, University of Alberta, Canada)

6th speech: “Research in automated text classification: trendsand perspectives” (FABRIZIO SEBASTIANI, Istituto diElaborazione della Informazione, Consiglio Nazionaledelle Ricerche, Pisa, Italy)

7th speech: “The Knowledge Organization research in Spain”(MARIA J. LÓPEZ-HUERTAS, Facultad de Bibiote-conomía y Documentación, Universidad de Granada,Spain)

Information and details about this conference are available onthe ISKO Spanish Chapter webpage(http://www.ugr.es/~isko)and inscriptions can be made onhttp://www.usal.es/precurextby selecting the line “Coloquio Documentacion”.

Further inquiries should be directed to:José Antonio Frías, Universidad de Salamanca, Departamentode Biblioteconomía y Documentación, Francisco Vitoria,6-16, E-37008 Salamanca, Tlf. 34-923 294 580, Fax 34-923 294582, Móvil 678 730 536, C.e.: [email protected]

ErratumKnowledge Organization 29(2002)No.1

Page 1:

Erratum: “School of Information Studies, University of Cen-tral England in Birmingham, Perry Barr, Birmingham, De-partment of Computer and Information Sciences, Universityof Strathclyde, 26 Richmond Street, Glasgow G1 1XH Scot-land, UK”

Correction: The authors, Ali Asghar Shiri, Crawford Revieand Gobinda Chowdhury, of the article, “Thesaurus-AssistedSearch Term Selection and Query Expansion: A Review of

User-Centred Studies”, were incorrectly indicated as affiliatedwith the School of Information Studies, University of Cen-tral England in Birmingham, Perry Barr, Birmingham. Pleasenote that these authors are indeed affiliated with the Depart-ment of Computer and Information Sciences, University ofStrathclyde, 26 Richmond Street, Glasgow G1 1XH Scotland,UK. Knowledge Organization extends a sincere apology,first, to the authors, and second, to its readers, for this errorin information.

Knowl. Org. 29(2002)No.2F.J. Devadason, N. Intaraksa, P. Patamawongjariya, K. Desai: Faceted Indexing Based System ...

65

Faceted Indexing Based System for Organizingand Accessing Internet Resources

Francis J. Devadason, Neelawat Intaraksa,Ponprapa Patamawongjariya, Kavita Desai

Computer Science and Information Management, Asian Institute of Technology,P.O.Box:4, Klong Luang, Pathumthani – 12120, Thailand

Dr. Francis J. Devadason, Director of the Center for Library & Information Resources, Asian Institute ofTechnology in Bangkok, is also a Faculty Member at the Computer Science and Information ManagementProgram at the same institute. He has served as Associate Professor at the Documentation Research andTraining Centre (DRTC), Bangalore, India. He studied in India under Dr. S.R. Ranganathan, famous forhis faceted classification scheme, the Colon Classification. He has about 10 years of teaching experience atthe Master’s level and an equal number of years of experience as a practicing librarian, systems librarian, in-formation officer and information scientist – the teaching and the practice interleaved and repeated in suc-cession every few years and sometimes both roles played at the same time. He has held the position of As-sistant Director and Head of the National Information Centre for Leather and Allied Industries at theCentral Leather Research Institute, Madras, as Senior Information Scientist of the Management of Tech-nology Information Center at the Asian Institute of Technology, Bangkok and as Senior Lecturer at theUniversity of Papua New Guinea. He has about 60 papers/ publications to his credit and has presented pa-pers at international and national conferences. He has also served in the Editorial Board of “InformationProcessing and Management: An International Journal” published by Pergamon Press, USA. Dr. Devada-son holds a B.Sc. (Physics) and a B.Lib.Sc., both degrees from the University of Madras, an Associateship inDocumentation from the Documentation Research and Training Centre (Indian Statistical Institute), Ban-galore and a Ph.D. (Library & Information Science) from Karnatak University, Dharwad, India. He can becontacted at: [email protected] or [email protected]

Neelawat Intaraksa has a B.A., in Library Science from Chulalongkorn University and an M.S., in Infor-mation Management from the School of Advanced Technologies, Asian Institute of Technology, Thailand.She worked as Research Associate in the Transaction Control Log and Accounting System Software De-velopment Project of the Revenue Department, Royal Thai Government. Later she was a web designer atthe Distributed Education Center of the Asian Institute of Technology. Presently, she is a Project Associ-ate at the AIT Library looking after system administration, web site maintenance and network and soft-ware maintenance. She is interested in search engines, Internet subject gateways, user interfaces and webdesign for libraries.

Ponprapa Patamawongjariya has an M.S. in Computer Science and Information Management from theAsian Institute of Technology. She worked as a Software Engineer at the AIS Telecommunication Services,Bangkok. At present she is with TA Orange Company Limited, 968 U Chu Liang Building, 14th Floor,Rama IV Road, Bangkok 10500 Thailand, developing software for mobile applications.

Kavita Desai has an M.S. in Information Management from the School of Advanced Technologies, AsianInstitute of Technology, Bangkok. At present she is working as Software Engineer at Vicinity Corpora-tion, Hall Hanover NH 03755 USA, carrying out research and developmental work in artificial under-standing of text and in text manipulation processes. She can be reached at: [email protected]

Knowl. Org. 29(2002)No.2 F.J. Devadason, N. Intaraksa, P. Patamawongjariya, K. Desai: Faceted Indexing Based System ...

66

Francis J. Devadason, Neelawat Intaraksa, Ponprapa Patamawongjariya, Kavita Desai. (2002). Faceted Indexing Based Systemfor Organizing and Accessing Internet Resources. Knowledge Organization, 29(2). 61-77. 20 refs.

ABSTRACT: Organizing and providing access to the resources on the Internet has been a problem area in spite of the availabil-ity of sophisticated search engines and other software tools. There have been several attempts to organize the resources on theWorld Wide Web. Some of them have tried to use traditional library classification schemes such as the Library of Congress Clas-sification, the Dewey Decimal Classification and others. However there is a need to assign proper subject headings to them andpresent them in a logical or hierarchical sequence to cater to the need for browsing. This paper attempts to describe an experi-mental system designed to organize and provide access to web documents using a faceted pre-coordinate indexing system basedon the Deep Structure Indexing System (DSIS) derived from POPSI (Postulate based Permuted Subject Indexing) of Bhattacha-ryya, and the facet analysis and chain indexing system of Ranganathan. A prototype software system has been designed to createa database of records specifying Web documents according to the Dublin Core and to input a faceted subject heading accordingto DSIS. Synonymous terms are added to the standard terms in the heading using appropriate symbols. Once the data are enteredalong with a description and the URL of the web document, the record is stored in the system. More than one faceted subjectheading can be assigned to a record depending on the content of the original document. The system stores the surrogates andkeeps the faceted subject headings separately after establishing a link. The search is carried out on index entries derived from thefaceted subject heading using the chain indexing technique. If a single term is input, the system searches for its presence in thefaceted subject headings and displays the subject headings in a sorted sequence reflecting an organizing sequence. If the number ofretrieved headings is too large (running into more than a page) the user has the option of entering another search term to besearched in combination. The system searches subject headings already retrieved and looks for those containing the second term.The retrieved faceted subject headings can be displayed and browsed. When the relevant subject heading is selected the systemdisplays the records with their URLs. Using the URL, the original document on the web can be accessed. The prototype systemdeveloped in a Windows NT environment using ASP and a web server is under rigorous testing. The database and index man-agement routines need further development.

Introduction

Some of the recent findings about the Internet arethat (a) it is huge (it has exploded to more than a bil-lion web pages), (b) it is growing very fast, (c) searchengines cover only a small fraction of the web, (d)even the combined indexes of the major search en-gines cover only a small fraction of the web and (e)even the most dedicated surfer using the best searchsystem would be able to find barely one-third of thepages (Dahn, 2000). According to the survey byRoper Starch Worldwide (NUA Internet surveys re-port, 2001-03-01), poor search engines and informa-tion overload are causing web-rage among Internetusers. On average, users get angry and frustrated aftertwelve minutes of fruitless searching. For seven per-cent of respondents, it only takes three minutes be-fore web-rage strikes. The overwhelming amount ofinformation on the web is turning people away fromthe medium and back to pre-Internet information re-sources.

The main reasons are that the quality of the re-sources retrieved is doubtful, there is no simple wayto index and organize the resources on the Internetand there are no central agencies like the national li-braries to get a submission of the document/resource

to grant copyright protection and to assign an Inter-national Standard Identification Number to each ofthe resources. Several search engines using differentmethods of indexing the resources, and ranking themaccording to “their own ranking algorithms includingthe payment of fee as a major factor” are being used,and their number is steadily growing. Even afterbrowsing through the first five to seven pages of theretrieved records one is tempted to believe that theremay be some thing more relevant on the nineteenthpage of the ranked results. Meta-search facilities thatoffer several engines to be used in parallel to searchfor information are available. However, “the goal ofinformation systems is to create order in a collectionof documents so that an information searcher neednot scan the entire collection in an attempt to find in-formation of interest” (Fugmann 1999).

There have been several attempts to organize theelectronic resources including the use of traditional li-brary classification schemes such as, the Dewey Deci-mal Classification, (Mundie, D. A, 1999) Library ofCongress Classification, Universal Decimal Classifica-tion as well as special subject classification schemes formedicine, computing, and so forth (McKiernan, 1997).Moreover, several subject-based information gatewayprojects have been ongoing (Kerr & MacLeod, 1998)


67

(Mitchel & Mooney, 1996) to provide organized ac-cess to the electronic resources. Although use of clas-sification schemes helps in organizing electronicdocuments broadly and helps in browsing, they can-not be used to organize the resources precisely andfunction as an effective retrieval tool. Analysis ofsome of the websites using standard classificationschemes for organizing the resources (Williamson,1997) (McIlwaine, & Williamson, 1999) identified thefact that the application of classification at the siteswas often superficial and poorly executed. This be-comes an acute problem as the electronic documentsare mostly on recent, developing and yet to be devel-oped subject fields and facets. Relying on such classifi-cation schemes may not fully serve the purpose as e-documents get updated fast and may change their sub-ject with each modification and updating. Moreover,assignment of class numbers is needed for documentsin a library as it is necessary to restore the documentsto the shelves in the classified sequence once they areborrowed and returned, or their arrangement dis-turbed otherwise. In other words, class numbers areused to mechanize the arrangement of documents onthe shelves. In the case of documents on the Web,only their surrogates are organized and the sequenceof these surrogates is not disturbed by the users. Infact there is no need for assignment of class numbersto restore their sequence in a mechanical way. What isneeded is a tool that provides a standard frameworkor formula for the e-document developers to fill-in asa subject heading which will have the capacity to pro-duce an organizing sequence when sorted using the as-signed subject heading as key, as well as provide thenecessary index terms or keywords or search terms toprovide access. Such a heading can be incorporated asa Meta tag in the resources and used for both organiz-ing and indexing the resources. This paper describes aprototype of such a system that can provide organizedaccess to networked resources.

Faceted Indexing

Categories of Postulate Based Permuted Subject IndexingPOPSI/ DSIS

Kaiser’s “Systematic Indexing” (Kaiser, J. 1911) wasperhaps the first category-based subject indexing sys-tem. The Deep Structure Indexing System (Devada-son, 1986) is based on the Postulate-based PermutedSubject Indexing (POPSI) (Bhattacharyya, 1979) de-rived from the Chain Indexing System of Rangana-than, especially his concept of facet analysis as applied

to subject headings (Ranganathan, 1964). There havebeen studies of the relevance of facet analysis to searchand organize the resources on the Web (Ellis &Vasconcelos, 1999). The component ideas in a subjectheading, can be deemed to fall into any one of theElementary Categories or Facets : Discipline (Base),Entity (Core), Property and Action. Each of thesemay be subdivided into: species/type, and part. TheEntity may also have constituents . In the case of‘bamboo’; ‘arundinarisae’, ‘bambusa lineata’, ‘Brazil-ian bamboo’ are species/types; ‘bud’ ‘flower’, ‘leaf’,‘rhizome’, ‘culm’, ‘root’ are parts; ‘ash’, ‘lignin’,‘peptin’, ‘cellulose’ are constituents. Apart from theelementary categories, a special component calledModifier is recognized; for example: “red” in ‘redrose’, “concrete” in ‘concrete bridge’. A modifier gen-erally creates a species/type of the concept modified.Modifiers can be Common Modifiers like Form,Time, Place and Environment or Special Modifiersbased on any of the elementary categories. Generallycommon modifiers can modify a combination of twoor more category occurrences in a subject heading.There are also two other kinds of special modifiers.Independent Modifiers can modify the focus inde-pendently without depending on any other modifier.Dependent Modifiers cannot modify the focus di-rectly but can modify only another modifier of thefocus; for example, Temperature, High temperature,Very high temperature. When concepts are repre-sented by terms there may be cases wherein a term(composite term) may have to be broken down (fac-tored) into two or more constituent terms and eachone of them identified as belonging to one or theother of the elementary categories. In Kaiser’s system-atic indexing system agriculture is represented as acombination of land + cultivation corresponding tohis categories, concrete and process. A compositeterm is considered as a synonymous term to the com-bination of the factored constituent terms.

Syntax of the Subject Headings

The basic rule of syntax for formulating subject head-ings is DISCIPLINE (BASE) followed by ENTITY(CORE OBJECT) which is followed by PROPERTYand/or ACTION. PROPERTY and/or ACTIONmay be further followed by PROPERTY and/orACTION as the case may be, followed by COM-MON MODIFIERS. The SPECIES/TYPES and/orMODIFIERS and/or PARTS and/or CONSTITU-ENTS for each of the Elementary Categories followimmediately the manifestation to which they are re-


68

spectively SPECIES/TYPES or MODIFIERS orPARTS or CONSTITUENTS without the manifesta-tion of any other Elementary Category intervening.The rules of syntax give rise to a context-dependentsequence of the components in the subject heading.

Indicators of the Structure

Certain numeric codes have been prescribed to indi-cate the categories and their subdivisions to which in-dividual concepts belong. One set of these codes isshown below:

0 Form Modifier 9 Discipline/Base2 Time Modifier 8 Entity/Core Object3 Environment Modifier .2 Property4 Place Modifier .1 Action

SUBDIVISIONS.3 Constituent.4 Part.5 Modifier of Kind 1 including

Phase Relation Modifier.6 Species/Type, including those

created by Modifiers of Kind 2

In the subject headings the indicators precede thecomponents to which they are indicators. The indica-tors for property and action as well as the sub-divisions (species, part etc.) are attached with the indi-cators for the elementary categories to which they arerespectively property or action or sub-divisions.

Formulation of Subject Headings

Taking the title of the resource as the starting pointeach of the specific subjects dealt with in the resourceis expressed in natural language. Each of the specificsubjects (topics) may warrant a separate subject head-ing. Each of the component ideas corresponding toeach of the elementary categories that are implied isexplicitly stated to form expressive titles. Let one ofthe expressive titles be “In Leather technology, drysalt curing of pig skin in Thailand”. This is then ana-lyzed to identify the elementary categories and theirsub-divisions to which each of the components in theexpressive title belongs. All composite terms are fac-tored into their constituent terms and identified as be-longing to one of the elementary categories. Thecomponent terms are written down as a formalizedexpression following the rules of syntax, as given be-low:

(Discipline/Base) Leather Technology (Entity/Core) Pig skin (Action) Dry salt curing (Place Modi-fier) Thailand.

Each of the component terms in the subject head-ing is then analyzed to find out its super-ordinateterms. This is done by finding out “of which the con-cerned component is a species/ type, or part or con-stituent” in the context of the subject as a whole. Thisprocess is continued with each of the super-ordinatesrecognized in the process till it ends up with the con-cept of the elementary category. For this purposeterminological sources such as thesauri, glossaries anddictionaries are to be used. Each of the super-ordinatesthus recognized are fixed prior to the concerned termsuccessively giving rise to a ‘modulated’ subject head-ing as follows:

(D/B) Leather technology (E/C) Hide and skin(Part of E) Skin (Type of E) Pig skin (A) Beam-houseoperation (Sub-action) Curing (Type of A) Salt curing(Type of A) Dry salt curing (Common modifier)Thailand.

NOTE: The reason for making each of the super-ordinates precede the respective component terms isto endow the subject headings with the capacity toproduce an organizing sequence effect resembling thesequence of class numbers when sorted alphabetically,along with similarly formulated other subject head-ings. Moreover, this will facilitate searching using anyof the super-ordinate terms also.

Each of the component terms in the subject head-ing is replaced with standard terms and the synonymsare attached to the standard terms with an appropriatesymbol such as an equals sign (=). For this purpose,vocabulary control tools such as thesauri and classauri(Bhattacharyya,1982) (Devadason & Ramanujam,1982) (Devadason, 1985) are used. Appropriate indica-tors for the elementary categories, their sub-divisions,and common modifiers of different kinds are insertedin the appropriate places. The resulting subject head-ing is as follows:

Leather technology 8 Hide and skin 8.4 Skin 8.6Pig skin 8.1 Beam-house operation 8.1.4 Curing 8.1.6Salt curing 8.1.6 Dry salt curing 4 Thailand

Organizing Sequence in Alphabetical Arrangement

A set of subject headings formulated according to thissystem when sorted alphabetically resembles an or-ganizing (classified) sequence as illustrated below:

Subjects according to Colon Class numbers:


69

L"aN5 = Medicine, Bibliography, 1950sL"k1' N3 = Medicine, Dictionary.L"v1'N = Medicine, History, 19th centuryL"44'N = Medicine, History, India, 19th centuryL = MedicineL, 0;2 = Medicine, AnatomyL,0;3 = Medicine, PhysiologyL,0;4 = Medicine, DiseaseL,0;4:3 = Medicine, Disease, DiagnosisL,0;4:6 = Medicine, Disease, TreatmentL,9C = Medicine, ChildL,9C;4 = Medicine, Child, DiseaseL,9C;42 = Medicine, Child, Disease, Infectious

diseaseL,9C;42:6 = Medicine, Child, Disease, Infectious

disease, TreatmentL,9C,4;4 = Medicine, Child, Respiratory system,

DiseaseL,9C,45;4 = Medicine, Child, Respiratory system,

Lung, Disease

The same subjects with headings formulated accord-ing to the indexing system and sorted alphabetically:

Medicine 0 Bibliography 2 Nineteen fiftiesMedicine 0 DictionaryMedicine 0 History 2 Nineteenth centuryMedicine 0 History 4 India 2 Nineteenth centuryMedicine 1Medicine 8.1 PhysiologyMedicine 8.2 AnatomyMedicine 8.2 DiseaseMedicine 8.2 Disease 8.2.1 DiagnosisMedicine 8.2 Disease 8.2.1 TreatmentMedicine 8 ChildMedicine 8 Child 8.2 DiseaseMedicine 8 Child 8.2 Disease 8.2.6 Infectious diseaseMedicine 8 Child 8.2 Disease 8.2.6 Infectious disease8.2.1 TreatmentMedicine 8 Child 8.4 Respiratory system 8.2 Disease8.2.6 Infectious disease 8.2.1 TreatmentMedicine 8 Child 8.4 Respiratory system 8.4 Lung 8.2Disease 8.2.6 Infectious disease 8.2.1 Treatment

It may be seen that the two sequences are almost simi-lar. In other words, it is possible to bring an organiz-ing sequence effect to the alphabetic arrangement ofsubject headings. Once the sequence is obtained, theindicators may be suppressed in the display orchanged to suitable punctuation marks.

Access System for Web Resources

A prototype system for web resources using thePOPSI/Deep Structure Indexing System has beenbuilt which has the following functions:

System functionsThe system staff -- namely, the administrator, man-ager, cataloger and indexer -- and the system userswho access and search, can perform the followingfunctions:

System Staff:– Select resources from sites recommended by users;– Add new resources to the database;– Summarize the resource descriptions;– Revise the resource summaries or abstracts;– Index the resources;– Maintain subject headings;– Revise and store resource descriptions as per stan-

dard metadata and create inverted index files; and– Monitor and review, link check and maintain the

system.

System Users:– Browse the subject headings displayed in hierar-

chies;– Search the database with the use of search words;– Display the retrieved records in a subject hierarchic

order;– Recommend new sites; and– Access original resources using the link provided in

the retrieved records.

System componentsThe main components of the system are:– Staff maintenance module;– Automatic text summarization and metadata mod-

ule;– Cataloging and Indexing module;– Database and Index (DSIS) generation and mainte-

nance module;– Search and Browse module; and– System maintenance module.

Figure 1, the Opening screen of the system for theadministrators, shows the major functions of the sys-tem staff as given below:


70

Figure 1

The Staff Profile function allows the administrators toadd new staff (cataloguer, indexer) and to maintainthe staff profile and password authentication. The Se-lection from Recommendation allows the system staff

to select the sites recommended by the users, evaluatethem and add them to the system. The followingscreen (Figure 2) shows the list of recommended sitesand whether they are selected or not.

Figure 2: Recommended Sites Listing


71

Figure 3: Description of Site

Figure 3 shows the description of the site as recom-mended by the users showing as far as possible themeta data identified and supplied by the user.

The selected resource can be added to the system,and the record is displayed as per the Dublin Coredata elements with a summary of the text formedautomatically as shown below:

Figure 4: The Data Elements


72

Figure 5: Data Elements (Continued.)



73


Figure 8: Data Elements (Continued.) Faceted Subject Heading

The automatic text summarization module requiresthree values to be input :

a) minimum frequency for a word to be taken as sig-nificant;

b) minimum number of significant words to form acluster; and

c) maximum distance between two significant wordsin a cluster to extract the sentence to form thesummary.


74

Figure 9: Summarization of Text

The summarizer uses a list of stop words to removecommon words in the beginning of the text process-ing. The size of the summary may be modified bychanging the values supplied. It may be necessary toedit the summary to resolve any dangling sentences. It

is possible to add a new site directly by specifying theURL. The system automatically accesses the resource,pulls out the metadata (cataloging elements), fills upthe template, takes input for summarization and addsa summary.

Figure 10: Search Screen


75

Figures 7 & 8 above show the portion of the inputtemplate that allows the faceted subject heading to beinput manually. The Discipline/Base can be selectedfrom the list displayed in the selection window. TheCategories Property and Action are repeatable as arethe addition of species and parts. The input templateaccommodates all of these. Moreover, more than onefaceted subject heading could be assigned to a re-source. Once the data is entered the record can bebrowsed and edited and corrected. The desired recordcan be retrieved by entering some of the input dataand edited and modified. Once the records are enteredand admitted to the system, the system stores the rec-ords in the data base and prepares the required indexes

from the faceted subject heading for searching. Figure10 shows the search screen (see page 74). Once asearch term(s) is input the system searches for facetedsubject headings having the term(s) and displays themin sorted sequence as shown in Figure 11.

If the subject headings span more than one page,the searcher can go browse through the pages to selectthe subject heading that best describes his or herquery or input another term to be searched in combi-nation. The faceted subject headings containing boththe terms would be selected and displayed along withthe number of records that would be retrieved foreach. The records can be displayed by clicking thisnumber at the right hand side.

Figure 11: Retrieved subject headings in Organized Sequence

The searcher may browse through the pages to selectthe subject heading that best describes his or herquery, or if the retrieved subject headings span morethan one page he or she may input another additionalterm to be searched in combination. The faceted sub-ject heading containing both the terms would be se-

lected and displayed along with the number of recordsthat would be retrieved for each. The records can bedisplayed by clicking this number on the right handside. The record display will show the URL whichcan be clicked to access the resource on the Net.


76

Figure 12: Retrieved Record Display

It is possible to change the display to produce an organ-ized index of different levels to make browsing easy. Bybringing the required category (along with its subdivi-sions) to the beginning of the subject chain, the organiz-ing effect can be changed. The sequence can be variedaccording to the user. It can be presented as discipline-based if the user of the system is an academic. It can bechanged to entity- (object) based or action- (process)based. When resources are collected for indexing and tobuild search files, it may be possible to exercise controlover synonyms and standardize the terms used. It mayalso be possible to have a change from non-standardterms used by the searcher to the standard ones. Whenany term used by the searcher is poly-hierarchic, thenthe subject chains in which the term occurs (only thesuper-ordinates in the category to which the term be-longs) can be displayed, perhaps along with the termdenoting the discipline or base in reverse order, and theuser can be asked to select the appropriate chain repre-senting his/her query. It is possible to incorporatesynonyms due to semantic factoring so that the systemsearches for the combination of factored terms to bepresent in faceted subject headings. It is possible to sortthe headings and present higher level indexes for easybrowsing. It is also possible to resolve the meaning ofhomonyms by displaying the full faceted subject head-ings, having the term and asking the user to select thesubject heading that represents the intended meaning ofthe searcher.

Limitations and Labor Intensiveness

An indexing approach of this analytical type usingfacet analysis perhaps is best suited for scholarly aca-demic resources having educational and researchvalue. Moreover, there may be cases wherein it wouldbe difficult to recognize the categories in certain sub-jects in spite of illustrative examples. Provision ofsuch subject chains for a resource as a whole and forthe significant sub-units would be time consuming asvocabulary control tools and terminological sourcesof different types may have to be consulted and used.Normally it is expected to take about three months totrain an indexer. To begin with an indexer would beable to provide the faceted subject heading for five re-sources a day which may increase to ten per day in aweek’s time. The optimum would be about twentyresources per day after working for about twomonths.

Conclusion

As mentioned by Bella Hass Weinberg (1996) “it ishoped that the systems for organizing informationdeveloped earlier will not be ignored, that their designflaws will not be replicated, and that our increasingknowledge of human factors will be incorporated intosystems for indexing the Internet”.


77

Further work will include a function to suggest in-dex terms for consideration by the cataloguer whileformulating subject headings using automatic index-ing techniques as well as by searching for similar itemsin the database is in progress. Another module toform tables of contents of different hierarchic levelsfor quick browsing as well as creation of a classaurusand thesaurus from the facet analyzed subject head-ings is also under development.

References

Bhattacharyya, G. (1979). POPSI : Its fundamentalsand procedure based on a general theory of subjectindexing languages. Library Science with a Slant toDocumentation 16 (1), 1-34.

Bhattacharyya, G. (1982). Classaurus : Its fundamen-tals, design and use. In Proceedings of the 4th Inter-national Study Conference on Classification Re-search. Augsburg. June 28- July 2, 1982 (pp. 139-148). Frankfurt: Indeks Verlag.

Dahn, M.. (2000). Counting angels on a pinhead :Critically interpreting web size estimates. Online24(1), 40.

Devadason, F.J., & Kothanda R. M. (1982). Com-puter aided construction of ‘alphabetic’ classaurus.In Proceedings of the 4th International Study Confer-ence on Classification Research. Augsburg. June 28-July 2, 1982 (173-182). Frankfurt: Indeks Verlag,

Devadason, F.J. (1985). Online construction of alpha-betic classaurus : A vocabulary control and index-ing tool. Information Processing and Management.21(1), 11- 26.

Devadason, F.J. (1986). Computerized deep structureindexing system: FID/CR Report. Frankfurt: IndeksVerlag.

Ellis, D., & Vasconcelos, A. (1999). Ranganathan andthe Net: Using facet analysis to search and organ-ize the world wide web. Aslib Proceedings , 51(1),3- 10.

Fugmann, R. (1999). The empirical approach in theevaluation of information systems. Knowledge Or-ganization, 26 (1), 3.

Kaiser, J. (1911). Systematic indexing. In The CardSystem Series Vol. II. London: Pitman & Sons.

Kerr, L., & MacLeod, R. (1998). Subject-basedInformation Gateways (updated.10 Dec., 1998).

Retrieved 7 November 2002 from http://www.ariadne.ac.uk/issue18/subject-gateways/

McIlwaine, I.C., & Williamson, N.J. (1999). Interna-tional trends in subject analysis research. Knowl-edge Organization. 26 (1), 27.

McKiernan, G. (1997, February). Hand-made in Iowa:Organizing the web along the Lincoln highway.D Lib Magazine. Retrieved 7 November 2002from http://www.dlib.org/dlib/february97/02mckiernan.html

Mitchel, S., & Mooney, M. (1996). INFOMINE : Amodel web-based academic virtual library. Infor-mation Technology and Libraries, 15, (1). .

Retrieved 7 November 2002 from <http://lib-www.ucr.edu/pubs/italmine.html>

Mundie, D. A. (1999). Organizing computer re-sources: Or, how I learned to stop worrying andlove the DDC. Retrieved May 12, 1999 fromhttp://ivory.lm.com/~mundie/CyberDewey/organizing_computers.html

Nua Internet Surveys (2001). Search engines cause ireamong Net users. Retrieved March 1, 2001 fromhttp://www.nua.com/surveys/

Ranganathan, S.R. (1964). Subject heading and facetanalysis. Journal of Documentation. 20(3) ,109 –119.

Tinker, A. J., Pollitt, A.S., O’Brien, A., & Brakevelt,P.A. (1999). The Dewey Decimal Classificationand the transition from physical to electronicknowledge organization. Knowledge Organization.26(2), 80 – 96.

Weinberg, B. H. (1996). Complexity in indexing sys-tems – abandonment and failure : Implications fororganizing the Internet. ASIS 1996 Annual Confer-ence. 19 – 24 Oct, 1996. Retrieved November 7,2002 from <http://www.asis.org/annual-96/ElectronicProceedings/weinberg.html>

Williamson, N.J. (1997). Knowledge structures andthe Internet. In: Knowledge Organization for In-formation Retrieval : Proceedings of the 6th Interna-tional Study Conference on Classification Research,University College London, 16-18 June 1997 (pp. 23-27). The Hague, Netherlands: International Fed-eration for Information and Documentation.

Knowl. Org. 29(2002)No.2 M. Žumer, G.J.A. Riesthuis: Consequences of Implenting FRBR

78

Consequences of Implementing FRBR:Are we Ready to Open Pandora’s Box?

Maja Žumer*, Gerhard J.A. Riesthuis**

*Department of Library and Information Science and Book Studies,University of Ljubljana, Slovenia (e-mail: [email protected])**Department of Information Science, University of Amsterdam,

The Netherlands (e-mail: [email protected])

Maja Žumer is an Assistant Professor in the Department of Library and Information Science and BookStudies, University of Ljubljana. She holds a PhD from the University of Zagreb and an MLS from KentState University. Her main research interests include information retrieval, design of user interfaces, and,more recently, implementation of the Functional Requirements for Bibliographic Records (FRBR) modelin the design of bibliographic databases.

Gerhard J.A. Riesthuis is a lecturer in the Department of Information Science at the University of Amster-dam (The Netherlands). He holds a Master’s Degree in Sociology from the University of Utrecht and aDoctorate from the Faculty of Humanities at the University of Amsterdam. He teaches courses in cata-loguing and indexing, and in information technology. His areas of research include subject access systemsand cataloguing theory. He is the Secretary of ISKO and the Literature Editor of Knowledge Organization.

Maja Žumer, Gerhard J.A. Riesthuis. (2002). Consequences of Implementing FRBR: Are We Readyto Open Pandora’s Box? Knowledge Organization, 29(2). 78-86. 24 refs.

ABSTRACT: The study Functional Requirements for Bibliographic Records (FRBR) was commissionedby IFLA and published in 1998. It defines the core functions of a catalogue (and bibliographic records)as a gateway to information. For that purpose an abstract entity-relationship model of a catalogue is

proposed. The FRBR model is revolutionary. The (computer) catalogue is not seen as a sequence of bibliographic records and areplica of the traditional card catalogue, but rather as a network of connected entities, enabling the user to perform seamlessly allthe necessary functions. So far there has been some theoretical discussion of the model and some limited experiments, but thereis a lack of research in how to implement this theoretical model in practice, in new-generation catalogues. In this paper some re-actions to the model are analysed. The main focus is on consequences of the model for the OPAC interface design, particularlythe searching functionality and display of results.

1. FRBR: re-examination of cataloguing

1.1. Background

In 1961, as a result of fundamental re-examination ofcataloguing theory and practice on an internationallevel, the so-called Paris Principles (Statement of Prin-ciples, 1971) were agreed upon. The second importantstep was the development of ISBDs, which started in

1971 with the International Standard Bibliographicfor Monographic Publications (ISBD (M)). Thus thefoundation for new and revised national and interna-tional cataloguing rules was established.

The last three decades of the last century alsobrought unprecedented changes to both the ways li-braries operated and to their users’ needs and expecta-tions. Library automation, development of large bib-liographic databases, union catalogue systems and

Knowl. Org. 29(2002)No.2M. Žumer, G.J.A. Riesthuis: Consequences of Implenting FRBR

79

shared cataloguing, and new forms of publishingwere some of the revolutionary developments. At thesame time, libraries were faced with the need to re-duce the high costs of operation, particularly for cata-loguing.

IFLA sponsored the Seminar on Bibliographic Re-cords in Stockholm in 1990 to address these issues.The Seminar acknowledged the need for libraries toreduce the cost of cataloguing, and identified a focusfor further research: meeting user needs associatedwith the use of various types of materials and thebroad range of eventual new requirements for biblio-graphic records. One of the resolutions of the Semi-nar was therefore to commission a study “to definethe functional requirements for bibliographic rec-ords”. The terms of reference for the study stated asits purpose and scope “to delineate in clearly definedterms the functions performed by the bibliographicrecord with respect to various media, various applica-tions, and various user needs. The study was to coverthe full range of functions for the bibliographic recordin its widest sense – that is, a record that encompassesnot only descriptive elements, but access points(name, title, subject, etc.), other ‘organizing’ elements(classifications, etc.), and annotations.” (FRBR, 1998,p. 2)

In 1992 the Standing Committee of the IFLA Sec-tion on Cataloguing accepted the terms of referenceand appointed a study group. The final report of thestudy was accepted at the IFLA Conference in Co-penhagen in 1997 and published the following year(FRBR, 1998, p.2-3).

The study group described as its aim “to produce aframework that would provide a clear, preciselystated, and commonly shared understanding of whatit is that the bibliographic record aims to provide in-formation about, and what it is that we expect the re-cord to achieve in terms of answering user needs.”(FRBR, 1998, p.2). The study had two objectives.“The first is to provide a clearly defined, structuredframework for relating the data that are recorded inbibliographic records to the needs of the users ofthose records. The second objective is to recommenda basic level of functionality for records created by na-tional bibliographic agencies.” (FRBR, 1998, p.7)

1.2. The catalogue described with an entity-relationshipmodel

Entity-relationship methodology was used for theFRBR model. This means that a set of objects of in-terest or entities is defined, and the relations between

the entities are listed. Further the important charac-teristics or attributes of each entity are identified.

There are three groups of entities. Group 1 entities(with no common name, but for which the authors ofthis article propose the term ‘bibliographical entities’)include work, expression, manifestation, and item.These entities represent the information traditionallyreflected in the formal cataloguing part of biblio-graphic records. Group 2 entities (‘name entities’)comprise persons and corporate bodies responsible forthe intellectual or artistic content, the physical pro-duction and dissemination or the custodianship ofbibliographic entities. Group 3 entities (‘subject enti-ties’) represent the subject of works and include con-cept, object, event, and place. Also the entities of thefirst and second group can be the subjects of works.

Relationships serve as a link between entities andenable the user to navigate within the bibliographicdatabase (catalogue) and beyond. Relationships canlink entities of different groups, different entities ofthe same group, and instances of entities of the sametype. There is, for example, a relationship between awork and all the expressions derived from it (entitiesof the same group), or a relationship between theauthor (name) and a work as examples of entities ofdifferent groups. The relationship between all expres-sions of one work (e.g. translations) is a relationshipof entities of the same type.

The study identifies four generic tasks that the us-ers perform while searching: “to find entities that cor-respond to the user’s stated search criteria (i.e., to lo-cate either a single entity or a set of entities in a file ordatabase as the result of a search using an attribute orrelationship of the entity); to identify an entity (i.e.,to confirm that the entity described corresponds tothe entity sought, or to distinguish between two ormore entities with similar characteristics); to select anentity that is appropriate to the user’s stated searchcriteria (i.e., to choose an entity that meets the user’srequirement with respect to content, physical format,etc., or to reject an entity as being inappropriate tothe user’s needs); and to obtain access to the entity de-scribed (i.e., to acquire an entity through purchase,loan, etc., or to access an entity electronically throughan online connection to a remote computer).” (FRBR,1998, p.82).

The Associazione Italiana Biblioteche (AIB, 1999)and Elaine Svenonius (2000) expressed the need for“navigation” as an additional user task. In our opin-ion, navigation is not a function of a catalogue but anecessary facility of an OPAC to fulfill the functionsof the catalogue; not a function or goal on its own.


80

Admittedly, navigation is only possible when all thenecessary data and the relations between entities arepresent in a catalogue: for example, two expressionsof the same work cannot be recognized as such unlessthey have the same title or explicit information on thework they are derived from.

1.3. Revolution or evolution: the two parts of the FRBR

As said before, the study had two objectives: toprovide a clearly defined, structured frameworkfor relating the data that are recorded in biblio-graphic records to the needs of the users of thoserecords, and to recommend a basic level of func-tionality for records created by national biblio-graphic agencies. The first objective is discussed inthe chapters 2 – 6 of the FRBR and the second ob-jective is dealt with in chapter 7.

The two parts are in many respects quite different.In the first part, discussing the structural framework,the FRBR model is revolutionary. The (computer)catalogue is not seen as a sequence of bibliographic re-cords, ordered according to strict rules, or as a replicaof the traditional card catalogue in a computer, butrather as a network of connected entities, enabling theuser to perform seamlessly all the functions of thecatalogue mentioned in the FRBR: “having for thefirst time considered a totally electronic catalogue, or-ganized according to a network structure, and notonly re-proposing in electronic form the manual cata-logue, which on the contrary is organized in a linearmanner”. (Weston, 2000). In principle, values of allcharacteristics (attributes) of all entities are available asaccess points, without the traditional restriction tothree values of each attribute for an individual de-scription. Chapter 6 contains four tables, which dis-cuss the importance of the characteristics (attributes)and relations of each of the bibliographic entities, forthe four tasks mentioned. The importance is given asthree grades: high, moderate and low. For example,the title of a work is said to have high value to find awork, as has the relation dependent component. In-tended audience has a low value for this task, and formof work a moderate value. The relations between awork and persons/corporations responsible for thework are of high value for the find task. For the selec-tion task, both the title of the work and the form ofthe work have a high value. The language of the ex-pression has a moderate value for the find task but ahigh one for the identify and select tasks.

The model discussed in the first part of the FRBRsuggests a work oriented approach, and that means

that the relations between bibliographic and authorityrecords should be re-evaluated (Eversberg, 1998). Pat-rick Le Boeuf (2001b) states:

No more ‘rule of three’? No more ‘one book inhand, one record’ principle? No more ‘authority rec-ords’ as such (for uniform titles at least)? Catalogingcodes, ISBDs and MARCs thrown into question?Those are not ‘minor changes’, that is an earthquake!The entire landscape cataloguers were used to is col-lapsing. Actually, to be honest, FRBR does not explic-itly call for such a revolution; but such a revolutionlogically ensues from FRBR. One might call FRBR a‘quiet revolution’ – or a time bomb. (p. 18)

The second part of the FRBR study is different. Itgives the basic level of functionality for national bib-liographies, and contains nine tables, which specifythe data elements and relationships that should be in-cluded as a minimum in the bibliographic records of anational bibliography. The list of these data elementsis more or less equal to the traditional canon of mostcataloguing rules, and contains only the characteristicsconsidered of high value in the lists of chapter 6 forone of the four tasks of the catalogue. Even with thisrestriction there are some inconsistencies (e.g., in therole of title(s)). The discussion of this issue is beyondthe scope of this paper and should be addressed sepa-rately (see Byrum and Madison, 2000, p.26-28 and 45-47).

2. Reactions to FRBR

The FRBR study prompted immediate response fromlibrarians. Two events should be mentioned in par-ticular. The ELAG (European Library AutomationGroup) held four consecutive workshops: “ELAGOO-oriented bibliographic model” at the ELAGseminar in The Hague, The Netherlands (March 25-27, 1998), “IFLA model for bibliographic records” atthe ELAG seminar (April 21-23, 1999 in Bled, Slove-nia) followed by another workshop titled “FRBR:time to act for ELAG” in 2000 (April 12-14, 2000 inParis) and “What benefits do we expect from anFRBR-based automated catalogue?” (June 6-8, 2001 inPragu, Czech Republic). The other event was the“Seminario FRBR” organized by the Cataloguing Sec-tion and the Tuscany Section of the Associazione Ital-iana Biblioteche in Florence, January 27-28, 2000(Seminario FRBR, 2000). This was the first conferencethat was exclusively devoted to FRBR.

ELAG discussions started immediately after thestudy was published. They dealt with a detailed in-


81

vestigation of the model, judgement of its possiblebenefits and identification of future research and de-velopment. Among the conclusions, the most impor-tant ones were that a lot would have to change (e.g.,cataloguing rules will have to be reviewed) and thatthe model would have to be verified in practice. Thediscussion in 1999 focused on detailed investigation ofentities, particularly those of the first group. The con-clusion was that there are in reality seven levels, be-cause parts of works, expressions and manifestationsshould be introduced as separate intermediate levels.In addition, there was a recommendation for an addi-tional highest level, a so-called “top hat”, describingthe original work from which other works are de-rived (Grinnen, 1999; Holm, 1999). The same conceptwas advocated by Martha Yee (2000); she introducedsix hierarchical levels (superwork, work, version, edi-tion, near-equivalent, copy). The recommendation ofELAG 1999 was withdrawn in 2000, with the com-ment that the same goal can be achieved introducing ahorizontal relationship (“related to” or “derivedfrom”) for works. The discussion focused on compari-son between two models: FRBR and ICOM/CIDOC( Crofts, 2001), the object-oriented model originatingfrom the museum community. The conclusion wasthat although there are some methodological differ-ences between the models, stemming particularlyfrom differences in scope (libraries and museums),there are enough similarities that a future merger ofmodels is feasible. The discussion in 2001 focused onimplementation of the model in (future) libraryautomation systems. Because of the complexity of themodel, more prototyping was proposed, particularlyin the area of displays and linking. An importantpoint was also the description of two independentprojects (Danish and Norwegian/Finnish) with thegoal of creating automatically FRBR records from ex-isting MARC records. The Norwegian/Finnish proj-ect is described by Knut Hegna and Eeva Murtomaa(2002).

The discussion of “Seminario FRBR” in Florence fo-cused first on the model itself, stressing that the ex-pression level may have to be subdivided further, be-cause expressions can be derived from previous ex-pressions, or, alternatively, can differ from the origi-nal expression by using only a different medium. An-other conclusion was that, especially in the beginning,the implementation of the FRBR model would resultin an increase of work. The participants of the Semi-nario also suggested that ‘navigation’ should be addedto the four functions of the catalogue. As mentioned

already, navigation is in our view, a necessary facilityof an OPAC, not of a catalogue.

The early reactions, while mentioning some problemsand shortcomings of the model, are in general quitefavourable, emphasizing that it is a good basis for rela-tional database design and, above all, a new concep-tual view of cataloguing.

In her book The Intellectual Foundation of InformationOrganization, Svenonius (2000) discusses in a chaptercalled Bibliographic Objectives, the objectives formu-lated in the FRBR. Her conclusion is that:

The IFLA statement is both timely and relevant inits generalization to embrace nonbook materials andinformation agencies other than libraries, in its mod-ernization of terminology, and in its resolution ofambiguity. However, another change it makes issomewhat problematic – the collapsing of traditionalfinding and collocating objectives. The traditionalfinding objective specifies that what is to be found is aparticular known document, while the traditional col-locating objective specifies that what is to be found is aset of documents, defined by criteria such as author,work, and subject. (page 17, italics from Svenonius)

Documents are for Svenonius what are manifesta-tions in the FRBR. She makes two alterations to thefirst objective of the FRBR. She replaces entities bydocuments – that is, one of the entities defined by theFRBR, manifestations – and she redefines the generalfinding function of the FRBR: she defines finding assearching for a particular document and collocation assearching for all documents that share common at-tribute values.

The find-function as defined in the FRBR com-prises both the find and the collocation-function ofcard catalogues: “to find entities that correspond tothe user’s stated search criteria (i.e., to locate either asingle entity or a set of entities in a file or database asthe result of a search using an attribute or relationshipof the entity)” (FRBR, 1998, p. 82).

In the context of the FRBR collocation is not men-tioned at all. As seen from the perspective of cata-logues as databases, collocation is only an auxiliaryfind function of card catalogues. Because there couldnot be enough access points in a card catalogue, re-lated records were filed together to overcome thelimitations. This mechanism enabled users to locateone bibliographic record using any of the accesspoints provided, and then browse through recordsfiled next to it to locate related records. The computer


82

catalogue is a database with no intrinsic order of rec-ords, thus without collocation. If enough accesspoints are available and if searching on several criteriasimultaneously is supported, users can always retrievethe relevant records.

On the other hand, the order of records is veryimportant when displaying the results of a query. Ameaningful order of resulting records enables the userto browse effectively through the list to evaluate therelevance of the result, and also to identify, select, andobtain. To summarise: in the card catalogue colloca-tion is provided when the catalogue is created, whilein a computer catalogue ordering (i.e., collocation) isperformed after the query.

Finding a given manifestation of a given work thatexists in many manifestations representing many ex-pressions was – and is – a real problem in a printedbibliography or a card catalogue. In these systems themany descriptions of the many manifestations have tobe ordered in a way that is understandable for the us-ers, otherwise the user is lost. Uniform titles, workheadings and also references are needed to make itpossible to get such a usable order in these systemsand thus to make searching possible. In a card cata-logue, without order, there can be no searching. Anonline bibliographic file has, however, no intrinsicorder and does not need it. It is an inventory with in-dexes, as recognized and regretted by Michael Carpen-ter(2000).

It would already be a big improvement if the findfunction as defined in Chapter 7 of the FRBR wereimplemented in full in our catalogues. This meansamong others that the “rule of three” has to be aban-doned and that when a manifestation contains morethan one work all the works have to be made accessi-ble. In the description of their work on the transfor-mation of MARC records to FRBR Hegna and Mur-tomaa (2002) regret the present lack of consistent useof cataloguing rules (e.g., in original title, language, re-lator codes, recording of separate works in a singlemanifestation, etc.).

3 FRBR and the users of the catalogue

3.1. FRBR and the builders of the catalogue

Interestingly, the vast majority of further work basedon FRBR focused on testing the model’s adequacy fordescribing various types and kinds of library materi-als. That approach was probably additionally encour-aged by the fact that the second part of the FRBRstudy is devoted to the list of basic data required for

records created by national bibliographic agencies.The list practically corresponds with current practiceand can be mapped perfectly into UNIMARC struc-ture, thus giving the impression that the model hasnot changed anything. But is that the case?

At the Lubetzky Symposium (Future, 2000) theFRBR were mentioned three times, but only more orless in passing. This is remarkable since one of thetopics of this Symposium was “Current research incataloging”. In the papers devoted to this topic theFRBR are not mentioned at all. Yet some findings inthese papers are relevant for the topics discussed inthe FRBR.

Sara Shatford Layne (2000) found that six principalpatterns might be desirable for access to art works. Indescending order of frequency they are:

1. people attributes, including name as an attribute(one should be able e.g., to search for ‘twentiethcentury Russian painters’);

2. kind of work, subdivided or qualified by date orplace;

3. subject or related literary work;4. titles (meaning creator plus title) of specific works;5. style;6. technique or material.

In this context, another paper of the Lubetzky Sym-posium is relevant. Carpenter (2000) recognizes that”What we have with self-standing bibliographic de-scriptions plus ’access points’ is essentially a registerwith an index”. And further “The register-index cata-log is not one that lends itself to arrangement of theentries in a way that readily fulfils the second func-tion of the catalog. It does, however, admirably fulfilthe first objective”. By first and second function aremeant the two functions mentioned under 2.1 and 2.2of the so-called Paris principles1.

He ends his paper with some remarks about thecatalogue of tomorrow. It should show all works ofan author in a usable arrangement when a user looksup a particular author. When searching with a title, alleditions of the work should be shown if there is onlyone work with that title; if there are more works withthe title sought, the user should choose which workshe or she wants and should be shown all editions.“We will then be working with a catalog that, as aminimum, is faithful to the Paris Principles.” (Car-penter, 2000) As said before, collocation by means ofadded entries and appropriate filing was a necessarytool only for card catalogues, because of lack of searchpossibilities. Giving the results of a search in a given


83

order is not necessary for the find function, but im-portant for the identify and select function.

As long as the original title of translations is storedin the inventory and indexed, all translations can befound by searching on author and/or original title.The only alternative is keyword searching in com-plete records. The disadvantage is that the precisionwith keyword searching is much smaller. For workslike Bibles, searching with combinations of subjectdata, languages, title words and the like can be used tofind manifestations of given expressions.

We question whether the very expensive provisionof uniform titles and so forth, just for display func-tions in an OPAC is worth the money. Who willsearch with the name Goethe, Wolfgang and then wanta nicely ordered list of all the manifestations of all theexpressions of all the works available in Die DeutscheBibliothek? In a paper Allyson Carlyle (2001) showsthat end-users often would prefer another order thanthe one provided by the traditional uniform headingof author plus uniform heading approach.

3.2. FRBR and the new generation catalogues

There has been very little discussion of how new cata-logues, designed according to FRBR, will look andwhat impact that will have on end-users and the waythey use library catalogues. In one example (Noerr etal., 1999), which derives from the 1999 ELAG discus-sions, authors list possible advantages to users: morestandardization and control of various indexes, clus-tering and intuitive relationships among entities, en-hanced navigation, and media integration.

It should not come as a surprise that librarians whowere trained in traditional cataloguing find the FRBRmodel difficult; there are problems with terminologyand definition of entities. As Kirsten Strunck (1999)reports, even students “find it unnecessarily compli-cated to operate with the abstract entities of themodel as you cannot study these entities per se. Theyfind the definition of the entities academic and airy.”

Patrick Le Boeuf (2001a and 2001b) explains thisattitude with the fact that present catalogues still relyon Paris Principles and are in fact a replica of a cardcatalogue, a model that is not appropriate any more inthe automated context. In an overview of the work ofthe FRANAR (Functional Requirements And Num-bering for Authority Records; this project is a follow-up of FRBR) Françoise Bourdon (2001) states that themembers – all librarians – of the working group havedifficulties with the entity-relation model that under-lies the FRANAR and also the FRBR.

If librarians and future librarians find the FRBRmodel difficult, we question whether end-users willnot find the model even more difficult. End-users areaccustomed to present-day catalogues (with all theirshortcomings, admittedly) and will find new entitiesand relations even stranger. The FRBR catalogue willbe a very different tool.

The FRBR model is endorsed by IFLA. Goal 3 ofthe Strategic Plan 2001-2003 of the Section on Cata-loguing is “to promote the Functional Requirementsfor Bibliographic records (FRBR) study and its rec-ommendations, and take follow-up action to developnew descriptive standards for access points and to de-velop a new approach to the bibliographic universe”(IFLA Section on Cataloguing, 2002).

Therefore it was to be expected that developers andvendors of library automation software would see itas a necessity to incorporate FRBR into future sys-tems. That has proven to be true: several vendorshave already prepared prototypes and even more areplanning to begin the development in the near future.

Therefore, there is no way back. Present cataloguesare not easy to use, as Christine Borgman states in hertwo famous papers (1986 and 1996): ”online catalogsare difficult to use because their design does not in-corporate sufficient understanding of searching behav-ior. …(we should) lay to rest the card catalog designmodel for online catalogs.” End-users will be using thenew catalogues and we have to make sure that thesecatalogues will fulfil their mission: to “respond moreeffectively to an increasingly broad range of user ex-pectations and needs” (FRBR, 1998, p. 1).

4. Proposals for OPAC-design: consequences forthe interface design

4.1. Goals of OPAC design

For the end-user the FRBR model has to be imple-mented in a transparent way, without an explicit useof terminology and/or concept. The end-users shouldnot be faced with the problems of intricate differencesbetween expressions and manifestations, problems ofhierarchy of expressions, and the boundaries of awork.

The FRBR model actually offers more accesspoints and, above all, the relationships really enablethe most important facility of an OPAC: the naviga-tion within the catalogue and the whole bibliographicuniverse.

One of the important characteristics of the FRBRmodel is that the traditional division between biblio-


84

graphic and authority records has disappeared; it hasbeen replaced by a network of interrelated entitieswith relationships serving as links between entities ofdifferent groups (e.g., between bibliographic entitiesand names, reflecting authorship), between differentlevels of bibliographic entities within one group (e.g.,between works and expressions). It will be interestingto see how the FRANAR project will further elabo-rate on these issues.

New OPACs will have to be designed. There willbe important differences in searching capabilities andin the way bibliographic information is presented onthe screen. But it must be noted that the formal im-plementation of the model itself does not change any-thing unless cataloguers change their practice funda-mentally -- particularly by implementing all relation-ships. Navigation cannot be improved if relationsamong entities are not recorded and if retrieval of bib-liographic records is not seen as the major goal of cata-loguing. The present state of catalogues can be attrib-uted to the fact that current cataloguing rules still re-flect the technology of the card catalogue and are wellsuited for that purpose. Even MARC formats havekept the card characteristics. Especially in the begin-ning, the purpose of MARC records was to make thedistribution of catalogue cards more effective (Hegnaet al., 2002, p. 35). Too much information in theMARC records is suited for display only and cannotbe searched effectively.

4.2. Searching functionality

OPACs should enable end-users to retrieve records,containing any attribute value, part of attribute valueor combination of attribute values. This can beachieved (and is achieved in present catalogues) bysearching on a string of characters within a particularaccess point, a combination of access points, or any-where in the record. While not advocating keywordsearching as a panacea, it is often the only means ofaccessing information that has been entered in un-structured fields such as notes. If the FRBR model isimplemented in full, the need for keyword searchingwill be greatly reduced.

It should be possible to navigate to related infor-mation directly from the displayed (bibliographic orauthority) records. Therefore a mechanism for linkingrelated information within the catalogue and beyondhas to be used; most probably that will be in the formof hyperlinks, a mechanism that has already proven aseffective and intuitive and has as a result been ac-cepted as a de facto standard. While hypertext links

are a broadly used and accepted technique for linkingattribute values, there should also be a mechanism forlinking entities as a whole (e.g., linking a work withall its related works, or all its expressions). The usershould get the information that related entities existand have a simple way of requesting a list of these en-tities; probably by clicking on a button.

Although we assume that graphical interfaces willbe used due to their user-friendliness and wide accep-tance, some special needs of users have to be takeninto account: visually impaired users and users withvery slow and/or unreliable computer network con-nections. Both groups will still need a text-orientedinterface; for the visually impaired that can enabletranscription into Braille or speech synthesis, and thequantity of data transmitted over the network is dras-tically reduced.

The designers of future systems will therefore haveto include a text-oriented interface as an option with-out sacrificing much of the functionality or flexibil-ity.

4.3. OPAC displays

As we have already stated, the FRBR model is concep-tually difficult – particularly because the databasemodel behind it is unfamiliar to most cataloguers andend-users – and end-users should not be exposed to itin its complexity. The data (results of queries) shouldtherefore be displayed in a transparent and intuitiveway, while keeping all the functionality of FRBR.

One of the important requests therefore, is thatonly bibliographic entities of the same level shouldappear in any list of hits. It can be assumed for exam-ple, that a query on an author’s name would result inone or more works, all expressions of these works, allmanifestations of those, and all items (copies) the li-brary owns. If all these records are displayed in a sin-gle list (even if it is ordered meaningfully), the userwill be confronted with too long a list of seeminglyidentical data. The situation can be avoided if onlyone level of entities is displayed in one list: onlyworks, only expressions, only manifestations, or onlyitems. The user should then be able to display thenext level of entities for the selected record(s).

It is intuitively easier to proceed from a higherlevel to a lower level, therefore the first list should bethe display of the highest appropriate level of entitiesregarding the search statement. The level which isdisplayed first is determined by the attributes and/orrelationships used in the search statement. If only oneattribute/relationship is searched, then this attrib-


85

ute/relationship determines the level (as defined inFRBR: it is the highest level with this attrib-ute/relationship). The relationship “is author of” isapplicable to all bibliographic entities, but work is thehighest; expression is the highest level for languageattribute; manifestation for year of publication andany copy-specific details of course for item, for exam-ple, call-number.

The decision becomes more complex when thesearch statement contains more (i.e., a combinationof) attributes/relationships. The level of entitiesshould then be the highest level that is common to allattributes/relationships concerned. For example,when searching on a combination of publication yearand language, the manifestation level should be dis-played, because the publication year only exists atthat level.

When the attribute cannot be determined (e.g.,when searching on keywords), the lowest biblio-graphic entity level in the catalogue has to be dis-played. That would be manifestation level in mostcases and item level where copy specific details in-cluded in bibliographic records cause the finding ofthe records.

5. Conclusions: do we open the Pandora’s box?

It is easier to put the question than to answer it. It de-pends a lot on which part of the FRBR one looks at,and also which reactions one takes into account.

If one looks at the second part of the FRBR and atthe ISBDs proposed and decided upon since the publi-cation of the FRBR, then the answer is simply a No;there is no need for it. Everything should remain asmuch as possible as it has been since 1876, the yearCutter published his list of functions of the catalogue,or preferably since 1961, the year of the Paris Princi-ples; that hallowed separation of subject indexing andformal cataloguing. To say it differently -- the OPACshould be a mirror of the card catalogue it replaced.

If we look at FRBR as the new conceptual model,the answer is YES. We should start exploring theways in which we could improve our catalogues.There is a lot of pressure from the vendors of libraryautomation systems and information retrieval sys-tems, who are aware of all the shortcomings of pres-ent catalogues and bibliographic databases as well asfrom informed end-users, who have been exposed tosimple, seemingly more user-friendly ways of search-ing the electronic universe.

We may even have no choice. New catalogues willbe developed. Librarians have to start seriously ex-

ploring the implementation issues of FRBR, as well asre-thinking the foundations of cataloguing in view ofnew technological possibilities. We also have to plancarefully for the conversion, migration and/or inte-gration of existing catalogue records into the new-generation catalogues. Some research on that has beenmentioned previously.

The vast body of traditional expertise in knowl-edge organisation should be used in the process. Thismay be the perfect opportunity to re-establish librari-ans as essential experts in the information chain of thefuture.

Notes

1.The catalogue should be an efficient instrument forascertaining

2.1 whether the library contains a particular bookspecified bya. its author and title orb. if the author is not named in the book, its

title alone, orc. if author and title are inappropriate or in-

sufficient for identification, a suitable sub-stitute for the title; and

2.2 (a) which works by a particular author and(b) which editions of a particular work are in

the library (Verona, 1961).

References

AIB. (1999). Functional Requirements for BibliographicRecords Seminar (Florence, January 27-28, 2000).Rome: Associazione Italiana Biblioteche. Re-trieved June 11, 2002, from http://www.aib.it/aib/sezioni/toscana/conf/cfrbr.htm

Borgman, C.L. (1986). Why are online catalogs hardto use? Lessons learned from Information-Retrieval Studies. Journal of the American Societyfor Information Science, 37 (6), 387-400.

Borgman, C.L. (1996). Why are online catalogs stillhard to use? Journal of the American Society for In-formation Science, 47 (7), 493-503.

Bourdon, F. (2001). Functional Requirements andNumbering of Authority Records (FRANAR): towhat extent authority control can be supported bytechnical means. Paper for the 67th IFLA Counciland General Conference August 16-25, 2001 (Paper096-152a-E). Retrieved June 11, 2002 from http://www.ifla.org/IV/ifla67/papers/096-152ae.pdf

Byrum, J.D., & Madison, O.M.A. (2000). Reflectionson the goals, concepts and recommendations of


86

the IFLA Study on Functional requirements forBibliographic Records. In M. Guerrini (ed.) Semi-nar FRBR Functional Requirements for Biblio-graphic Records Seminar (Florence, January 27-28,2000) (pp. 11-51). Rome: Associazione ItalianaBiblioteche. Retrieved June 11, 2002, fromhttp://www.aib.it/aib/sezioni/toscana/conf/frbr/byrmadis.htm

Carlyle, A. (2001). Developing organized informationdisplays for voluminous works: a study of userclustering behaviour. Information Processing andManagement, 37 (5), 677-699.

Carpenter, M. (2000). Main and added entries. In TheFuture of Cataloging: Insights from the LubetzkySymposium April 18, 1998, University of CaliforniaLos Angeles, (pp. 60-71). Chicago: ALA.

Crofts, N. (2001). Introduction to theICOM/CIDOC conceptual reference model. InArchives, Libraries and Museums Convergence, 24th

ELAG seminar (Paris 2000), (pp. 168-177). Paris:Cité des Sciences et de l’Industrie.

Eversberg, B. (1998). The part ! whole relationshipin German and American cataloguing data: resultsand suggestions. In REUSE+ : joint project ofOCLC and Niedersächsische Staats und Univer-sitätsbibliothek Göttingen [online]. Göttingen:Niedersächsische Staats und Universitätsbiblio-thek Göttingen, 1998. Retrieved June 11, 2002,from http://www.biblio.tu-bs.de/allegro/formate/reusep.htm

FRBR. (1998). Functional Requirements for Biblio-graphical Records: Final report / IFLA study groupon the functional requirements for bibliographicalrecords ; Approved by the Standing committee ofthe IFLA Section on Cataloguing. München :Saur. Retrieved June 11, 2002, fromhttp://www.ifla.org/VII/s13/frbr/frbr.htm

The Future of Cataloging: Insights from the LubetzkySymposium, April 18, 1998, University of CaliforniaLos Angeles (2000). Chicago: ALA.

Grinnen, E.B. et al. (1999). IFLA model for biblio-graphic records ELAG OO edition II. Discussionpaper In Managing Multimedia Collections, 23rd

ELAG seminar (Bled, 1999), (pp. 110-119) Ljubl-jana: National and University Library.

Hega, K., & Mutomaa E. (2002). Data mining Marc tofind: FRBR. 68th IFLA Council and General Con-ference, Glasgow, August 18-24, 2002. Nr053-133-E.Retrieved June 11, 2002, from http://www.ifla.org/IV/ifla68/papers/053-133e.pdf .

Holm, L. (1999). IFLA model for bibliographic rec-ords: ELAG OO edition II. In: Managing Multi-media Collections, 23rd ELAG seminar (Bled, 1999,(pp. 120-124) Ljubljana: National and UniversityLibrary.

Layne, S.S. (2000). Modelling relevance in art history.In The Future of Cataloging: Insights from the Lu-betzky Symposium, April 18, 1998 University ofCalifornia Los Angeles, (pp. 33-41). Chicago: ALA.

Le Boeuf, P. (2001a). The impact of the FRBR modelon the future revisions of the ISBDs: a challengefor the IFLA section on cataloguing. Paper for the67th IFLA Council and General Conference August16-25, 2001. Nr 095-1521a-E. Retrieved June 11,2002, from http://www.ifla.org/IV/ifla67/papers/095-152ae.pdf

Le Boeuf, P. (2001b). FRBR and further. Catalogingand Classification Quarterly, 32(4). 15-52

Noerr, P., Goossens, P., & Matei, D. (1999). Userbenefits from a new bibliographic model: follow-up of the IFLA Functional Requirements study.International Cataloguing and Bibliographic Con-trol. 28( 3). 80-81

Petrucciani, A. (2000). New requirements for newcatalogues. In M. Guerrini (ed.) Seminar FRBRFunctional Requirements for Bibliographic RecordsSeminar (Florence, January 27-28, 2000), (pp. 138-146). Rome: Associazione Italiana Biblioteche.

Strunck, K. (1999). About the use of ‘Functional Re-quirements for Bibliographic Records’ in teachingcataloguing. IFLA 1999 Bangkok Conference Pro-ceedings (Code Number 1089-131-E) Retrieved June11, 2002, from http://www.ifla.org/IV/ifla65/papers/108-131e.htm

Svenonius, E. (2000). The Intellectual Foundation of In-formation Organization. Cambridge: The MITPress.

Verona, E. (1961). Statement of Principles, annotateded. with commentary and examples by Eva Verona.International Conference on Cataloguing Principles,1961, Paris. 1971. London: IFLA Committee onCataloguing.

Weston, P. G. (2000). FRBR and the user: considera-tions on research. In M. Guerrini (ed.) SeminarFRBR Functional Requirements for BibliographicRecords Seminar (Florence, January 27-28, 2000).Rome: Associazione Italiana Biblioteche.

Yee, M. (2000). Lubetzky’s work principle. In: TheFuture of Cataloging: Insights from the LubetzkySymposium April 18, 1998, University of Califor-nia Los Angeles, (pp.72-104). Chicago: ALA.

Knowl. Org. 29(2002)No.2U. Miller, R. Teitelbaum: Pre-Coordination and Post-Coordination

87

Pre-Coordination and Post-Coordination:Past and Future

Uri Miller*, Ruth Teitelbaum**

*Library of the Wingate Institute for Physical Education and Sports,Netanya, Israel (E-mail: [email protected])

**Social Sciences Information Center, The Henrietta Szold Institute,Jerusalem, Israel (E-Mail: [email protected])

Dr. Uri Miller is the Senior Indexer at the Library of the Wingate Institute for Physical Education andSports, Netanya, Israel, and the Thesauri Construction Coordinator at the MOFET Institute – Research,Curriculum & Program Development for Teacher Educators, Tel-Aviv, Israel.

Mrs. Ruth Teitelbaum is the Director of the Social Sciences Information Center at the Henrietta Szold In-stitute in Jerusalem and teacher at the School of Library, Archive & Information Studies, The HebrewUniversity of Jerusalem.

Uri Miller, Ruth Teitelbaum. (2002). Pre-Coordination and Post-Coordination: Past and FutureKnowledge Organization, 29(2). 87-93. 31 refs.

ABSTRACT: This article deals with the meaningful processing of information in relation to two sys-tems of information processing: pre-coordination and post-coordination. The different approaches arediscussed, with emphasis on the need for a controlled vocabulary in information retrieval. Assigned in-dexing, which employs a controlled vocabulary, is described in detail. Types of indexing language canbe divided into two broad groups – those using pre-coordinated terms and those depending on post-coordination. They represent two different basic approaches in processing and information retrieval.The historical development of these two approaches is described, as well as the two tools that apply tothese approaches: thesauri and subject headings.

Introduction

It is only natural that librarianship and informationscience, similar to any other theoretical foundation ofa widespread and important sphere of practical activi-ties, are highly sensitive to the kind of sharp, rapid hi-tech progress and ‘technological revolution’ that con-tinues to transform our professional environment.Computers and the Internet continue to cause theo-retical and practical turmoil even in experts’ minds.We professionals find ourselves caught up in a race ofsorts, ever hurrying and driven on by the technologi-cal onslaught (Miller, 2001).

While these conditions are capable of generatingmany new decisions and practical ‘breaks’, they leavetoo little time to comprehend the theoretical founda-tions of continuing changes, or, to use military par-lance, of ‘fortifying the rear’. In addition, there are anumber of theoretical principles that, from the outset,appear to be clear and unshakable, that is, axiomatic.The concepts of ‘pre-coordination’ and ‘post-coordi-nation’, and their interconnections are among these.A striking demonstration of the slump of interest inthis problem is evident from a search of these terms inthe Library and Information Science Abstracts database.Undoubtedly, the high level of their elaboration inthe already classical works of Dagobert Soergel (1974),

Knowl. Org. 29(2002)No.2 U. Miller, R. Teitelbaum: Pre-Coordination and Post-Coordination

88

Antony Charles Foskett (1996), Frederick WilfridLancaster (1986) and others is a very important reasonfor this established situation. (For interesting histori-cal reviews see: Cleveland & Cleveland, 1990, pp. 60-62; Kilgour, 1997, pp. 340-348).

But nevertheless, now and again, the return to re-analysis of some axiomatic propositions can be veryrelevant. The need to revisit our foundations is evi-dent from publications that have distorted the realmeaning of these concepts. These distortions are rep-resented by two examples written over a decade apart:articles by Ann Schabas (1982) and David Bodoff andAjit Kambil (1998). Schabas, without ceremony orexplanation, tested the Library of Congress SubjectHeadings (a pre-coordinated structure) using post-coordinated searching.

Although the recently published investigation byBodoff and Kambil proposes several new searchmethods suited to the modern information environ-ment, it nevertheless reveals a certain vagueness of in-terpretation of some of the basic conceptions underly-ing subject indexing and searching (both pre- andpost-coordinated). First, their definition of “subjectqueries which return zero hits” as “subject search fail-ures” (Bodoff & Kambil, 1998a p.1254) is only admis-sible in the case of the fixed presence of relevantdocuments in the information massive, otherwise,“the lack of information is also information”. Second,the authors compare pre-coordinated indexing andsearching on the ground of subject headings (specifi-cally, Library of Congress Subject Headings) to post-coordinated indexing and searching using full-text freeindexing and searching, where every meaningful wordis counted as a keyword. Such contraposition cannotbe considered appropriate in the comparison of pre-coordination and post-coordination. Rather, Bodoffand Kambil are comparing the presence or absence ofvocabulary control. Such comparison can be relevantonly to vocabulary-controlled pre-coordinated index-ing and searching (based on subject headings) as op-posed to vocabulary-controlled post-coordinated in-dexing, and thesaurus-based searching. Ignorance ofthesauri as the basic tool of vocabulary-controlledpost-coordinated indexing and searching is not onlyvery strange, it also weakens efforts at comparison,due to the inherent impossibility of a different basisfor comparison. This methodological confusion al-lows the authors to reach such questionable and un-proven conclusions as the “lower precision” of post-coordination versus pre-coordination and “lack of atheory for selecting post-coordinated keywords” (Bo-doff & Kambil, 1998a, p. 1257), the unfounded attack

on “narrow keyword terms” (Bodoff & Kambil,1998a, p. 1257-1258), and so forth.

The widespread evidence of such misapprehensionswas vividly illustrated by Catherine Robinson andJanet Knight (1997) in the work especially producedfor the elimination of all misunderstandings – the in-ternational standard for thesaurus construction (“In-ternational Standard ISO-2788. Documentation--Guidelines for the Establishment and Development ofMonolingual Thesauri, second edition -- 1986-11-15”)and the related discussions.

But perhaps this is useful in prompting us to re-examine our theoretical ‘cornerstones’. Without goinginto details regarding the modern theoretical andmethodological principles of thesaurus constructionwhere we can find clear answers to the questionsraised (see Miller 1997), we can establish the urgentneed for ‘stock-taking’ of some basic theoretical prin-ciples of indexing. In the following discussion, we willreview lexical control in the form of pre-coordinate(subject headings) and post-coordinate (thesauri) vo-cabularies, their application and their philosophicaland historical foundations, revealing certain conclu-sions about contemporary indexing and searching.

The Need for Lexical Control

Once more, the key role played by the retrieval processbrings the discussion of the role of lexical control in in-formation processing back on the agenda. Again thisraises the ‘old’ question of the use of free text versuscontrolled language (lexical control) in information re-trieval. While there are many advantages to searchingusing free text, mostly in terms of availability and rapidinformation retrieval, the main disadvantages lie in theamount of ‘noise’ (irrelevant information) accompany-ing information searching, low precision in search out-comes and a high degree of information loss (low re-call). This is the consequence of the lack of control andlack of uniformity inherent in natural language. A freetext search means having to cope with a flood of notalways relevant information (for a detailed analysis ofthe central problems of information retrieval, see forinstance Blair, 1990, chapter 4). It is perfectly reason-able to search using free text when searching for newterms or concepts which have not yet been expressed incontrolled language. A free text search provides a veryhelpful complement in information extraction, butcannot be the sole search tool (Peters & Kurth, 1991).In some instances, the combined use of a free and a con-trolled search will produce the best results (Dubois,1987; Fidel, 1987, 1991, 1992; Rowley, 1990).


89

The use of free text when searching the Internet issubject to the same constraints. In this case, technol-ogy does not entirely solve the ‘problem’ for us.When searching the Internet we confront a vast arrayof often irrelevant information because Internet“keywords” are the same elements of free text search-ing arranged according to the frequency of their oc-currence, but not according to all the classical key-words of lexically controlled databases. Neither com-puters nor the Internet can replace the human mindin processing information meaningfully. Quality dataretrieval is impossible without purposeful data proc-essing, particularly since the range of data increaseswith geometric progression (Weinberg 1995).

Subject Indexing

One form of meaningful data processing is subject in-dexing. Subject indexing is part of the system knownas ‘assigned indexing’. Here, a computer program as-sists the information scientist in processing the data.The program supports the routine work of the in-dexer with a thesaurus facility, online indexing andindexing quality control. It can look up terms in thethesaurus, proof read and compile indexes.

On the whole, subject indexing uses a controlledindexing language. Indexing languages have two ele-ments: a vocabulary and a syntactical structure. Index-ing language, in contrast to natural language, is artifi-cial and is constructed with information processing inmind. The terms used in an artificial language relate toconcepts as opposed to words. Lexical control providescontrol of synonyms, homographs, alternative spell-ings and obsolete terminology. Lexical control meansthat all material related to a concept will be listed forthat concept, along with pertinent references to otherterms.

The syntactical structure consists of rules for ter-minology construction and for determining the con-nection between terms. Indexing languages differ intheir syntactical structures which become the basis oftheir designation as either pre-coordination or post-coordination. With pre-coordination, the connection(or coordination) between index terms occurs as theindexer processes the data. With post-coordination,the connection between the index terms occurs duringdata retrieval. Both approaches employ coordinationbetween the terms, but they differ in timing.

Pre-coordination and post-coordination are twofundamentally different data processing and retrievalphilosophies. When pre-coordination is used for in-dexing, the indexer determines the connection be-

tween the index terms in advance. These connectionsare fixed and rigid. The best examples of this methodare subject headings and book subject indexes, whichhave their own reasons for such methodical ap-proaches (but this is a subject for a separate analysis).With post-coordination, connections between indexterms are not decided in advance. The connections areonly constructed during the data retrieval process.The thesaurus works on this principle.

The pre-coordination approach used by the indexerduring the indexing process only partially resemblesthe post-coordination indexing process. The stagescommon to both methods are the essential itemanalysis and subject selection stages that occur duringindexing. An indexer using the pre-coordinationmethod will use a syntactical structure that requiresthe connection between terms to be decided in ad-vance. The combination order of facets to be prede-termined constructs a subject heading that is analo-gous to a multi-link chain. When using post-coordination, the indexer identifies the appropriatefacets, but does not decide on the connection betweenthe index terms. An item will be assigned index terms,while each term retains its independence. Thus, theterms assigned to a particular document are uncon-nected.

With pre-coordination, there is no clear separationbetween data processing and its retrieval – they aretwo stages within a single process. The indexer decidesthe connections between terms while indexing andcreates the subject heading strings. The subject head-ing string represents the outcome of the search. Inother words, the indexer knows the outcome of thesearch since the outcome is predetermined. When in-formation is extracted using this method, a group of‘ready’ subject heading strings is obtained, after whichthe most appropriate one is selected. This is carriedout by browsing through the subject headings which,as we have already noted, are prepared in advance. Butsuch an approach holds the danger of a vocabularyexplosion because of the enormous number of subjectheading strings required.

With post-coordination, indexing and retrieval aretwo separate processes, although there is a close rela-tionship between them. In post-coordinate indexing,the indexer tries to express all of the relevant subjectscontained in the indexed item according to the facetsdetermined by the indexing policy. The indexer can-not ‘predict’ the search results since he or she does notmake the connection between the terms. Each termstands alone. Each user will construct his or her ownsearch strategy via which the terms are linked, based


90

on the user’s requirements. The indexer using thepost-coordination method is required only to carryout the objective indexing. During indexing, the in-dexer must focus on the indexing process that will in-fluence the retrieval results.

Following this distinction, we can divide databasesinto the ‘traditional’, based on the pre-coordinated in-dexing and retrieval principle (subject headings), and‘non-traditional’ databases based on the post-coordi-nated indexing and retrieval principle (thesauri).

Pre-coordinate terms or complex terms?

Subject headings resemble a chain with multiple links.During pre-coordinate indexing, the indexer connectsthe various links, that is, he or she expresses pre-coordination via the syntactic structure. Each link inthe chain represents a ‘subject’, for example: “Chil-dren and Television”. This subject heading has twoelements: “Children” and “Television”, but is definedas a single ‘subject’, with two separate distinguishableconcepts. The subject heading can have pre-coordi-nation in its lexicon as well; for example: “Physicaleducation for children – Curricula – France”. In thisheading, we find three different ‘subjects’ (links in thechain), and four different ‘concepts’. The four con-cepts are: physical education, children, France, curric-ula.

In contrast, the thesaurus contains descriptors asopposed to subject headings. Descriptors refer to con-cepts. They may be comprised of one or severalwords, but will always relate to a single concept. Anindex term comprising more than one word andwhich denotes a concept, or ‘compound term’ (Miller,1996, 1997; Teitelbaum, 1994), may not be brokendown into individual words without there being anessential change of its meanings. Such a break canproduce a vast amount of irrelevant information inthe process of retrieval, for example: ‘high schools’,‘latch-key children’, ‘short working week’, ‘soap op-era’ and ‘venetian blinds’.

The term ‘agricultural schools’ is not the equiva-lent of the sum of ‘schools’ and ‘agriculture’. Simi-larly, the combination of the terms ‘group’ and ‘dis-cussion’ gives us two different terms ‘group discus-sion’ and ‘discussion group’ (Batty, 1989; Milstead,1984). ‘Children and Television’ is not a compoundterm. It is a pre-coordinated term made up of twoseparable elements. The thesaurus has no place for en-tries of this kind. The descriptors appear individually,‘children’ will be found on its own, as will ‘televi-sion’, and each entry will be accompanied by its own

set of references to concepts. The searcher connectsthem during data retrieval.

It is important to bear the following in mind: agiven composite of words may be considered a com-pound term according to the scope of one thesaurusand a pre-coordinated term according to another. Forexample, ‘sports psychology’ is a compound term insport or psychological thesauri, but it is also a pre-coordinated term in thesauri on more general or dif-ferent spheres. The conceptual framework in whichthe term appears is sometimes the decisive factor andthe type of interpretation system (pre-coordinated orpost-coordinated) in which such a term functions isnot important. Information retrieval requirements,which among other things include a high level of pre-cision and which prevent search noise, will determinethe use of compound terms in all information re-trieval systems (Batty, 1989; Milstead, 1984, Soergel,1974).

The Development of Pre-coordination and Post-coordination

Pre-coordination and post-coordination are two fun-damentally different data processing and retrieval phi-losophies which were influenced by different generalphilosophical doctrines. Pre-coordination and post-coordination emerged at different times and representthe moods and fashions then influencing society.When indexing employing subject headings appearedon the American library scene, pragmatism was theprevailing philosophical doctrine in American society(Miller, 1980). The subject heading doctrine positsthat the information consumer must trust the in-dexer, because it is the indexer who determines theconnections in advance, and thereby determines theway that information retrieval occurs.

In contrast, post-coordination emerged during therise of the ‘free information’ movement. This point ofview allowed the consumer a free hand in choosing theinformation in which he or she was interested and incarrying out combinations of subjects not linked in ad-vance. Here, the indexer’s role is to facilitate rapid ac-cess to information and ensure its availability to theconsumer. The indexer has no control over data re-trieval, and all is open before the user. The history ofthese information processing and retrieval approaches islinked to the different tools employed by each method.

Examples of the tools used for pre-coordination aredictionary catalogues and subject catalogues. The bestexample of a pre-coordination indexing language isthe Library of Congress Subject Headings (LCSH).


91

Subject headings applied the principles developed byCharles Ammi Cutter who, at the end of the 19th cen-tury, laid down the rules for constructing dictionarycatalogs, and began to address subject analysis. It is in-teresting that pre-coordination as a method has acomplex character because it was used not only insubject approach, but also in its antipode, a classifica-tion approach (as in Cutter’s and Dewey’s classifica-tion systems). LCSH was distributed to and adoptedby a large number of libraries throughout the world(see, for instance: Heiner-Freiling, 2000). Its impacthas been reinforced by two projects: MARC and theLibrary of Congress cataloging-in-publication data.These projects used subject headings since they werecrucial to the data processing conducted.

With the advent of technological advances, whichpresented an opportunity for change, subject analysiswas transformed into a control-based doctrine. At thispoint, the crisis worsened since the Library of Con-gress management was not prepared for a ‘technologi-cal revolution’ and had difficulty in adapting subjectheadings to the new developments. The well-knownresearcher Hans Wellisch compared computerized da-tabases that use subject headings to a jet plane riggedwith a nineteenth century steam engine. The planewill indeed move, and might even fly, but it will beaccident prone, unreliable and its body design will bewasted due to the low speed it moves at (Wellisch,1972). The Library of Congress Subject Headingshave been met with widespread criticism, as have thelatest attempts to change the subject headings andmake their structure a quasi-thesaurus (Conway, 1993,Dykstra, 1988; Gerhan, 1989; Nuckolls, 1994).

Currently, post-coordination is in the librarianshipand information science spotlight. Post-coordinationis the outcome of a dialectic process. Although post-coordination was an outgrowth of pre-coordination,it conflicts with its principles and represents an an-tagonistic approach. Post-coordination emerged withthe advent of library computerization and uses com-puters for information processing and retrieval. Nowwe see that even the natural adherents of pre-coordination recognize new realities. Thomas Mannemphasized at the “Bicentennial Conference on Bib-liographic Control for the New Millennium: Con-fronting the Challenges of Networked Resources andthe Web” held at the Library of Congress on Novem-ber 15-17, 2000: “Neither I nor anyone else is arguingfor precoordination rather than postcoordination. Weneed both browse displays of precoordinated stringsand the possibility of postcoordinate combinations ofindividual elements” (Mann, 2000).

Pre-coordinate information processing and re-trieval may be compared to a ‘craft’, post-coordinateprocessing to a ‘science’ and retrieval using this ap-proach to ‘an art’. This analogy should not be under-stood as an attempt to detract from the importanceand prestige of pre-coordination; each activity needsto be considered in light of its compatibility with li-brary needs or with the indexed collection. Pre-coordination can be used to provide a high quality,professional standard of indexing. The above analogyrelates to the fact that post-coordination informationprocessing activities require associative thinking onthe indexer’s part, allowing not always visible connec-tions between terms to be exposed. The data retrieverusing this approach also requires the necessary skillsfor performing the ‘art’ of information retrieval.

Subject headings are constructed pragmatically be-cause the subjects are connected to the material con-tained in a particular library: new subject headings areincorporated only after material relating to a specificconcept appears in the library. The thesaurus has nosuch close link with any concrete database. The pres-ence of real concepts is the main reason for building athesaurus. The existence of real material is a secon-dary issue. Building a thesaurus is like building amodel, which is not closely connected with the exis-tence of real material in the database. The thesaurus isa lexical-semantic model of a conceptual reality.

Conclusions

Several conclusions emerge from the discussion onlexical control and developments in the area of subjectindexing:

a. Post-coordination is the most sophisticated methodof processing data by subject, since it does not im-pose decisions made by the indexer on the user,and allows flexible data retrieval consistent withdiverse and changing data requirements. Post-coordination works efficiently with the employ-ment of syntactical devices that help to increaseprecision. The quality of the thesaurus as a lexicalcontrol tool is affected by numerous factors,among these being: correct choice of terms, build-ing terms according to accepted rules and a consis-tent, precise method of working. These rules needthesauri with a high degree of order and structure,so that indexers can search for appropriate termswithout much effort.

b. At this stage, there is no substitute for lexical con-trol, which offers the most accurate means of data


92

processing and retrieval. It is worthwhile investingin developing lexical control mechanisms. Obtain-ing a reasonable amount of highly accurate infor-mation justifies the expenditure.

References

Batty, D. (1989). Thesaurus Construction and Main-tenance: a Survival Kit. Database, 12.13- 20.

Blair, D.C. (1990). Language and Representation in In-formation Retrieval. Amsterdam: Elsevier SciencePublishers.

Bodoff, D., & Kambil, A. (1998a). Partial coordina-tion I. The best of pre-coordination and post-coordination. Journal of the American Society forInformation Science, 49 (14). 1254-1269.

Bodoff, D., & Kambil, A. (1998b). Partial coordina-tion II. A preliminary evaluation and failureanalysis. Journal of the American Society for Infor-mation Science, 49 (14). 1270-1282.

Cleveland, D.B., & Cleveland, A.D. (1990). Introduc-tion to Indexing and Abstracting. 2nd ed.Englewood, CO: Libraries Unlimited.

Conway, M.O. (1993). Characteristics of SubjectHeadings in the Library of Congress BOOKSMDatabase. Library Resources & Technical Services,37. 47-58.

Dubois, C.P.R. (1987). Free Text vs. Controlled Vo-cabulary: a Reassessment. Online Review, 11. 243-253.

Dykstra, M. (1988). LC Subject Headings Disguised asa Thesaurus. Library Journal, 113. 42-46.

Fidel, R. (1987). Controlled Vocabulary and Free-Text Searching: Searchers’ Selection of SearchKeys. In C. Chen (Ed.), Proceedings of the 50thASIS Annual Meeting (pp. 71-73). Medford, NJ: In-formation Today.

Fidel, R. (1991). Searchers’ Selection of Search Keys 2:Controlled Vocabulary or Free-Text Searching.Journal of the American Society for Information Sci-ence, 42. 501-514.

Fidel, R. (1992). Who Needs Controlled Vocabulary?Special Libraries, 83. 1-9.

Foskett, A.C. (1996). The Subject Approach to Informa-tion 5th ed. London: Library Association.

Gerhan, D.R. (1989). OCSH in vivo: Subject Search-ing Performance and Strategy in the OPAC Era.Journal of Academic Librarianship, 15. 83-89.

Heiner-Freiling, M. (2000). Survey on Subject Head-ing Languages used in National Libraries and Bib-liographies. Cataloging and Classification Quar-terly, 29 (1/2), 189-198.

Kilgour, F.G. (1997). Origins of coordinate searching.Journal of the American Society for Information Sci-ence, 48. 340-348.

Lancaster, F. W. (1986). Vocabulary Control for In-formation Retrieval 2nd ed. Arlington, VA.: Infor-mation Resources Press.

Mann, T. Is Precoordination Unnecessary in LCSH? AreWeb Sites more Important to Catalog than Books? AReference Librarian’s Thoughts on the Future of Bib-liographic control. Available in: http://lcweb.loc.gov/catdir/bibcontrol/mann_paper.html

Miller, U. (1980). Methodological Challenges of SubjectAnalysis – A Comparative Analysis of the Principlesof Constructing a Subject Heading Language in theUS and USSR. Moscow: Nauka. (In Russian).

Miller, U. (1996). The Sport Database: Some Com-ments. Online & CD-ROM Review, 19. 67-74.

Miller, U. (1997). Thesaurus Construction: Problemsand Their Roots. Information Processing and Man-agement, 33. 481-493.

Miller, U. (2001) Thesaurus construction and new in-formation environment. In A. Kent (Ed.), Ency-clopedia of Library and Information Science. vol. 68,suppl. 31, (pp. 319-345). NY: Marcel Dekker.

Milstead, J.R. (1984). Subject Access Systems: Alterna-tives in Design. Orlando, FL Academic Press.

Nuckolls, K.A. (1994). Subject Access to DiversityMaterials: The Library of Congress Subject Head-ing Shortfall. Reference Librarian, 45-46. 241-251.

Peters, T.A. & Kurth, M. (1991) Controlled and un-controlled vocabulary subject searching in an aca-demic library online catalog. Information Technol-ogy and Libraries, 10. 201-211.

Robinson, C., & Knight, J. (1997). ContemporaryRecordkeeping: The Records Management The-saurus – Response. (Article in response to paperdelivered at the 1997 Records Management Asso-ciation of Australia’s National Convention). (Elec-tronic article). Available in: http://www.records.nsw.gov.au/publicsector/rk/aaa/response.htm.

Rowley, J.E. (1990). A comparison between free lan-guage and controlled language indexing and search-ing. Information Services and Use, 10. 147-155.

Schabas, A.H. (1982). Postcoordinate retrieval: acomparison of two indexing languages. Journal ofthe American Society for Information Science, 33 (1).32-37.

Soergel, D. (1974). Indexing Languages & Thesauri:Construction & Maintenance. Los Angeles, CA.:Melville Publishing.


93

Teitelbaum, R. (1994). The thesaurus in informationretrieval systems: Structure and components. YadLaKoreh, 27. 48-55. (In Hebrew)

Weinberg, B.H. (1995). Why postcoordination failsthe searcher. The Indexer, 19. 155-159.

Wellisch, H. (1972). Subject retrieval in the Seventies,methods, problems, prospects. In H. Wellisch, &T.D. Wilson (Eds.), Subject Retrieval in the Seven-ties: New Directions ( pp. 14-15). Westport, CT:Greenwood Press.

Knowl. Org. 29(2002)No.2KO Reports

94

KO Reports

Challenges in Knowledge Representation and Organizationfor the 21st Century: Integration of Knowledge Across Boundaries

Seventh ISKO International ConferenceGranada, Spain – July 10-13, 2002

Nancy Williamson

Participants at the Seventh International ISKO Con-ference, Granada, Spain, July 10-13, 2002 spent a veryfull and busy four days. They listened to an impres-sive range of papers covering much ground and mov-ing from general fundamental theories through manyindividual research projects demonstrating the appli-cation of new and old methods of organizing knowl-edge, and stretching from facet analysis to artificial in-telligence in the struggle to cope with the tremendousflood of information pouring towards us through theInternet. The theme of the Conference was “Chal-lenges in Knowledge Representation and Organiza-tion for the 21st Century: Integration of KnowledgeAcross Boundaries.” Papers were presented in sixteensessions over the four days. The published proceed-ings contains 80 papers, a few of which were not pre-sented because the authors were unable to attend theConference.

In her keynote address entitled “Conceptual Universalsin Knowledge Organization and Representation,” Re-becca Green drew our attention to the importance ofuniversal concepts in Knowledge Organization andRepresentation, stressing that it is the concept, ratherthan the terminology, that remains constant across lan-guages and cultures. On the assumption that whereuniversals exist knowledge integration is most likely tooccur, her paper explored “a representative inventoryof semantic and lexical universals” that should be ac-counted for in any information system. Indeed, she setthe scene for the subsequent presentations, touching onthe problems of vocabularies both actual and artificial,(i.e., controlled vocabularies, including classifications),on the need for a clear analytical approach and the pos-

sible potential for switching languages. Green also high-lighted the challenges presented by cultural differences.All of these themes were revisited throughout the con-ference. In conclusion Green stated that problems ofcomprehension and retrieval associated with lack ofphysical structure on the web have caused informationprofessionals to return to traditional principles, but shesaw this as only a partial answer. “The development ofeffective information systems hinges on understandingsocial and cognitive environments as well as knowledgeof organizational structures and technological capabili-ties.”

The opening session included four papers on “Theo-retical Models and Universals in Knowledge Organiza-tion and Representation,” all dealing with aspects ofcognition. Jack Andersen addressed the problem of“Ascribing Cognitive Authority to Scholarly Docu-ments: On the (Possible) Role of Knowledge Organiza-tion in Scholarly Communication.” His concern wasthe examination and discussion of how, and to what ex-tent, knowledge organization as an epistemic instru-ment in scholarly communication, can contribute tocognitive authority of scholarly documents. In her pa-per on “Augmenting Human Capabilities: Classifica-tion as Cognitive Scaffolding,” Elin Jacob sought “toextend the notion of the classification scheme as a cul-turally-transmitted knowledge tool by emphasizing thecognitive value of the schemes internal patterns of rela-tionship.” In doing so, she considered classificatorystructures in terms of their constraints, selections andexpectations. In this same session, Clare Beghtol exam-ined “Universal Concepts, Cultural Warrant and Cul-tural Hospitality.” Her analysis started from the prem-


95

ise that “regardless of linguistic and other domainboundaries or cultural traditions” access to informationcan be examined on the basis of how cultural universalsare implemented in specific cultures at specific timesand places.” “Time,” and its implementation in calen-dars, was used to illustrate how this concept is treatedin systems of knowledge organization. “Hospitality”was interpreted broadly and a “theoretical principle ofcultural hospitality” was derived to provide the theo-retical framework for decisions about types of accessthat could be “globally useful and ethically balanced.”Time was also a factor in the final paper in this group.In “Subject Ontogeny: Subject Access through Timeand the Dimensionality of Classification,” Joe Tennisbased his presentation on the understanding thatknowledge changes, making constant revision of classi-fication a necessity. He sees this as the opportunity to“add dimensionality to classification” and a chance touse classification schemes to explain conceptual war-rant, to explain the shift from disciplinary to multidis-ciplinary knowledge, and to serve as a tool for domainanalysis. Some applications of subject ontogeny wereproposed, including metadata preservation models, on-line access tools and interoperability frameworks. Theauthor sees subject ontogeny as an addition to, or anenhancement of, existing classification systems – actingas an interpretive layer to serve the needs of informa-tion workers.

The second session of the conference focused on “Epis-temological Foundations for Knowledge Structures andAnalysis.” A paper by Nuno Silva and João Rocha enti-tled “Merging Ontologies Using a Bottom-Up Lexicaland Structural Approach” took the view that classicalintegration approaches do not meet current require-ments and sought to propose new strategies. Theauthors suggest an “alignment process” as opposed tothe merging of data sources. This process, applied at theontological level, is described in six phases – normaliza-tion, similarity measuring, bridging, representation,transformation and negotiation. “Ontological Analysisof Literary Works of Art” by Gillola Negrini and Pa-trizia Zozi focused on the way that ontological struc-tures can aid the understanding and modelling of worksof art. The starting point was an analysis of the authors’own work on the Thesaurus of Italian Literature, thatused categories derived from the work of IngetrautDahlberg and some of its significant aspects were high-lighted. Following from this, the theories of NicolaiHartman and Roman Ingarden were examined in asearch for commonalities between the Dahlberg ontol-ogy and the two theories, hoping for some easy solu-

tions. Jarmo Saarti, in his paper on “The Analysis ofthe Information Process of Fiction: a Holistic Ap-proach to Information Processing,” challenged the con-cept of “aboutness” in subject catalogues. It comparedthe work of the author with previous studies of the in-formation process by such authors as Clare Beghtol,Annelise Pejterson and Jutta Austin. Findings suggestedthat users vary in the search elements that they use, es-tablishing the need to include documents in their total-ity in the information process. Systems have to bemulti-faceted to meet all the needs of various users. Thefinal paper in the group by NairYumiko Kobashi,Johanna W. Smit and M. de Fátima G.M. Tálamo onthe “Constitution of the Scientific Domain of Informa-tion Science” attempted to delimit the terminology ofinformation science on the assumption that the “cen-tral” terms of the discipline should form the core of thediscipline and the peripheral terms should be derivedfrom knowledge areas with which information scienceinterfaces. The Information Science discipline was de-scribed as one that is “in development” and does not yethave clearly defined boundaries or interfaces. Theauthors state that the terminology borrows from anumber of domains including logic, administration,linguistics, computer science, sociology, communica-tions, and cognitive science, as well as librarianship.Procedures were set up for detecting terms. Based onthe results, it was concluded that central terms come inlarge measure from librarianship and constitute thefield of “Information Organization and Representa-tion,” while terms from other areas showed little adap-tation. In the majority of cases the terms were adoptedas is, but “their original conceptual structure” wasabandoned. Further, it was determined that “the inter-disciplinary character of Information Science is clearlyupheld in the majority of discourse … from the 1990s.”Information science is described as a social science,rarely presenting itself as a science. While the authorsfeel that it is not a full-fledged science, it appears to as-pire to intervene “practically in society through opti-mization of information flow.”

A major segment of the conference focused on “Modelsand Methods” for knowledge organization and re-trieval. Perhaps this should not be surprising since, afterall, this is what the conference was about. Five sessionswere devoted to this general topic, each dealing with adifferent aspect – representation, tools and systems, re-trieval, the electronic environment (i.e., the Internet),conceptual relationships and information systems. Inthe session on “representation,” five papers were pre-sented, each dealing with an unusual situation. Anita


96

Coleman discussed “A Classification of Models” them-selves. As stated by the author, our systems of knowl-edge organization do not provide a unified location forworks about “models.” This paper presented a frame-work for such a class of one category of models (i.e.,scientific models) using facet analysis. Gain Piero Zarriaddressed the “Indexing and Querying of NarrativeDocuments, a Knowledge Representation Approach”in a paper that described a system called NarrativeKnowledge Representation Language (NKRL). Thesystem takes into account narrative multimedia docu-ments (e.g., news stories, corporate documents, legaltexts, and representations of patients’ medical records)and uses knowledge representation principles and somehigh-level inference tools. Jeremy Shapiro’s “Interdisci-plinary Knowledge Integration and Intellectual Creativ-ity” explored and analyzed the “use of culturallyprominent metaphors, symbols, archetypes, myths andnarrative patterns as metadata, as a method of facilitat-ing the discovery and retrieval of information and theintegration of knowledge across both disciplinary andcultural boundaries.” A Universal Cultural SymbolThesaurus is described as a potential subject languagefor a lexicon of metaphors and symbols that could beused to classify information objects. Theoretical back-ground is given, the metaphorical and symbolic meta-data are explained and the thesaurus described. The fi-nal paper of the session, on “Images and Words” byCatalina Naumis Peňa, focused on a new resource forindexing images. The differences between traditionalindexing and digital audiovisual information were ex-plored and the use of the thesaurus as a tool for imageindexing re-evaluated. The author proposes ErwinPanofsky’s principles as the basis for a new thesaurusfor image indexing.

In a “Models and Methods” session on “Tools andSystems,” five papers were presented with particularemphasis on the need for tools and systems to handledigital resources. A presentation by Maria Inés Cor-deiro and Aida Slavic examined “Data Models forKnowledge Organization Tools” and considered theneed for such tools to be “fully disclosed and avail-able” in the open network environment. The placeand value of traditional library and information sci-ence tools – classification schemes, thesauri andauthority files – and their interoperability was ex-plored. Some of these tools are already available; oth-ers need to be adapted and reshaped. In contrast,Vanda Broughton, in a paper entitled “Facet Analyti-cal Theory as a Basis for a Knowledge OrganizationTool in a Special Subject Portal” described a project

being carried out at University College London(UCL), which focused on the way classificationschemes using facet analysis principles can be appliedto the organization of digital resources. The UCL sys-tem builds on the classificatory principles developedby the Classification Research Group and derives itsmethodology from the Bliss Bibliographic Classifica-tion. The nature of classificatory structures, the appli-cation of faceted classification and the facet analyticalmethodology are explained. Facet analysis was the ba-sis recommended by many at the conference, thoughnot everyone put the same interpretation on thismuch-used and frequently misunderstood term.Broughton gave a clear definition in her paper as fol-lows: “It is taken to mean that rigorous process ofterminological analysis, whereby the vocabulary of agiven subject is organized into facets and arrays, re-sulting in a complex knowledge structure with bothsemantic and syntactic relationships clearly deline-ated.” In this same session, Stella Dexter Clarke ad-dressed the problem of “Planning Controlled Vocabu-laries for the UK Public Sector.” Public sector infor-mation is being made available through an e-Government Interoperability Framework that isbased on a set of standards, one of which is a “a smalland simple taxonomy,” a controlled vocabularyknown as the Government Category List (GCL). Im-provement in subject access is needed and six thesauraloptions (or models) were discussed here. In a similarvein, Widad Mustafa el Hadi focused on the need fortools to satisfy access to new media in the form ofNatural Language Processing (NLP) techniques. Inthis context, the author surveyed the “recent trends ofterminology acquisition and management in relationto information retrieval.” New trends, new tech-niques and most particularly the needs and means forbuilding ontologies were discussed. It was suggestedthat some testing of existing tools on a representativecorpus is needed before drawing up the ontology. Inthe final paper of the session Hur-Li Lee and AllysonCarlyle gave a presentation on “Academic LibraryGateways to Online Information: A Taxonomy ofOrganizational Structures.” The authors describe apreliminary analysis of schemes applied by academiclibraries in arranging their electronic resources. Mostof the gateways considered were created locally anddo not conform to the well known universal schemes.When the websites in academic libraries in 10 coun-tries were examined, there were many differences andfew recognizable patterns found. Cultural differenceswere the exception to this. No final conclusions couldbe drawn but there were implications for future re-


97

search. Two findings did emerge: among academic li-brarians very simple schemes of organization wereconsidered appropriate; and there is a tendency for li-braries of the English speaking world to make moreeffort to organize than their counterparts in otherparts of the world.

Closely related to “Tools and Systems” of knowl-edge organization was a session of “Models and Meth-ods” in “Retrieval.” The first paper by Gerhard Ri-esthuis and Maja Žumer on “The Functional Require-ments for Bibliographic Records and Knowledge Or-ganization” dealt primarily with descriptive cataloguingdata, as opposed to subject cataloguing. The catalogue isviewed as an interconnected network with special at-tention to access points and relationships between enti-ties and the changes they will bring to the new cata-logue model. A paper by Snunith Shoham and RochelleKedar on “The Subject Cataloguing of Monographswith the Use of a Thesaurus” presented the findings of astudy of indexing procedures with the use of a thesau-rus for post-coordination. The article outlines numer-ous subject cataloguing mistakes attributed to the use ofpre-coordinate systems – specifically LCSH – and ex-plores the alternative of a post–coordinate system.Findings indicated that post-coordination resulted in anincrease in the number of descriptors assigned and thelonger the item catalogued the greater the number ofdescriptors. Mistakes in the assignment of descriptorstended to be for topics with less than 20 percent literarywarrant in the item. Other errors could be attributed tothe lack of sufficient scope notes and the lack of famili-arity of the indexers with the post-coordinate type ofindexing. In the paper on the “Application of the Can-tor Set Theory in Making Decisions about CollectionsDevelopment” by Ana Pérez López, Mercedes de laMoneda Corrochano and Ángel Moros Ramírez, Can-tor Set Theory is used as a decision-making tool in col-lection development. It takes a “structuralist approach”and depends on the identification of underlying struc-ture of the subject provided by relations between de-mand, knowledge and patterns of publication. The re-sulting structure is then used to formulate a collectionsdevelopment policy. The application of the theoryproved useful as it aided in the identification and repre-sentation of the core of knowledge, the peripherals andthe degree of interdisciplinarity in particular thematicareas. Hemalata Iyer and Jeanne Keefe described an ex-ploratory study to assess the value of “WordNet as anAuxiliary Resource to Search Visual Image Database inArchitecture.” Carried out at the Rensselaer Polytech-nic Institute, Troy, New York, the Art & ArchitectureThesaurus provided the terminology and the study re-

vealed problems encountered by “lay users” and ob-served how WordNet was useful. In the final paper ofthe session, a paper by Douglas Tudhope, Ceri Bind-ings, Dorothee Blocks, and Daniel Cunliffe on “Repre-sentation and Retrieval in Faceted Systems” discussedthe retrieval potential of faceted thesauri and XML rep-resentations of fundamental facets. The work describedis related to the “FACET” project, collaborative re-search between the University of Glamorgan in Walesand the UK National Museum of Science and Industry.

Another “Models and Methods” session that focused on“Conceptual Relationships” contained five papers, threeof which were domain related. In their paper on “Struc-tured Models of Scientific Concepts for Organizing,Accessing and Using Learning Materials’ TerenceSmith, Marcia Zeng and the ADEPT Knowledge Or-ganization team “introduce a ‘strongly-structured’model of scientific concepts that provide the founda-tion for a knowledge base.” The domain-related systemis described and applied in a digital library environ-ment. Another domain-related presentation on “Refer-ence Linking in Economics: the CitEc Project” by JoséManual Barrueco and Vicente Julián Inglada described asoftware system that would provide for the automaticlinking of each work cited in a document with its elec-tronic full text. CitEc is the agent that automates theprocess. Also, domain-related, “Equivalence in Tillett’sBibliographic Relationships Taxonomy: A Revision”by Allyson Carlyle and Lisa Fusco, challenged BarbaraTillett’s definition of equivalence as being too restric-tive by excluding relationships among items “that may,based on contexts of use, act as equivalent.” A reviseddefinition is offered. Closely related to the Carlyle andFusco paper was the final presentation of the session byJosé Antonio Fraís and Ana Belén Ríos Hilario entitled“Visibility and Invisibility of the Kinship Relationshipsin Bibliographic Families of the Library Catalogue.”This work also focuses on the study of the “FunctionalRequirements for Bibliographic Records.” As such, it is adetailed analysis of that document and is more closelyrelated to descriptive cataloguing than to the represen-tation of topical subjects. Jonathan Furner described “AUnifying Model of Document Relatedness for HybridSearch Engines” that focused on providing a systemthat would give information seekers the opportunity“to exploit multiple sources of evidence of documentrelatedness in the process of search.” The proposed sys-tem is graph-theoretical and would give users controlover multiple ways in which document collectionsmight be ranked and re-ranked for retrieval.


98

In the final of the six sessions on “Models and Methods”the focus was on ‘Integration in Information Systems.”At the most general level Rebecca Green, Carol Beanand Michèle Hudon discussed “Universality and BasicLevel Concepts.” In doing so, they questioned whethera concept’s hierarchical level affects its universalityacross schemes for representing and organizing knowl-edge. As a basis for analysis, equivalents were drawnfrom a bilingual thesaurus, a pair of biomedical vocabu-laries and two ontologies. Findings suggested thatequivalent concepts occur most often at basic levels asopposed to either subordinate or superordinate levels.As a consequence, the authors recommended that at-tempts to integrate knowledge representation and or-ganization tools should be made at the basic level. Theremainder of the papers in this session focus on specificsystems. In “Chronotope and Classification: HowSpace-Time Configurations Affect the Gathering of In-dustrial Statistical Data” Grant Campbell used MikhailBakhtin’s theory to investigate how representations ofspace and time affect the first six classes of the NorthAmerican Industry Classification System. Closely relatedto Campbell’s work was the presentation by FernandoElichirigoity and Cheryl Knott Malone, entitled “Rep-resenting the Global Economy.” In conjunction withthe North America Free Trade Agreement (NAFTA)the three countries involved – Canada, Mexico and theUnited States -- had developed a unified North AmericanIndustry Classification System to replace separate eco-nomic classification schemes used in each country. Theauthors took the novelty of this situation as an oppor-tunity to explore the implications of the creation ofsuch a tool, to explore the role of economic theory andpolitical context as they affected classificatory theoryand to ponder the fact that the organizing data aboutindustries “serves an essentially different purpose thanschemes for organizing library materials.” In a paperentitled “Corporate Thesauri – How to Ensure Integra-tion of Knowledge and Reflections of Diversity,”Marianne Lykke Nielsen and Anna Gjerluf Esau com-pared three methodologies for constructing thesauri.The methods used were literary scanning, word associa-tion tests, and the involvement of subject expertgroups. Their analysis indicated that the methods arecomplementary and should be used together in con-structing the same thesaurus. In her paper on ‘Knowl-edge Integration and Classification Schemes,” NancyWilliamson used the three major universal schemes in apreliminary analysis on the handling of the domain“environmental sciences.” This work in progress is con-cerned with problems of primary location, interdisci-plinarity, degree of scatter, terminology and structure.

“Semantic Views over Heterogeneous and DistributedData Repositories,” by M.V. Hurtado, L, García and J.Parets was concerned with providing users with inte-grated access to data located in different repositories.The authors propose a two step process which wouldpreserve the semantics of the data stored. A first stepwould provide an integrated schema or semantic over-view and an integrated terminology as a basis for de-scribing the content of the repositories. In the secondstep the schema and terminology would need to be en-riched to allow for “evolutionary” aspects of the sys-tem.

A session on “Organization of Integrated Knowledge inthe Electronic Environment” brought together fivepresentations on the Internet. A paper on the “Organi-zation of the Information about Health Resources onthe Internet” by J.A. Salvador Oliván, J.M. Agnós Ul-late and Ma. J, Fernández Ruíz, focused on the fact thatthe usual search engines are inadequate to meet theneeds of users. As an alternative they propose the de-sign of a medical information retrieval system that usesMeSH as the controlled language and fields structuredprimarily on the basis of the Dublin Core. Criteria forquality and credibility were provided and the goal was asystem that would be an improvement on other searchengines and subject gateways. A “Practical Method toCode Archive Finding Aids in Internet” by EduardoPéis, Antoni Ruiz, Francisco José Muñoz Fernandez,and Francisco de Alba Quiñones dealt with anotherspecialized retrieval problem. Taking advantage of theneed for specialized handling, the authors propose theuse of the Encoded Archival Description (EAD). Basedon SGML this tool “makes information interchangepossible.” In contrast, Marthinus van der Walt’s presen-tation described “An Integrated Model for the Organi-zation of Electronic Information/Knowledge in Small,Medium and Micro Enterprises (SMMEs) in South Af-rica.” This paper explored the “feasibility of using abusiness process model as a framework for the inte-grated organization of electronic information in thecontext of business enterprises in general” and in par-ticular one-owner businesses and firms with fewer than100 employees. Three contexts were examined – docu-ment creation, document collection and informationretrieval systems – based on surrogates. Findings sug-gest that this approach is “theoretically feasible.” Fi-nally, Roberto Poli presented a paper on “Framing In-formation” that distinguished between semiotic, seman-tic and ontological classifications.


99

Another area of concentrated interest was artificial in-telligence, for which there were three sessions. One canconclude from an examination of these sessions thatmuch active work in this area is occurring in academiccommunities in Spain where all but two of the paperson this topic were produced. One paper came fromeach of Singapore and France. The Spaniards appear tobe a closely knit research group since some authors areinvolved in more than one of the papers. In the firstsession on “Applications of Artificial IntelligenceTechniques to Information Retrieval” three paperswere presented. Christopher Khoo, Karen Ng and Shi-yan Ou carried out “An Exploratory Study of HumanClustering of Web Pages.” Here the purpose was to findout how human beings cluster web pages naturally.Based on ten queries, a group of web pages were re-trieved using the Northern Light search engine. The re-sults were then sorted by three human subjects into“meaningful” categories. Findings indicated that differ-ent users clustered the same web pages quite differently,creating different categories. The goal of the authors isto develop automatic methods of categorizing webpages based on the categories that the users create. Sté-phane Chaudiron, Majid Ihadjanene and François Rolepresented a system designed to provide for “AuthorialIndex Browsing in an XML Digital Library.” The sys-tem proposed would offer a possible delivery formatfor controlled vocabularies and might “offer a single in-terface to different controlled vocabularies and classifi-cation systems.” Xavier Polanco set out a proposal touse “Clusters, Graphs, and Networks for AnalyzingInternet-Web Supported Communication Within aVirtual Community.” Co-citation analysis was used inretrieval from the websites of countries of the Euro-pean Union and examined using graph theory and so-cial network analysis. Further research, in the frame-work of pattern recognition and exploratory dataanalysis will be based on graph theory. Pedro Cuesta,Alma Gómez and Francisco Rodríguez investigated“Using Agents for Information Retrieval” with a focuson intelligent information retrieval agents. Such agentsare seen as a “key technology” for the future of theInternet. The functions of such agents are described asassisting users in finding useful relevant information,managing and overcoming difficulties associated withinformation overload, informing users of new relevantdata that has been published, and carrying out tasks in-dependently or working in a co-ordinated way withother agents. A prototype system called MASIR is in-troduced and described.

In the second session on “Applications of Artificial In-telligence,” four papers were presented on topics rele-vant to aspects of the field. All contributors come froma group of researchers at the Universities of Granadaand Extremadura. An “Evaluation of the Applicationof Genetic Algorithms to Relevance Feedback” pre-pared by Cristina López-Pujalte, Vincente P. GuerreroBote and Félix de Moya-Anegón described different ge-netic algorithms that could be applied to relevancefeedback using documents found in literature. A vectorspace model was used and comparison was made with atraditional feedback algorithm previously used in astudy by Gerard Salton. The experiment was carriedout using the Cranfield collection. Two other paperswere presented by Félix de Moya-Anegón on behalf ofcolleagues. “An Inductive Query by Example Tech-nique for Extended Boolean Queries Based on Simu-lated Annealing-Programming“ addressed the problemof the user who does not know the terminology of asubject needed to express a valid query. The system de-scribed would automatically derive queries from a set ofrelevant documents supplied by the user. A paper enti-tled “Graphic Table of Contents (GTOC) for LibraryCollections” introduced the application of UDC codesfor subject mapping. Multi-dimensional scaling (MDS),cluster analysis and neural networks were the method-ologies used. A conclusion describes the viability of theapplication of such an approach. In the final paper inthis session on artificial intelligence, Luis de Campos,Juan Fernández-Luna and Juan F. Huete provided abrief survey of the literature on applications and mod-els for “Managing Documents with Bayesian BeliefNetworks.” These tools deal with the problem of “un-certainty” in information retrieval.

In the third session on “Applications of Artificial Intel-ligence,” there were four papers, all prepared by groupsof researchers from the University of Granada, and theUniversities of Jaén and Extremadura in Spain. In-cluded were papers on “Using Neural Networks forMultiword Recognition in IR;” “Artificial NeuralNetworks Applied to Information Retrieval” and twopapers using fuzzy set theory: “On the Evaluation ofXML Documents Using Fuzzy Linguistic Techniques”and “Fuzzy Logic for Measuring Information RetrievalEffectiveness.”

Two sessions focused on “Integration of Knowledge inMulticultural Domain- Oriented and General Systems.”Significant work is being done in this domain and theeight papers on this topic were presented in two ses-sions. The papers took a variety of approaches to multi-


100

lingual problems. “Knowledge Organization in a Multi-lingual System for the Personalization of Digital NewsServices” by Antonio Garcia Jiménez, Alberto DíazEsteban and Pablo Gervás was concerned with sendingperiodic news selections to subscribers of digital news-papers by means of electronic mail. It describes a sys-tem called “Hermes” that applies existing techniquesfrom the field of text classification and categorization.Based on user profiles it selects items from differentnewspapers in different languages (in this case Spanishand English) – that is, a kind of automated bilingualSDI system which also allows for relevance feedback.Similarly, in “Knowledge Representation of GenderStudies on the Internet,” Maria López-Huertas andMario-Guido Barité Roqueta focus on retrieval of in-formation in a variety of languages. One domain, “gen-der,“ was chosen and eight search engines, two interna-tional, one international affiliate and five local (in Span-ish) systems were used. The research took two ap-proaches – qualitative and quantitative. Analysis basedon retrieval on the word “gender” was carried out, us-ing the structural and semantic contextual analysis ofthe search engines. Not surprisingly, the internationalsearch engines were superior. And some search enginesperformed inconsistently. Victoria Frâncu, in her paperon “Language-Independent Structures and MultilingualInformation Access,” was concerned with the need forthe creation of adequate tools to enable efficient re-trieval of information across “geographical, linguisticand cultural boundaries.” She suggested a classificationsystem, such as UDC, as a solution to the problem.The paper is a case study that describes in detail a multi-lingual access system that uses two approaches to in-formation retrieval languages – the mapping of a tradi-tional classification system onto an interdisciplinarythesaurus and the use of either UDC codes or UDC-based descriptors in various languages. Languages in-cluded in the project were English, French and Roma-nian. In their paper on “Models for Collaborative Inte-gration of Knowledge,” Annelise Mark Pejtersen andHanne Albrechtsen introduce a cognitive engineeringapproach to modelling collaborative integration ofknowledge in work domains. A generic means-endsmodel was used to provide a theoretical foundation,and a decision task model was used in decision making.The problem of the integration of knowledge was ex-plored in a web-based study of collaborative film index-ing. Richard Smiraglia’s paper took a somewhat novelapproach to the investigation of multiculturalism. Hispaper, “Crossing Cultural Boundaries,” provided per-spectives on the popularity of creative works (knownvariously as opera, oeuvres, works). The investigation

was based on evidence that the popularity of workscontributes to “mutation and derivation” of their intel-lectual content and that the more popular the work,the more likely change will be observed over time. Inthis context, Smiraglia studied a sample of best-sellingworks (fiction and non-fiction) from 1900 to 1999. Thisis a work in progress, but preliminary results suggestthat his basic assumption will be supported. In “SomePatterns of Information Presentation, Organization andIndexing for Communication Across Cultures andFaiths,” A. Neelameghan and Hemalata Iyer see devicessuch as indexes, glossaries of terms and presentation oftext in different languages and scripts as aids to com-munication across cultures, faiths and linguisticboundaries. Their research used documents from thespiritual domain.

As might be expected, the conference also included asession on “Epistemological Approaches to Classifica-tion Principles” in which there were several papers on“design and construction.” The presentation by BirgerHjørland addressed the state-of-the art in “The Meth-odology of Constructing Classification Schemes.” Hepostulated that previous methodologies neglect theempirical basis of classification and provided evidencefor the necessity of historical and pragmatic methods.His presentation covered the methods of constructingclassifications from Ranganathan to the design of on-tologies in computer science. Further he referred tothe recent ‘paradigm shift’ in classification research.Hope Olson, Juliet Nielsen and Shona Dippie dis-cussed “Encyclopaedist Rivalry, Classificatory Com-monality, Illusory Universality.” In it they describethe cultural construction of classification as seenthrough the work of two French Encyclopaedists,Jean d’Alembert and Denis Diderot, and the work ofSamuel Taylor Coleridge. As presented at the confer-ence, their work is part of a larger study which is ex-ploring the cultural construction of classification andconstitutes an example of the first phase of the projectwhich is “a deconstruction developed from relevanttexts.” The texts are being encoded and when the en-coding has been fully developed “it will allow analysisof texts from multiple perspectives.” In this same ses-sion, a paper on “Evolving Paradigms of KnowledgeRepresentation and Organization” by Jain Quin, pro-vided a comparative study of classification, XML/DTD and ontology in which she based her compari-son on “how they specify concepts, build data modelsand encode knowledge organization structures.” Dif-ferences and relations in knowledge organization wereexamined and four paradigms were generalized – inte-


101

grative and disintegrative pragmatism, and integrativeand disintegrative epistemologism. One of the find-ings in the comparison was the indication “that morerecent approaches to knowledge representation andorganization are developed using the foundations es-tablished by precursors.” Significantly, this findingmirrors comments and results from a number of pre-senters at this Conference. In this same session, I.C.McIlwaine’s paper, “Where Have All the FlowersGone? An Investigation into the Fate of Some SpecialClassification Schemes,” took the conference back tothe roots of classification theory. The ClassificationResearch Group (CRG) was a driving force in thecreation of “a framework and body of principles thatremain valid for the retrieval needs of today.” High-lights of this paper were the history and backgroundof the CRG and its special schemes and fundamentalprinciples that remain valid. In contrast to this, Jens-Erik Mai asked the question “Is Classification TheoryPossible?” Mai is “Rethinking Classification Research”and questions “the assumption that bibliographic clas-sification theory can resemble scientific theories.” Indoing so, he argues that expertise is not achievedthrough the correct application of classification the-ory but that it is the more practical experience and ad-justment to context that operates in the real world.

Perhaps the most unusual session of the conferencewas the session on “Professional Ethics. Users and In-formation Structures. Evaluation of Systems.” It wasunusual in two ways. First of all ”ethics” is a topicthat has seldom, if ever, been a major topic for discus-sion at conferences of this kind. Will this be a newtrend? One can only surmise that ethics has becomean issue because of globalisation, national and interna-tional sharing of databases, the Internet and advancesin technology. Secondly, unlike the other sessions,this one dealt with three distinct topics – ethics, usersand evaluation. As for the mix of topics, it is clearthat there were insufficient papers focused directly onthese topics to warrant their own independent ses-sions. Three of the papers focused on the ethics issue.“Ethical Aspects of Knowledge Organization andRepresentation in the Digital Environment” by J.Carlos Fernández-Molina and J. Augusto C. Gui-marães aimed to identify the principal ethical valuesrelated to the representation and organization ofknowledge and to find out to what degree these valuesare addressed in the codes of the professional organi-zations. Findings suggested that ethics are addressedbut not in sufficient detail and that there is need todeal with the problems that have come about with

generous sharing of information and the advent of theInternet. A second paper, “Knowledge Organizationfrom a ‘Culture of the Border,’” presented by Anto-nio García Gutiérrez dealt theoretically and meta-phorically with multiculturalism and the problems oftranscultural ethics. Then Christopher King, DavidMarwick and M. Howard Williams, presented a paperentitled “The Importance of Context in Resolving ofConflicts When Sharing User Profiles.” Their concernis that while user profiles have been an important fac-tor in the everyday life of information retrieval forsome time, such profiles are being “assembled by in-formation providers with no regard to future sharing”of this information. Their paper discusses the prob-lems involved and a prototype has been constructedto examine their ideas on this topic. Two of the pa-pers in this session focused on users. “Ariadne’sThread: Knowledge Structures for Browsing inOPACs” by Carmen Caro Castro and CríspuloTravieso Rodríguez presented a discussion on subjectsearching with particular reference to “how users’ ex-pressions match subject headings” and to determinewhether “knowledge structure used in online cata-logues enhances searching effectiveness.” In a some-what different look at users, Ali Asghar Shiri, Craw-ford Revie and Gobinda Chowdhury presented astudy “Assessing the Impact of User Interaction withThesaural Knowledge Structures.” This was a pilotstudy to determine the extent to which a thesaurus-enhanced search interface to a web-based database aidsend users in their selection of search terms. Finally,Linda Banwell explored the issues surrounding “De-veloping an Evaluation Framework for a Suprana-tional Digital Library” as seen through the TEL (TheEuropean Library) project. The project, funded bythe European Union, is a project to set up a co-operative framework for access to the major national,mainly digital collections in European national librar-ies. As described by Banwell, the focus is on “consen-sus building, and also includes preparatory technicalwork to develop testbeds, which will gauge to whatextent interoperability is achievable.” Her paper de-scribes the work on the project to date and seeks “toestablish what are the key drivers, priorities and bar-riers encountered in developing such a framework.”

While the conference was presented in what were in-tended to be clearly defined sessions, these sessionswere far from being mutually exclusive and manytopics, problems and issues often spilled over intoother presentations and discussions, in that they per-meated the whole conference. At the beginning of the


102

conference twelve participants were invited to act asrapporteurs or relators. At the end of the conferencetheir views were brought together as a summary andpresented in a wrap-up by Ia McIlwaine. In conclu-sion, a selection of those comments are included here.Clearly, the thesaurus predominates as the favouredtool for retrieval and the concept of ontologies is be-coming more and more prominent. However, asMcIlwaine indicated, “under every good thesaurusthere should lurk a classification scheme.” And whilea number of papers “showed an awareness of tensionsbetween classification as a culturally specific activityand classification as an activity based on universalconcepts,” classification, in particular facet analysis, isstill fundamental to knowledge organization and rep-resentation of knowledge. Not surprisingly the prob-lems of interdisciplinarity, the need for interoper-ability and the shortcomings of the traditional ap-proaches to knowledge organization were all in evi-dence and highlighted by several contributors. Thedesirability of capitalizing upon the expertise of allbranches of the information world, and the experi-ence of those in libraries, archives and museums wasstressed. The representation of what Elaine Svenoniuscalls “aboutness” and the challenge of coping withcreative thought and theoretical innovation in thehumanities and the social sciences were explored. Thepossibility of combining classificatory approacheswith a mark-up language, such as XML, was investi-gated and its potential as a tool to assist with prob-lems posed by interoperability between systems wasexplored. Also, artificial intelligence was seen as animportant player in future information systems and,of course, the Internet was not forgotten.

The theme of many sessions included the words“Models and Methods for Knowledge Organization”which, to some extent, could be claimed to be the un-

derlying intention of the whole conference. As de-scribed by Ia McIlwaine in the wrap-up, another pre-dominating feature was the reporting of individualprojects that have been carried out by discrete groupsof people or individuals. Of course, proper attentionhas been paid to the work of others in the field, butthe prevailing problems of retrieval in a networkedenvironment do call for some corporate activityacross international boundaries and this is not yetvery evident. Many of the enterprises that we heardabout have rejected the tried and tested approaches infavour of ontologies, taxonomies and approaches fa-voured by computer scientists.

In summary, the principal themes that emerged arethe fundamental need for facet analysis and a plea for“back to basics,” together with the use of categoriesand a systematic approach to the design of retrievaltools. The general preference is for the thesaurus asthe favoured tool for subject retrieval, and the sadlack of up to date standards for the creation of re-trieval tools has been commented upon. Related stud-ies in Artificial Intelligence and in lexicography andthe need to draw upon others working in parallelfields, and to make our methods known to them,were emphasised, especially in the design of ontolo-gies for accessing data on the Web. XML has featuredas a useful tool on a number of occasions. Social andethical problems and the needs of users from differentbackgrounds and at different levels of sophisticationwere emphasized. Overarching all of the papers therewas the challenge of effective retrieval of informationfrom the Web and the need for work on interdiscipli-nary fields and across different cultural divides in or-der to resolve the problems posed by the organizationof resources in a networked environment.

Knowl. Org. 29(2002)No.2Book Reviews

103

Book Reviews

Edited by Michèle Hudon

Book Review Editor

SCHMITZ-ESSER, Winfried. Expo-Info 2000 :visuelles Besucherinformationssystem für Weltaus-stellungen [Expo-Info 2000 : a visual informationsystem for visitors of World Exhibitions]. Berlin,Germany : Springer, 2000. xii, 119 p. ISBN 3-540-67307-5.

This book describes a blueprint for an information sys-tem for the World Exhibition (Expo) 2000 in Han-nover (Germany). Although the system was never im-plemented, the ideas developed here were, and remainvery interesting. The purpose was to develop a systemwith a very high degree of user-friendliness, and withfunctionalities that would enable the many visitors ex-pected at Expo 2000 to find their way to those parts ofthe exhibition of interest to them. The proposed sys-tem is based on the idea of a “Thought Space Travel”supported by a multilingual thesaurus.

The book is divided into two parts. The first part de-scribes the device to be used for the searching while thesecond part describes the supporting thesaurus.

SIVIT (Siemens’ Virtual Touch screen) was selectedas a searching device. It consists of an elliptical tablewith a camera and projector above it. The camera de-tects the movements of the hand of the user and reactsto these movements. The results are subsequently pro-jected on the table.

Access to the system is possible by keywords (freetext) and by way of a network of concepts. The key-words access works with a KWIC-index. Selection ofone of the lines of the index gives a multimedia result: aphoto or short video plus some text. Using the first re-sults, further information can be found by touching lit-tle ellipses with terms like “geographical”, “Statistics”,and “Desired effect”. Documentary sources and a mapillustrating the route to a pavilion where exhibits con-nected with the subject will be found can also be re-quested.

The “Thought Space Travel” is supported by a largemultilingual thesaurus. For each subject for which in-formation is available at the Expo, a so-called proxy-

text is made, a descriptive profile consisting of one ormore combinations (“tupels”) of four descriptors se-lected in the thesaurus. The first three descriptors areanswers to the What, Where, and When questions. Thefourth descriptor – called aspect – is less traditional; itdenotes the purpose for which the topic could be inter-esting or useful. Twelve aspects are suggested; for ex-ample, the aspect “Recommendation” is used to de-scribe recipes, instruction, and how-to-do; “Impact de-sired” describes goals envisaged, or target projected; and“Organizational” qualifies social and judicial conditions.The subject Doctors without Frontiers is assigned the fol-lowing three tupels:

What-Descriptor Where-Place

When-Time

Aspect

Free medical care Earth > 1971 organizational

Private helpinitiative

Earth > 1971 definition

Humanitarian aid Earth > 1971 organizational

The Super-Telescope “Parascope” gets four tupels:

What-Descriptor Where-Place

When-Time

Aspect

Very large tele-scope

AtacamaDesert

> 1987 technical

Discovering space AtacamaDesert

> 1998 impactdesired

Observationof space fromthe Earth

AtacamaDesert

> 1998 result

Atmosphericaldiffraction

Earth <> 0 assessment ofconsequences

The What-descriptors are general descriptors. Propernames are not used as descriptors but they can be foundthrough the free text keyword system.

An important aspect of the system is that there aresyntactic relations between all the descriptors. The“discovering space”, in the indexing of the Parascope


104

telescope, will always be connected with “impact de-sired”, the year “> 1998”, the country “Chile” and thecontinent “South-America”. In the searching system,the what-descriptors and the place-descriptors are con-nected to broader and narrower descriptors.

The thesaurus itself is multilingual, including termsin English, French, Spanish and German. There are sixrelations between the terms in the thesaurus: in addi-tion to the equivalence relation they are: generic/abstract, partitive, geographical, beneficial and detri-mental. A few simple examples: Overfertilization is det-rimental to Biotopes; Automatic translation is a part ofComputer linguistics.

It is of interest to note that Schmitz-Esser has alreadypublished several papers in relation to this thesaurusmodel. At the 2000 ISKO Conference, in Toronto,Canada, he presented a thesaurus for describing dyna-mism in the world with partly different relations1. TheExpo thesaurus does not include the instrumental rela-tion that is present in the author’s other projects. In theExpo thesaurus, the instrumental relation appears to beincluded as a partitive relationship, as in Salk vaccina-tion which is a part of the Fight against poliomyelitis.This example from the book shows a weakness in theuse of these relations. Why would this be a “is part of”relation rather than a “beneficial for” relation?

The book is interesting and, in my opinion, also im-portant. It gives a sketch of a complete information sys-tem built for a clearly defined user group. The thesau-rus and the automated search system are developed asan integrated whole. The extra search possibilities thatcomputers can offer for information systems with con-trolled vocabularies are used in an intelligent way. It isregrettable that this project could not be put in applica-tion; it would have been interesting to be able to evalu-ate the experiences of the users.

Notes

1. Schmitz-Esser, W. (2000). How to cope with dy-namism in ontologies. Dynamism and Stability inKnowledge Organization : Proceedings of the SixthInternational ISKO Conference 10 – 13 July 2000.Toronto, Canada., 83-89.Würzburg, Germany : Er-gon Verlag.

Gerhard J.A. Riesthuis

Dr. G.J.A. Riesthuis, Leerstoelgroep BAI, Univer-siteit van Amsterdam, Oude Turfmarkt 147, 1012 GCAmsterdam, The Netherlands.E-mail: [email protected].

WRIGHT, Sue Ellen, and BUDIN, Gerhard, eds.Handbook of Terminology Management : Vol.2Applications-oriented Terminology Management.Amsterdam, The Netherlands : John Benjamins, 2001.549 p (pp. 371-920). ISBN 9-0272-2155-3.

1.Overview

The second volume, under review, has taken four yearsto materialise after its announcement when the firstvolume appeared in 1997. Little wonder that the actualpresentation has taken some liberty with what hadbeen projected. Time takes its toll and, more perti-nently, the situation depicted of terminology manage-ment (TM) applications, as the compilers emphasize, isone of permanent change, so that this second volume,by presenting as it were, “a kind of snapshot of the topic(s)in question, reflecting the conditions at a particular mo-ment in time” (p.371), takes the risk of quicker obsoles-cence than the first volume which covered more dura-ble issues. The risk is well taken, since despite the sus-penseful state of some fields, like exchange formats, andthe resultant expectancy, the general impression is oneof coherent validity, mainly on account of the compil-ers' talents for evening odds up. However, the websiteoffered for updating the information provided (http://appling.kent.edu) leaves the curious in the lurch. Noteven the present volume is presented, but the site is saidto be “revamped” along with expected updates. Thepersevering reader, when reverting to the print docu-ment, will be rewarded for his persistence by an infor-mative collection of articles made available by the com-pilers.

The present volume keeps to the layout announced ofthree additional chapters on(6) Information Management,(7) Commercial & Industrial Applications, and(8) Computer Applications for Terminology.

Even though these topics are liable to suffer under theendemic blast of change of the computer world rufflingthe terminological scene, the main features of that sceneare steadfast enough to warrant an inventory at anytime. The three chapters will be evaluated in more de-tail later.

The natural evolution of the discipline has justifiedthe reshuffling of appendices and the excellent info-boxes, as well as other minor manipulations of the pro-jected volume. It is frustrating, however, to find thatthe additional selected bibliography promised for Vol-ume 2 appears to have vanished.


105

2. Evaluation

A good part of the book’s Introduction repeats whathad been said under that title in the first volume. It istempting to consider repetitions fastidious, but theymay be justified by the compilers' concern to makeboth ends meet: that of user-friendliness with that of ef-ficiency.

The good idea of setting keywords in the margin ofthe text throughout has continued to be implemented,facilitating easier referencing and permitting the in-tended hypertextual usage of the information pre-sented. This layout favours only the odd pages; the set-ting of keywords in the inner right margin of the evenpages does not offer the same degree of facility for rapidhypertextual navigation and takes some practice.

The index serves only for this volume. There is nogeneral index for the whole handbook, nor are thereseparate ones for authors and subject-matter respec-tively. This is regrettable since the 1st volume index wasalready considered scanty. It is true that the many hy-pertext-like links by keywords included in the textcompensate partly for this weakness, although not allkeywords are found in the index.

Chapter 6 – Information Management – starts with adescription of computer-assisted thesaurus manage-ment, yet there is no definition of information; the re-lation of information to thesauri is hard to guess fromthe presentation. Even the three attempts at defining athesaurus are a non-starter. First a thesaurus is a docu-mentary language, then it is a list of terms semanticallystructured, and six pages later it ends up being a systemof data sets and relations. The reader required for sortingout this sort of statements would find it degrading towaste his time on such an exercise, when, in addition,he is reading under “computer assistance” that thesauriare paper-hungry. Maybe he would be relieved to learnthat the “facets” of ISO 5964 are not covered by the ar-ticle, whatever they are. The discussion of tools for the-saurus construction and maintenance with a view tochoosing the right solution (for whom?) skips the mainconcern of terminologists, namely the vital link be-tween their daily bread, that is, terms, and the appro-priate structuring of their findings. This poses the moregeneral question of who is supposed to be the targetreader. It seems that a method showing just a possibleend to reach without mapping out the way thereto isnot very helpful.

Terminology and indexing for information retrievalare the two next subsections soberly presented by thesame author, Strehlow, providing a brief introductionto standard indexing practice, and exploring the critical

principles of precision and recall of randomly accessibleinformation. It appears that terminology, by its suc-cinct technicality, facilitates topical location of informa-tion, whereas Zipf's rule refines its occurrence; bothconstitute ready tools for the indexer's job. However,finding that the FAQs quoted do not relate to indexing,and, since the question, “when is my data saved?” (p.423)appears twice, this raises the suspicion that precisionmay have been overrated. On the other hand, the ISOstandards quoted in the bibliographical lists of bothsubsections do not show up in the standards listed inAppendix II, which seems to show that its frame is toonarrow to allow coherence of main normative referencequotes within the handbook. After all, “terminologymanagers in their daily work” might find it frustrating tocome across lacunae of this sort, quite apart from con-sistency. To make standards work at an operative level,they need to be listed in a manual like this, to quote thecompiler, but it may not be the best idea to relegate thewhole of them to an index.

Chapter 7 – Commercial & Industrial Applications –starts off with terminology in technical document pro-duction, like user manuals, specifications, design proto-cols etc., where terminology is dedicated more to in-struction, skill acquisition and action than to informa-tion. The many examples given disperse the reader'smain interest in how terminology actually functions intechnical writing. The action verb focus as opposed tothe noun predominance of most terminologies justshows the marginality of this type of TM. On the otherhand, the insistence on definition is a nudge with abarge pole. It again raises the general question of thetarget readership of the collection, and the sad experi-ence of seminar attendance is confirmed to the effectthat many a contributor feels like starting from scratch,taking the patient audience as hostages in a vain hold-upexercise.

The subsection on Industrial Management sets outfirst “Terminology as an Organisational principle inComputer Integrated Manufacturing (CIM) Environ-ments”. The author and co-compiler S.E.Wright de-scribes the central function of terminology as a drivingforce behind the assignment of data element names, al-though the role of terminology in CIM is somewhatcurtailed by the well perceived compartmentalisationcharacterising large industrial corporations. Rare are thefirms that have understood the need for streamliningthe internal flow of information and adopting a firm-wide uniform terminology, and even fewer have beenable to pacify the departmental trenches and vanquishresistance in order to debug knowledge circulation. Sothe situation envisaged by the author pictures a rather


106

idealistic frame of integrated corporate data exchangesystems with a core terminology management difficultto implement. The question remains of how practicablethe principles and models exposed in a handbookshould be to help the reader on his way. Typically, thereferences mostly include standards, norms, and guide-lines which are either abstruse or self-evident, thoughthey figure a convention which a handbook has, in-deed, to record. On the other hand, it could be arguedthat the theoretical interest is great for computer appli-cations in the so-called knowledge engineering drive ofmodern management techniques. But then the funda-mental question arises about the role of terminology asenvisaged in chapter 7, especially in hightech industrialapplications. If the claim is that “the conventional (ob-ject) numbering techniques can no longer keep pace withthe increasing complexity of modern products and thegrowing stream of technical information” (p.480) and thatterminology is to fill the gap, the question immediatelyarises as to the limit, the purpose, and, above all, thecontent of the term. If identification is the main con-cern, numbers beat terms. The infinity of objects/concepts stands against the boundaries of language, butcan, indeed, be matched by the infinity of numbers. Inaddition, the computer with its numerical approach re-duces all input, including terms, to its binary code.Terminology, like people's names, is of little concern toit. The man-machine relations require relics like termi-nology for man. The machine functions with numbersonly. This should be borne in mind when implicatingterminology in computer-assisted management systems.It is true that linking coherent numbering systems toconsistent terminology streamlines information man-agement and facilitates understanding of ultra-thintechnicalities, like the 21-thousandth shade of the deepblue sea. A number will easily permit identification ofthis object. Language usage does not cut as thin, andterms are semantically relatively cumbersome, thickitems. A code may be the answer, because even termscan be coded, and codes are numbers. So it is an illusionto have both terms and numbers, since all boils downto 0-1. Take the Unicode standard! It clearly evidencesby its comprehensiveness, requiring for all data objectsnot only terms but also all other identifying attributesto “meaningfully” identify such objects, the superiorityof number coding. We leave the object world, when weembark on meaning. And what's worth an opinion onwhether a code is still a term or not? Use and usage willfinally clinch the matter.

The trouble with papers like Knowledge engineeringin enterprises, by Düsterbeck and Hesser, is that theythrow up a misleading title and then report on a very

specific project, in this particular case, stock control ina specific company. It is indeed hard to see why a spe-cially designed system, which may have its casual mer-its, should deserve general interest in a handbook meantto provide a systematic overview of terminology man-agement practice. One cannot but sympathetically nodat the compilers' disclaimer that their work does notmerely represent yet another collection of potentially unre-lated or uncoordinated articles selected at random by theirauthors (p.377). Examples are always welcome if they il-lustrate some general principle and are linked to it notonly by a vague title.

The following article on Total Quality Managementby the co-compiler S.E. Wright leaves a better impres-sion, since it develops a scenario designed to demonstratehow terminology management practices provide qualityassurance support for multilingual document generation(p.501), and thus provides an example to follow, which,alas, remained unheeded by many other contributors.

Case studies like that reported in Project-integratedterminology management for technical writing & transla-tion document again specifics as presented hotchpotchat seminars and which are not even controversial forlack of general interest. What is the point then of point-ing out that the proposed four text types for categorisa-tion are no text types at all or that anecdotes are gener-ally more amusing to the teller than to the reader? Andthere is no point either in remarking that a search onterms is vain if it ignores the concept-level.

Terminological localisation of computer software,which is the topic of the last article of chapter 7, is aparticularly hot issue in the wake of globalisation. It re-flects the readiness of target markets to import a newproduct with its foreign name or, on the contrary, toreject the name while adopting the product under anacceptable local designation. The generalisation of PCshas borne out the varying sensitiveness of nations withregard to their language feeling. In Europe, German-speaking countries have shown a relative insensitivitytowards importing US terminology, (the present Chan-cellor buttonholing “unsere amerikanischen Freunde”),whereas the French and Spanish-speaking world hasrisen in arms against the invasion, led by those (Quebecand Cuba), who live nearest to the US-German friends,since a neighbour rarely is a friend. The localisation is-sue impersonates the well-perceived “implied user” in atarget culture. The form and format of localising is onlycasually attended to, since “there are [software] toolsavailable for a variety of budgets and purposes” (p.533),and, indeed, the dominant Windows interface tends toreduce that variety to nominal. As regards the contents,that is the cultural implication, some good examples


107

highlight the principles that should underlie any searchfor acceptability both in word and image.

Chapter 8 – Computer applications for terminology– occupies over 70 percent of this 2nd volume with 15contributions, which virtually all document a reversedrelation, namely terminology in a computer environ-ment. Both terminology and computers are ancillary toknowledge organisation and management, so that data-bases, data categories, entry structures, data extraction& retrieval, interchange, compilation, and informationhandling & structuring, which all are competently andinformatively discussed here, clearly bear out what theyactually sub-serve, and that terminology is part & par-cel of a bigger deal, that of knowledge engineering. Thetechnicalities presented highlight particular concerns ofterminologists, localizers, and information specialistsworking in the language industry, while the last contri-bution on TM in artificial intelligence and knowledgeacquisition (Khurshid Ahmad) strides a wider area andfills the register.

3. Conclusion

It is always hard to assess the merits of information.Bias is a feature which, in scientific publications, cannotafford to come out as clearly as in publicity, but it iswell known that, like in journalism, selection, rejec-tion, and presentation of data forms opinion and createsfollowers. When confronted with a fuller picture in-cluding further information or with information basedon a different arrangement of facts or terms, the firstimpression will have to be revised. This is what isbound to happen after reading this outstanding secondvolume, and hopefully, the opinion gained from usingthis manual will brave it out. The reader has plenty ofnodes of hypertext to joggle off on. It is up to him toenjoy the game or disagree with ontology as picturednaively on p.888. Since hypertext allows no conclusion,no exit, he should bridle his curiosity and look out forthe first stop to take a return ticket, as Alice was ad-vised.

H. Eisele

Herbert Eisele, Terminologist, Research director fortranslation and terminology, ISIT, Paris.E-mail: [email protected].

SMIRAGLIA, Richard P. The nature of “a work”:Implications for the organization of knowledge.Lanham, MD : Scarecrow Press, 2001. ISBN 0-8103-4037-5.

As any cataloguer knows, the concept of “the work,”which is central to bibliographic description, receivesalmost no overt treatment in our cataloguing codes.Richard Smiraglia has therefore produced a timely con-tribution: the first book-length exploration of the con-cept of the work in bibliographic description. Smiragliaposes three questions: What is the nature of a work?What is a work? And who is concerned about the na-ture of works? The author attempts to provide pre-liminary answers to these questions by summarizingprevious research, by embedding the work concept in atheoretical context, and by providing empirical evi-dence of the presence of works and of bibliographic re-lationships in catalogues.

Smiraglia begins with an analysis of the work as itfigures in the primary writings on Anglo-Americancataloguing. Using an effective conceptual frame thatoriginates in Patrick Wilson’s concept of the biblio-graphic universe, he surveys the chief writings on bib-liographic description in a comprehensive and me-thodical fashion, beginning with Hyde and followingthrough such important names as Panizzi, Jewett, Cut-ter, Pettee, and Lubetzky. He covers the Paris Princi-ples, and moves on to the work of Domanovsky, Tan-selle, Wilson, Carpenter, and the IFLA Report on theFunctional Requirements of Bibliographic Records. Hethen goes on to survey the work of Tillet, Smiraglia,Leazer, Yee, Velluci and Carlyle on bibliographic rela-tionships.

From these scholars, Smiraglia extracts some generalpoints of consensus on the work: primarily that it is anabstract concept that can sustain a variety of physicalmanifestations. He separates intellectual content into“ideational” and “semantic” content, and argues that asignificant change in ideational content, semantic con-tent or both, results in the creation of a new work. Ataxonomy of bibliographic relationships can be createdto chart the growth of new works from a progenitorwork.

Smiraglia then moves on to a series of reflections onthe social and cultural importance of works, whichproduces what can only be described as a whirlwindtour of structuralist and post-structuralist theory. Hegrounds the work concept in linguistics and semiotics,drawing on Saussure, Yngve, and Peirce. He also drawson Barthes and Mark Poster, as well as music theoristsJean-Jacques Nattiez and Lydia Goehr, and uses Fou-


108

cault to emphasize the theoretical tangles we encounterwhen dealing with problems of authorship, ownershipand attribution in works. On the basis of this survey,Smiraglia draws two chief conclusions: first, that thework is a social and cultural phenomenon, and second,that the work concept rests on a tension between im-mutability, which enables multiple physical manifesta-tions to be linked with the same work, and mutability,which causes works to mutate over time.

This mutability becomes the focus of the third partof the book, in which Smiraglia presents recent empiri-cal research on derivative bibliographic relationships,much of it his own. He summarizes a series of studiesthat quantitatively measure the existence of works andof bibliographic relationships in a variety of catalogues.These studies examine the impact of various features onbibliographic families: the age of the progenitor work,the work’s form and genre, the language and place oforigin, and the discipline and subdiscipline. He followsthis with a qualitative investigation of various biblio-graphic families, complete with diagrams and models.The author comes to the conclusion that bibliographicrelationships are indeed complex, that the age of theprogenitor work is a factor in the number of works itspawns, and that a work’s popularity may be factor indegree of derivation. He concludes the book with anoutline of a potential theory of the work. A work, heclaims, is the intellectual content of a bibliographic en-tity, which functions in society as a sign functions inlanguage, and possesses the qualities of the Peirceansymbol.

This book has a number of significant strengths. Forone thing, it is badly needed, particularly at a timewhen the Joint Steering Committee for AACR2R isworking to incorporate a more articulate theoreticalbase for the Anglo American Cataloguing Rules. Smi-raglia grounds his discussion in both a theoretical andan empirical context, which should appeal to research-ers of various methodological inclinations. He providesa conscientious and detailed outline of his methodologyin the various studies that form the heart of the laterchapters, and while much of his research on derivativebibliographic relationships has already appeared else-where, it certainly benefits from being assembled andintegrated into a broader discussion. Above all, he pro-vides a superb summary of the various writings on thework, drawing together disparate voices into a concise,effective narrative on the growth of this concept overthe last two hundred years. This alone will make thebook required reading for many cataloguing courses.

The weaknesses of the book are largely mitigated bySmiraglia’s own admission that this constitutes only a

preliminary exploration of a concept that needs muchmore attention. Some of them, however, do deservementioning, if only to stimulate debate. First, Smira-glia’s questions are at odds with his method, to somedegree. The book claims to be working towards a defi-nition of the work, and that this working definition,however tentative, will be the “product” of the book’scontent. But because much of the discussion is based onan empirical analysis of the presence of works in cata-logues, he needs to have a working definition before hebegins. As a result, the empirical chapters assume thatthe primary questions of the book have already beenanswered, and the framework he produces at the end ofthe book is not significantly different from the defini-tion he produced at the end of the literature survey.Furthermore, the third question (“Who is concernedabout the nature of works?”) is answered, but the an-swer reads like a justification of the importance of theresearch, and does not really derive from the researchitself.

Certain concepts could use more clarification, par-ticularly the division of content into ideational content,which he defines as the propositions expressed in awork, and semantic content, which is the expression ofthose propositions in a particular set of linguisticstrings. The division itself causes no particular difficul-ties, and has precedents in the work of Wilson and oth-ers. But Smiraglia needs to explain how these two facetsof content can be operationally separated, and to show,perhaps through examples, how ideational content canchange without the semantic content changing as well.

Smiraglia’s use of the term “triad” when discussingPeirce, although not incorrect, may lead to some confu-sion. I would advise using Peirce’s term “trichotomy”to express the relationship between icon, index, andsymbol. It would prevent readers from confusing itwith the “triadic sign” of sign-object-interpretant,which is often used in information retrieval, particu-larly to distinguish the trinary Peircian model of thesign from the binary Saussurean model.

Smiraglia began the book by asking, “what is awork?” His book leaves me with another nagging ques-tion. Is there a work at all? Smiraglia is convinced thatthere is, and he uses semiotics and post-structuralism asa source of theory to be empirically tested. But the im-plications of the theoretical discussion work against thepremise that these things are empirically testable, atleast in the way that Smiraglia has tested them. If, asSmiraglia argues, the work is a product of social andcultural forces, what is there beyond what members ofa culture see by mutual consensus? And if “the work” isa cultural construction, then I would argue that we


109

need to pay greater attention to the social processes thatfabricate our conception of the “work.” These socialprocesses will be reflected in catalogues, but not articu-lated there. The articulations may, however, be foundin, or derived from, closer and more comprehensivereadings of social and cultural theory, and the opera-tionalization of a fresh set of variables for empiricalanalysis. Smiraglia has made a good beginning, but, ashe would be the first to admit, the task is far from over.

Grant Campbell

Dr. Grant Campbell, Faculty of Information and Me-dia Studies, University of Western Ontario, London,Ontario, N6A 5B7, Canada.E-mail: [email protected].

TAYLOR, Arlene G. Wynar’s Introduction toCataloging and Classification. 9th ed. Assisted byDavid P. Miller. Englewood, CO : Libraries Unlim-ited, 2000. xx, 552 p. ISBN 1-56308-857-6(pbk).

First published in 1964, and at the time solely authoredby Bohdan S. Wynar, this textbook has kept constantpace with the rapidly changing scene of bibliographicand information organization. Its nine editions chartthe course this field has treaded over the last four dec-ades, and the regular editions speak of its popularity. Ithas deservedly grown into a textbook which has shapedthe curricula of many library schools around the world.Associated with it since the sixth edition (1980), Profes-sor Taylor is a teacher of repute and a prolific writer inthe area of knowledge organization. A recipient of thecoveted Margaret Mann Citation and member of theDecimal Classification Editorial Policy Committee, sheis an authority in the field. Professor Taylor has en-listed the help of many colleagues and experts to keepthe text up to date and authentic, maintaining a policyof inviting widely based suggestions. After a nine yearperiod, Wynar’s Introduction to Cataloging and Classifi-cation needed drastic revisions.

The 20 chapters that make up the text have been or-ganised into six parts. Part I (the first two chapters) ex-plains the meaning, purpose and history of cataloguingand cataloguing codes. Part II (chapter 3) describes theMARC format, more specifically the MARC 21 ver-sion, as well as SGML, HTML and other formats for

encoding information. Part III (Chapters 4 to 7) discussdescriptive cataloguing and choice of access points ac-cording to the latest versions of the AACR and ISBDsfor different types of information packages. Rules havebeen illustrated with judiciously chosen examples.

Part IV (Chapters 8 to 17) deals with the theory ofclassification, shelf classification and subject indexing.Major classifications discussed in adequate details areDDC and LCC. Unused classification systems, such asthose of Brown and Cutter, are presented briefly. Re-gretfully, there is no description of Ranganathan’s CC,though its method is discussed cursorily in a section onfaceted classification. This part has detailed chapters,each on the Library of Congress Subject Headings and theSears List of Subject Headings. Some space is also given tothe presentation of a few of the best known thesauri,such as the Art and Architecture Thesaurus.

Part V (Chapter 18) deals with authority control andwork for consistency of headings to identify and collo-cate materials. Part VI (Chapters 19-20) deals with ad-ministrative issues relating to centralised and coopera-tive cataloguing, and describes the functions of biblio-graphic utilities such as OCLC. The last chapter hasbeen rewritten to be on the cutting edge and developsthe theme with a focus on the internal administrationof a cataloguing department.

The 1980 ALA filing rules are printed in an appen-dix. A glossary provides definitions of varying lengthfor some 325 select basic terms and abbreviations. Acurrent bibliography of about 300 books, journal pa-pers and standards, available either both in print or onWeb, will be useful for advanced researchers in thefield. The detailed index is exemplary, as is the analyti-cal table of contents. Figures and illustrations enhancethe lucidity of the book. Each chapter has been dividedinto sections with headings, is fully referenced, and isfollowed by suggested readings.

Wynar’s Introduction to Cataloging and Classificationdeals with a large number of topics both current andclassic. Yet its growing obesity has been skilfullychecked. The book reflects the current state of the artin the turbulently changing field of information anddocument organization, and so is indispensable for stu-dents, teachers and working cataloguers.

M.P. Satija

Dr. M.P. Satija, Guru Nanak Dev (GND) University,Amritsar, 143 005, India.E-mail: [email protected].

Knowl. Org. 29(2002)No.2Knowledge Organization Literature

110

Knowledge Organization LiteratureGerhard Riesthuis: Literature Editor

0 Form Division

02 Literature ReviewsSee also: 0189, 0196, 0243, 0261, 0262, 0263, 0279, 0297,0312

06 Conference Reports & Proceedings

0186 06.01-05-13/16; 692Andersen, J.- Works as signs, symbols, and canons : the epis-temology of the work (Lang.: eng). - In: Knowledge Organi-zation, 28(2001)4, p. 203-204.- 45 refs.

0187 06.01-11-02/08; 211Justice, A.- 12th American Society for Information ScienceTechnology, Special Interest Group/Classification Re-search.Classification Research Workshop (Lang.: eng). - In:Knowledge Organization, 29(2002)1, p. 40-44.- Refs.

0188 06.02-07-10/13; 111Challenges in Knowledge Representation and Organizationfor the 21st Century : Integration of Knowledge acrossBoundaries ; Proceedings of the Seventh International ISKOConference 10 - 13 July 2002 Granada, Spain / Organized bythe ISKO Spanish Chapter and The University of Granada ;ed.by María J.Lopez-Huertas ; with the assistance of Fran-cisco J.Muñoz-Fernandández (Lang.: eng). - Würzburg : Er-gon-Verlag, 2002.- 607 p.; 23 cm.- (Advances in KnowledgeOrganization); Vol.8). - Refs. - ISBN 3-89913-247-5* For separate papers, see this issue of the Bibliography

07 Textbooks (whole field)

0189 07.43; 43; 02Satija, M.P.- (Book review of) Mortimer, M.: Learn DeweyDecimal Classification (Edition 21). - Canberra: DocuMatrix,1998.- 132 p.- ISBN 0-876283-02-5 (Lang.: eng). - In: AsianLibraries, 8(1999)1, p. 389-390

0190 07.75; 756; 347Olson, H.A., Boll, J.J.- Subject access in online catalogs(Lang.: eng). - 2nd ed.. - Englewood, CO : Libraries Unlim-ited, 2001.- xv, 333 p.. - Refs. - ISBN 1-56308-800-2* Rev.ed.of Subject analysis in Online Catalogs / R.Aluri,D.A.Kemp, J.J.Boll

09 Standards, Guidelines

0191 09.26; 268National Information Standards Organization. - Codes forthe Representation of Languages for Information Interchange.-Bethesda, MD : NISO Press, 2001.- vii, 16p.. - (ANSI/NISOZ.39-53-2001). - ISBN 1-880124-52-1»» The newly revised standard of 3-letter codes for lan-guages is based on the MARC language codes and ISO 639-2/B

1 Theoretical Foundations and General Problems

11 Order and Knowledge OrganizationSee also: 0188, 0199

0192 111Jacob, E.K.- Augmenting human capabilities : classification ascognitive scaffolding (Lang.: eng). - In: Challenges in Knowl-edge Organization ...(See 0188), p. 38-44.- 23 refs.

0193 111González de Gómez, M.N.- Knowledge, communication, in-formation : intersubject links, institutional and technologicalmediations (Lang.: eng). - In: Challenges in Knowledge Or-ganization ...(See 0188), p. 50-53.- 8 refs.

0194 111Kobashi, N.Y., Smit, J.W., Tálamo, M.F.G.M.- Constitutionof the scientific domain of information science (Lang.: eng). -In: Challenges in Knowledge Organization ...(See 0188), p.80-85.- 4 refs.

0195 111García Gutiérrez, A.- Knowledge organization from a "Cul-ture of the Border" : towards a transcultural ethics of media-tion (Lang.: eng). - In: Challenges in Knowledge Organiza-tion ...(See 0188), p. 516-522.- 19 refs.

0196 111; 02Williamson, N.J.- (Book review of) Content organization inthe new millennium / ed.by A.Neelameghan, K.N.Prasad.-Bangalore : Sarada Ranganathan Endowment for Library Sci-ence, 2001.- 148p.- ISBN 81-900957-1-4 (Lang.: eng). - In:Knowledge Organization, 28(2001)4, p. 205-206


111

0197 119Andersen, J.- Ascribing cognitive authority to scolarly docu-ments : on the (possible)role of knowledge organization inscholarly communication (Lang.: eng). - In: Challenges inKnowledge Organization ...(See 0188), p. 28-37.- 18 refs.

12 Conceptology in Knowledge OrganizationSee also: 0214

0198 121Beghtol, C.- Universal concepts, cultural warrant, and cul-tural hospitality (Lang.: eng). - In: Challenges in KnowledgeOrganization ...(See 0188), p. 45-49.- 20 refs.

0199 121; 111Green, R.- Conceptual universals in knowledge organizationand representation (Lang.: eng). - In: Challenges in Knowl-edge Organization ...(See 0188), p. 15-27.- 28 refs.

0200 123Garciá de Quesada, M., Fuertes Olivera, P.A., Montero, S.-Propuesta de estructura definicional terminográfica en OntoTerm [Proposal for a structure for terminological definitionin Onto Term] (Lang.: spa). - In: Terminology, 8(2002)1, p.57-90.- 38 refs.

0201 124Oussalah, M., Giret, F., Khammaci, T.- A KR multi-hierarchies/multi-view model for the development of complexsystems (Lang.: eng). - In: Challenges in Knowledge Organi-zation ...(See 0188), p. 240-244.- 13 refs.

13 Mathematics in Knowledge Organization

0201a 131Pérez López, A., Moneda Corrochano, M.- Application ofthe Cantor set theory in making decisions about the collectionsdevelopment (Lang.: eng). - In: Challenges in KnowledgeOrganization ... (See ), p.181-185.- 7 refs.

14 Systems Theory and Knowledge Organization

0202 147; 751Guerrero-Bote, V.P., López-Pujalte, C., Faba Pérez, C.,Reyes Barragán, M.J., Zapico Alonso, .F.- Artificial neuralnetworks applied to information retrieval (Lang.: eng). - In:Challenges in Knowledge Organization ...(See 0188), p. 572-577.- 32 refs.

0203 147; 757Martínez, F., Martin, M.T., Rivas, V.M., Díaz, M.C.,Ureña, L.A.- Using neural networks for multiword recogni-tion in IR (Lang.: eng). - In: Challenges in Knowledge Or-ganization ...(See 0188), p. 559-564.- 14 refs.

0204 149Eíto Brun, R.- Software development and reuse as a knowl-edge management practice (Lang.: eng). - In: Challenges inKnowledge Organization ...(See 0188), p. 218-224.- 2 refs.

16 Science and Knowledge Organization

0205 162Shapiro, J.J.- Interdisciplinary knowledge integration and in-tellectual creativity (Lang.: eng). - In: Challenges in Knowl-edge Organization ...(See 0188), p. 100-106.- 19 refs.

2 Classification Systems and Thesauri (CS & T).Structure and Construction

21 General Problems of CS & TSee also: 0187, 0216

0206 211Tennis, J.T.- Subject ontogeny : subject access through timeand the dimensionality of classification (Lang.: eng). - In:Challenges in Knowledge Organization ...(See 0188), p. 54-59.- 8 refs.

0207 211Poli, R.- Framing information (Lang.: eng). - In: Challengesin Knowledge Organization ...(See 0188), p. 225-231.- 10refs.»» Distinction between semiotic, semantic and ontologicalclassifications

0208 211Williamson, N.J.- Knowledge integration and classificationschemes (Lang.: eng). - In: Challenges in Knowledge Organi-zation ...(See 0188), p. 332-337.- 13 refs.

0209 211Qin, J.- Evolving paradigms of knowledge represenatationand organization : a comparative study of classification,XML/DTD, and ontology (Lang.: eng). - In: Challenges inKnowledge Organization ...(See 0188), p. 465-471.- 9 refs.

0210 211Mai, J.E.- Is classification theory possible? : Rethinking classifi-cation research (Lang.: eng). - In: Challenges in KnowledgeOrganization ...(See 0188), p. 472-478.- 7 refs.

0211 211Andersen, J.- Communication technologies and the concept ofknowledge organization : a medium-theory perspective (Lang.:eng). - In: Knowledge Organization, 29(2002)1, p. 29-39.- 42refs.


112

0212 212McIlwaine, I.C.- Where have all the flowers gone? : An inves-tigation into the fate of some special classification schemes(Lang.: eng). - In: Challenges in Knowledge Organization...(See 0188), p. 479-486.- 17 refs.

0213 214; 918Aldana, J.F., Roldán, M.M., Gómez, A.C., Moreno, N.,Nebro, A.J.- Metadata functionality for semantic web integra-tion (Lang.: eng). - In: Challenges in Knowledge Organiza-tion ...(See 0188), p. 298-304.- 27 refs.

0214 221; 124Smith, T.R., Zeng, M.L.- Structured models of scientific con-cepts for organizing, accessing, and using learning materials(Lang.: eng). - In: Challenges in Knowledge Organization...(See 0188), p. 232-239.- 9 refs.

22 Structure and Elements of CS & TSee also: 0249

0215 224Green, R., Bean, C.A., Hudon, M.- Universality and basiclevel concepts (Lang.: eng). - In: Challenges in KnowledgeOrganization ...(See 0188), p. 311-317.- 12 refs.

0216 225; 212Orrico, E.G.D.- Metaphorical representation of the thematicidentity of social groups in the assistance of information re-trieval (Lang.: eng). - In: Challenges in Knowledge Organi-zation ...(See 0188), p. 552-558.- 12 refs.

0217 225; 357Tudhope, D., Binding, C., Blocks, D., Cunliffe, D.- Repre-sentation and retrieval in faceted systems (Lang.: eng). - In:Challenges in Knowledge Organization ...(See 0188), p. 191-197.- 18 refs.

0218 225; 752.3Broughton, V.- Facet analytical theory as a basis for a knowl-edge organization tool in a subject portal (Lang.: eng). - In:Challenges in Knowledge Organization ...(See 0188), p. 135-141.- 10 refs.

23 Construction of CS & T

0219 232Hjørland, B.- The methodology of constructing classificationschemes : a discussion of the state-of-the-art (Lang.: eng). - In:Challenges in Knowledge Organization ...(See 0188), p. 450-456.- 21 refs.

0220 232; 357Olson, H.A., Nielsen, J., Dippie, S.R.- Encyclopaedist ri-valry, classificatory commonality, illusory universality (Lang.:eng). - In: Challenges in Knowledge Organization ...(See0188), p. 457-464.- 8 refs.

0221 235Lykke Nielsen, M., Gjerluf Eslau, A.- Corporate thesauri -how to ensure integration of knowledge and reflection of di-versity (Lang.: eng). - In: Challenges in Knowledge Organi-zation ...(See 0188), p. 324-331.- 11 refs.

0222 237; 651/4Gu, H., [et al.] - A methodology for partitioning a vocabularyhierarchy into trees (Lang.: eng). - In: Artificial Intelligencein Medicine, 15(1999)1, p. 77-98.- Refs.

24 Relationships

0223 241Barrière, C.- Hierarchical refinement and representation ofthe causal relation (Lang.: eng). - In: Terminology, 8(2002)1,p. 91-111.- 22 refs.

0224 241Bertrand-Gastaldy, S.- (Book review of) Relationships in theorganization of knowledge / ed.by C.A.Bean and R.Green.-Dordrecht ; Boston: Kluwer Academic Publishers, 2001.- ix,232 p.- (Information and knowledge management; v.2). - ISBN0-7923-6813-4 (Lang.: eng). - In: Knowledge Organization,28(2001)4, p. 208-210

26 Notations, CodesSee also: 0191

0225 267Chachra, V.- Globalization and the Unicode Standard(Lang.: eng; fre). - In: Archive, Libraries and MuseumsConvergence : 24th Library Systems Seminar Paris, 12 - 14April 2000 / ed.by M.Witt and M.Ihadjadene.- Paris, Citédes sciences et de l'industrie, 2001, p. 99-105 [eng], p. 106-111 [fre]. - Refs.

27 Maintenance and Storage of CS & T

0226 273Madalli, D.P., Prasad, A.R.D.- VYASA : a knowledge repre-sentation system for automatic maintenance of analytico-synthetic scheme (Lang.: eng). - In: Challenges in KnowledgeOrganization ...(See 0188), p. 113-119.- 5 refs.

28 Compatibility of CS & T

0227 285Silva, N., Rocha, J.- Merging ontologies using a bottom-uplexical and structural approach (Lang.: eng). - In: Challengesin Knowledge Organization ...(See 0188), p. 60-66.- 22 refs.

0228 286Meo-Evoli, L., Negrini, G.- CoReC : a model for integratingclassification systems (Lang.: eng). - In: TKE '99 Terminol-ogy and knowledge engineering / ed.by P.Sandrini.- Vi-enna: TermNet, 1999, 293-306.- Refs.


113

0229 287; 757Hurtado, M.V., García, L., Parets, J.- Semantic views overheterogeneous and distributed data repositories : integration ofinformation system based on ontologies (Lang.: eng). - In:Challenges in Knowledge Organization ...(See 0188), p. 338-344.- 13 refs.

3 Classing and Indexing (Methodology)

32 Subject AnalysisSee: 0272, 0346

34 Automatic Classing and IndexingSee also: 0190, 0289

0230 344Downey, L.L., Tice, D.M.- A usability case study usingTREC and ZPRISE (Lang.: eng). - In: Information Process-ing and Managament, 35(1999)5, p. 589-603.- Refs.

0231 344Konings, E., Gramsbergen, E.- Ervaringen van deBibliotheek TU Delft : automatische onderwerpsondexeringvan een bibliotheekscatalogus [Experiences at Delft TechnicalUniversity: automatic subject indexing of a librarycatalogue] (Lang.: dut). - In: Informatie Professional,3(1999)10.p. 43-47.- Refs.

0232 344Leont'eva, N.N.- O soderzhanii informatsii [Identificationof information] (Lang.: ice; rus). - In: Nauchno-Tekhnicheskaya Informatsiya.Series 2, 8(1999)2, p. 9-14.-Refs.

0233 346Spitters, M., J.- Adjust : automatische thesauriële ontsluitingvan grote hoeveelheden krantenartikelen [Adjust: automatedindexing by thesaurus of large newspaper articles collec-tions] (Lang.: dut). - In: Informatie Professional,3(1999)10.p. 29-31.- Refs.

35 Manual and Automatic OrderingSee also: 0217, 0220, 0290, 0291

0234 357Cordeiro, M.I., Slavic, A. - Data models for knowledge or-ganization tools : evaluations and perspectives (Lang.: eng). -In: Challenges in Knowledge Organization ...(See 0188), p.127-134.- 23 refs.

0235 357Péis, E., Herrera-Viedma, E., Herrera, J.C.- On the evalua-tion of XML documents using fuzzy linguistic techniques(Lang.: eng). - In: Challenges in Knowledge Organization...(See 0188), p. 565-571.- 10 refs.

0236 357Pritchard, E.- XML : the future of web markup (Lang.: eng). -In: Information Research, 5(2000)2 [electr.]. - Refs.* http://panizzi.shef.ac.uk/elecdiss/edl0003/index.html

0237 357; 753Chaudiron, S., Ihanjadene, M., Role, F.- Authorial indexbrowsing in an XML digital library (Lang.: eng). - In: Chal-lenges in Knowledge Organization ...(See 0188), p. 358-363.-13 refs.

0238 357; 757Riggs, K.R.- XML and free text (Lang.: eng). - In: Journal ofthe American Society for Information Science and Tech-nology, 53(2002)6, p. 526-528.- Refs.

39 Evaluation of Classing and IndexingSee also: 0303

0239 396; 753Riesthuis, G.J.A.- Vrije tekst en gecontroleerd vocabulaires[Free text and controlled thesauri] (Lang.: dut). - In:Informatie Professional, 3(1999)10.p. 26-28.- Refs.

4 On Universal Classification Systems andThesauri

41 On Universal Systems in General

0240 411Rafferty, P.- The representation of knowledge in library classi-fication schemes (Lang.: eng). - In: Knowledge Organization,28(2001)4, p. 180-191.- 18 refs.

0241 411Sa'ari, C.Z.- Classification of sciences : a comparative study ofIhya'ulum al-din and al=Risalah al-laduniyyah (Lang.: eng). -In: Intellectual Discourse, 7(1999)1, p. 53-77.- Refs.

42 On the Universal Decimal ClassificationSee: 0246

43 On the Dewey Decimal Classification

0242 43Intner, S.S.- An interview with Dewey's Peter Paulson (Lang.:eng). - In: Technicalities, 19(1999)3, p. 2-3

0243 43; 02Satija, M.P.- (Book review of) Gupta, S.: Decimal Classifica-tion System: a bibliography for the period 1876 - 1994.- NewDeli: MD Publications Pvt Ltd., 1997.- ISBN 81-7533-058-9(Lang.: eng). - In: Asian Libraries, 8(1999)1, p. 344-345


114

48 On other Universal Classification Systems andThesauriSee also: 0344

0244 489; 663Dextre Clarke, S.G.- Planning controlled vocabularies for theUK public sector (Lang.: eng). - In: Challenges in KnowledgeOrganization ...(See 0188), p. 142-148.- 5 refs.

6 On Special Subjects CS & T

61 On CS & T in Logic, Mathematics and other FormalSciences

0245 614Coleman, A.- A classification of models (Lang.: eng). - In:Challenges in Knowledge Organization ...(See 0188), p. 86-92.- 13 refs.

0246 614; 42Herrera-Solana, V., Moya-Anegón, F.de - Graphical table ofcontents (GTOC) for library collections : the application ofUDC codes for the subject maps (Lang.: eng). - In: Challengesin Knowledge Organization ...(See 0188), p. 437-442.- 20refs.

65 On CS & T in Human Biological and PsychologicalSciences, Sport and HouseholdSee also: 0222, 0270

0247 651/4Lovis, C., [et al.] - Medical dictionaries for patient encodingsystems (Lang.: eng). - In: Artificial Intelligence in Medicine,14(1998),1/2, p. 201-214.- Refs.

66 On CS & T in Sociology, Politics, Social Policy, Law,Area Planning, Military Science, HistorySee also: 0244

0248 669Beghtol, C.- Knowledge representation and organization inthe Iter project : a web-based digital library for scholars of theMiddle Ages and Renaissance (http:\\iter.toronto.ca) (Lang.:eng). - In: Knowledge Organization, 28(2001)4, p. 170-179.-20 refs.

67 On CS & T in Economy, Management Science, Me-chanical Engineering, Building, Transport

0249 673; 225Campbell, G.- Chronotope and classification : how space-timeconfigurations affect the gathering of industrial statistical data(Lang.: eng). - In: Challenges in Knowledge Organization...(See 0188), p. 318-323.- 10 refs.

0250 673; 489Elichrigoity, F., Knott Malone, C.- Representing the globaleconomy : the North American Industry Classification System(Lang.: eng). - In: Challenges in Knowledge Organization...(See 0188), p. 345-350.- 19 refs.

69 On CS & T in Language, Literature, Music, Arts,Philosophy, ReligionSee also: 0186, 0293

0251 692Negrini, G., Zozi, P.- Ontological analysis of literary work ofart (Lang.: eng). - In: Challenges in Knowledge Organiza-tion ...(See 0188), p. 67-73.- 12 refs.

0252 692Saarti, J.- The analysis of the information process of fiction : aholistic approach to information processing (Lang.: eng). - In:Challenges in Knowledge Organization ...(See 0188), p. 74-79.- 16 refs.

0253 692Saarti, J.- Fiction indexing and the development of fictionthesauri (Lang.: eng). - In: Journal of Librarianship and In-formation Science, 31(1999)2, p85-92.- Refs.

7 Knowledge Representation by Language andTerminology

71 General Problems of Natural Language in Relationto Knowledge Organization

0254 711Mustafa el Hadi, W.- Terminology and information retrieval: new tools for new needs (Lang.: eng). - In: Challenges inKnowledge Organization ...(See 0188), p. 149-157.- 30 refs.

0255 715; 757Jones, K.S.- Information retrieval and artificial intelligence(Lang.: eng). - In: Artificial Intelligence, 114(1999)1/2, p.257-281.- Refs.

72 Semantics

0256 725Priss, U.- Alternatives to the "Semantic Web" : multi-strategyknowledge representation (Lang.: eng). - In: Challenges inKnowledge Organization ...(See 0188), p. 305-310.- 13 refs.

73 Automatic Language Processing

0257 733Marshman, E., Morgan, T., Meyer, I.- French patterns for ex-pressing concept relations (Lang.: eng). - In: Terminology,8(2002)1, p. 1-29.- 24 refs.


115

75 Online Retrieval Systems and Technologies

751 General and Theoretical ProblemsSee also: 0202, 0312

0258 751Blair, D.C.- The challenge of commercial document retrieval :Part I: Major issues, and a framework based on search exhaus-tivity, determinacy of representation and document collectionsize (Lang.: eng). - In: Information processing and manage-ment, 38(2002), p. 273-291.- 33 refs.

0259 751Blair, D.C.- The challenge of commercial document retrieval :Part II: A strategy for document searching based on identifi-able document partitions (Lang.: eng). - In: Informationprocessing and management, 38(2002), p. 293-304.- 10 refs.

0260 751Wondergem, B., Van Bommel, P., Van der Weide, T.-Information retrieval : een uitdagend onderzoeksgebied[Information retrieval: a challenging field for research](Lang.: dut). - In: Informatie Professional, 3(1999)10.p. 20-25.- Refs.

0261 751; 02Riley, B.- (Book review of) Information seeking in context:proceedings of an International Conference on Research in In-formation Needs, Seeking and Use in Different Contexts, 14-16August 1996, Tampere, Finland / ed.by P.Vakkari,R.Savolainen, B.Dervin.- London: Taylor Graham, 1997.-ISBN 0-947568-71-9 (Lang.: eng). - In: Asian Libraries,8(1999)1, p. 337-338

0262 751; 02Pinfield, S.- (Book review of) Large, A., Tedd, L.A., Hartley,R.J.: Information seeking in the online age: principles andpractice.- East Grinstead: Bowker Saur, 1999.- 336 p.- ISBN 1-85739-260-4 (Lang.: eng). - In: Program, 34(2000)1, p. 128-129.- Refs.

0263 751; 02Cullen, R.- (Book review of) Large, A., Tedd, L.A., Hartley,R.J.: Information seeking in the online age: principles andpractice.- East Grinstead: Bowker Saur, 1999.- 336 p.- ISBN 1-85739-260-4 (Lang.: eng). - In: Asian Libraries, 8(1999)1, p.386-387.- Refs.

752 Dialogue Systems. Interactive CataloguesSee also : 0218, 0280, 0313, 0316

0264 752Haas, S.W., Hert, C.A.- Finding information at theU.S.Bureau of Labor Statistics : over the barriers of scope, con-cept, and language mismatch (Lang.: eng). - In: Terminology,8(2002)1, p. 31-56.- 25 refs.

0265 752.2Holm, L.A.- Profiling the ALM sector (Lang.: eng; fre). - In:Archive, Libraries and Museums Convergence : 24th Li-brary Systems Seminar Paris, 12 - 14 April 2000 / ed.byM.Witt and M.Ihadjadene.- Paris, Cité des sciences et del'industrie, 2001, p. 62-72 [eng], 72-85 [fre]. - Refs.

0266 752.2Scolari, A.- Efficacia vs Facilita? [Effectiveness versus user-friendliness] (Lang.: ita). - In: Biblioteche Oggi, .17(1999)5,p. 18-26.- Refs.

0267 752.2; 756Shiri, A.A., Revie, C., Chowdhury, G.- Assessing the impactof user interaction with thesaural knowledge structures : aquantitative analysis framework (Lang.: eng). - In: Chal-lenges in Knowledge Organization ...(See 0188), p. 493-499.-11 refs.

0268 752.2; 756Caro Castro, C., Travieso Rodriguez, C.- Ariadne's thread :knowledge structures for browsing in OPACs (Lang.: eng). -In: Challenges in Knowledge Organization ...(See 0188), p.500-508.- 25 refs.

0269 752.3Kravchyna, V., Hasting, S.K.- Informational value of mu-seum web sites (Lang.: eng). - In: Firstmonday, 7(2002)2[electr.]. - 12 refs.* http://www.firstmonday.dk/issues/issue72/kravchyna/index.html

0270 752.3; 651/4Salvador Oliván, J.A., Angós Ullate, J.M., Fernández Ruiz,M.J.- Organization of the information about health resourceson the Internet (Lang.: eng). - In: Challenges in KnowledgeOrganization ...(See 0188), p. 198-204.- 8 refs.

0271 752.3; 879Herrera Morillas, J.L., Fernández Falero, M.- Informationand resources about bibliographic heritage on the Web sites ofthe Spanish universities (Lang.: eng). - In: Challenges inKnowledge Organization ...(See 0188), p. 291-297.- 2 refs.

0272 752; 321Hirsch, S.G.- Children's relevance criteria and informationseeking on electronic resources (Lang.: eng). - In: Journal ofthe American Society for Information Science, 50(1999)14,p. 1265-1283.- Refs.

753 On-line Access, Queries, Free Text SearchingSee also: 0237, 0239

0273 753Khoo, C.S.G., Ng, K., Ou, S.- An exploratory study of hu-man clustering of Web pages (Lang.: eng). - In: Challenges inKnowledge Organization ...(See 0188), p. 351-357.- 3 refs.


116

0274 753Cordón, O., Herrera-Viedma, E., Luque, M., Moya-Anegón, F.de, Zarco, C.- An inductive query by exampletechnique for extended Boolean queries based on simulated-annealing programming (Lang.: eng). - In: Challenges inKnowledge Organization ...(See 0188), p. 429-436.- 15 refs.

755 Problems of on-line systems. Types of searches

0275 755Fuhr, N.- Probabilistic Datalog : implementing logical infor-mation retrieval for advanced applications (Lang.: eng). - In:Journal of the American Society for Information Science,51(1999)2, p. 95-110.- Refs.

756 Classification and Thesaurus-based AccessSee also: 0190, 0267, 0268

0276 756Arsenault, C.- (Book review of) Olson, H.A., Boll, J.L.: Subjectanalysis in online catalogs.- 2nd ed.- Englewood, CO : Librar-ies Unlimited, 2001.- xv, 333 p.- ISBN 1-56308-800-2 (Lang.:eng). - In: Knowledge Organization, 28(2001)4, p. 206-208.-3 refs.

0277 756Huisman, F.- Anders zoeken met een classificatie [An alterna-tive method of searching with the aid of classification](Lang.: dut). - In: Informatie Professional, 3(1999)10, p. 49-53.- Refs.

0278 756Sieverts, E., Mastenbroek, O., Grygierczyk, N.- Eenuniform retrieval-systeem voor de Universiteit Utrecht [Auniform retrieval system for Utrecht University] (Lang.:dut). - In: Informatie Professional, 3(1999)10.p. 34-40.- Refs.

0279 756; 026Hjørland, B., Kyllesbech Nielsen, L.- Subject access points inelectronic retrieval (Lang.: eng). - In: Annual Review of In-formation Science and Technology, 35(2001), p. 249-298.-Ca 150 refs.

0280 756; 752.2Shiri, A.A., Revie, C., Chowdhury, G.- Thesaurus-assistedsearch term slection and query expansion : a review of user-centred studies (Lang.: eng). - In: Knowledge Organization,29(2002)1, p. 1-19.- 80 refs.

757 Expert Systems in Searching. Search EnginesSee also: 0203, 0229, 0238, 0255, 0317-0319

0281 757Furner, J.- A unifying model of document relatedness for hy-brid search engines (Lang.: eng). - In: Challenges in Knowl-edge Organization ...(See 0188), p. 245-250.- 13 refs.

0282 757Herrera-Viedma, E., Cordón, O., Herrera, J.C., Luque, M.-An IRS based on multi-granular linguistic information(Lang.: eng). - In: Challenges in Knowledge Organization...(See 0188), p. 372-378.- 13 refs.

0283 757Cuesta, P., Gómez, A.M., Rodriguez, F.J.- Using agents forInformation Retrieval (Lang.: eng). - In: Challenges inKnowledge Organization ...(See 0188), p. 379-385.- 12 refs.

0284 757López-Huertas, M.J., Barité Roqueta, M.G.- Knowledge rep-resentation and organization of gender studies on the Internet :towards integration (Lang.: eng). - In: Challenges in Knowl-edge Organization ...(See 0188), p. 393-403.- 6 refs.

0285 757De Campos, L.M., Fernández-Luna, J.M., Huete, J.F.- Man-aging documents with Bayesian belief networks : a brief surveyof applications and models (Lang.: eng). - In: Challenges inKnowledge Organization ...(See 0188), p. 443-449.- 20 refs.

0286 757Hipola, P., Vargas-Quesada, B., Montes, A.- Description yevaluacion de agentes multibuscadores [Description andevaluation of lulti-searcher agents] (Lang.: spa). - In: Profes-sional de la Informacion, 8(1999)11, p. 15-24.- Refs.

0287 757Hipola, P., Vargas-Quesada, B.- Agentes intelligantes : defin-tion y tipologia - los agentes de informacion [Intelligentagents: definition and categories - information agents](Lang.: spa). - In: Professional de la Informacion, 8(1999)4,p. 13-21.- Refs.

0288 757Dunning, A.- Do we still need search engines? (Lang.: eng). -In: Ariadne, (1999)22 [electr.]. - Refs.* http://www.ariadne.ac.uk/issue22/dunning/

0289 757; 348Fong, A.C.M.- Mining a Web citation database for documentclustering (Lang.: eng). - In: Applied Artificial Intelligence,16(2002)4, p. 283-292.- Refs.

0290 757; 357Chinenyanga, T.T., Kushmerick, N.- An expressive and effi-cient language for XML information retrieval (Lang.: eng). -In: Journal of the American Society for Information Sci-ence and Technology, 53(2002)6, p. 438-453.- Refs.

0291 757; 357Cohen, S., Kanza, Y., Kogan, Y., Sagiv, Y., Nutt, W., Sere-brenik, A.- EquiX - a search and query language for XML(Lang.: eng). - In: Journal of the American Society for In-


117

formation Science and Technology, 53(2002)6, p. 454-466.-Refs.

758 Online Systems in Subject Fields

0292 758-82Péis, E., Ruiz, A., Muñoz Fernández, F.J., Alba Quiñnones,F.- Practical method to code archive finding aids in Internet(Lang.: eng). - In: Challenges in Knowledge Organization...(See 0188), p. 205-210.- 8 refs.

0293 758-93; 693Byrd, D., Crawford, T.- Problems of music information re-trieval in the real world (Lang.: eng). - In: Information proc-essing and management, 38(2002), p. 249-272.- Refs.

759 Evaluation of Online Information Retrieval

0294 759Banwell, L.- Developing an evaluation framework for a su-pranational digital library (Lang.: eng). - In: Challenges inKnowledge Organization ...(See 0188), p. 509-515.- 7 refs.»» The TEL (The European Library)project.

0295 759Blanco, I., Martin-Bautista, M.J., Sánchez, D., Vila, .M.A.-Fuzzy logic for measuring information retrieval effectiveness(Lang.: eng). - In: Challenges in Knowledge Organization...(See 0188), p. 578-585.- 13 refs.

0296 759Redalen, A., Miller, N.- Evaluating Website modifications atthe National Library of Medicine through search log analysis(Lang.: eng). - In: D-Lib Magazine, 6(2000)1 [electr.]. - Refs.* http://www.dlib.org/dlib/january00/redalen/01redalen.html

77 Problems of Terminology

0297 771; 02Bowker, L.- Handbook of terminology management: Volume2 / ed.by S.E.Wright and G.Budin.- Amsterdam: John Ben-jamin, 2001.- p. 371-920.- ISBN 90-272-2155-3.- In: Termi-nology, 7(2001)2, p. 287-290

0298 773.4Collier, N., Nobata, C., Tsujii, J.- Automatic acquisition andclassification of terminology using a tagged corpus in the mo-lecular biology domain.- In: Terminology, 7(2001)2, p. 239-257.- 27 refs.

79 Multilingual Systems and Translation

0299 791Neelameghan, A., Iyer, H.- Some patterns of informationpresentation, organization, and indexing for communicationacross cultures and faiths (Lang.: eng). - In: Challenges inKnowledge Organization ...(See 0188), p. 539-545.- 2 refs.

0300 797Frâncu, V.- Language-independent structure and multilingualinformation access (Lang.: eng). - In: Challenges in Knowl-edge Organization ...(See 0188), p. 404-411.- 1 ref.

0301 797Hudon, M.- Accessing documents and information in a worldwithout frontiers (Lang.: eng). - In: Indexer, 21(1999)4, p.156-159.- Refs.

0302 797; 844Garcia Jiménez, A., Diaz Esteban, A., Gervás, P.- Knowl-edge organization in a multilingual system for the personaliza-tion of digital news services : how to integrate knowledge(Lang.: eng). - In: Challenges in Knowledge Organization...(See 0188), p. 386-392.- 14 refs.

8 Applied Classing and Indexing

81 General Problems, Subject Catalogues, Guidelines,Rules, Indexes (see also 34)See: 0303

84 Classing and Indexing of Primary LiteratureSee also: 0302

0303 841; 392; 811Shoman, S., Kedar, R.- The subject cataloging of monographswith the use of a thesaurus (Lang.: eng). - In: Challenges inKnowledge Organization ...(See 0188), p. 173-180.- 7 refs.

0304 844Zarri, G.P.- Indexing and querying of narrative documents, aknowledge representation approach (Lang.: eng). - In: Chal-lenges in Knowledge Organization ...(See 0188), p. 93-99.- 4refs.

0305 844Oliveira, M.O.E.de - Knowledge representation from Ama-zonian narratives : culture and oral tradition (Lang.: eng). -In: Challenges in Knowledge Organization ...(See 0188), p.546-551.- 8 refs.

0306 847Hutchinson, T.- Punti di accesso agli archivi per soggetto :L'esperienza nord amaricana [Access points to subject-basedarchives: the North American experience] (Lang.: ita). - In:Archivi and Computer, 7(1997)6, p. 375-385.- Refs.


118

0307 848; 918Smith, M.- DFAS : the distributed finding aid search system(Lang.: eng). - In: D-Lib Magazine, 6(2000)1 [electr]. - Refs.* http://www.dlib.org/dlib/january00/01smith.html

85 Back-of-the-Book Indexing

0308 856Bell, H.K.- Kiss and tell and index (Lang.: eng). - In: Indexer,21(1999)4, p. 180-181.- Refs.

0309 857Mulvany, N.C.- Software tools for indexing : revisited (Lang.:eng). - In: Indexer, 21(1999)4, p. 160-163.- Refs.

86 Secondary Literature Classification and Indexing

0309a 864; 245Barrueco, J.M., Inglada, V.J.- Reference linking in economics: the CitEc project (Lang.: eng). - In: Challenges in Knowl-edge Organization ... (See @@@), p.251-257.- 7 refs.

87 Classing and Indexing of Non-book MaterialsSee also: 0271

0310 871Naumis Peña, C.- Images and words (Lang.: eng). - In: Chal-lenges in Knowledge Organization ...(See 0188), p. 120-126.-10 refs.

0311 871Ciocca, G., Schettini, R.- A relevance feedback mechanismfor content-based image retrieval (Lang.: eng). - In: Informa-tion Processing and Management, 35(1999)5, p. 605-632.-Refs.

0312 871; 751; 02Bruza, P.D.- (Book review of) Maubury, M.T.: Intelligent mul-timedia information retrieval.- Cambridge, MA: MIT Press,1997.- 478 p. (Lang.: eng). - In: Artificial Intelligence inMedicine, 14(1998)1/2, p. 235-236

0313 871; 752Iyer, H., Keefe, J.M.- The WordNet as an auxilary resource tosearch visual image database in architecture (Lang.: eng). - In:Challenges in Knowledge Organization ...(See 0188), p. 186-190.- 2 refs.

0314 875Pejtersen, A.M., Albrechtsen, H.- Models for collaborativeintegration of knowledge (Lang.: eng). - In: Challenges inKnowledge Organization ...(See 0188), p. 412-421.- 15 refs.

0315 879Henninger, M.- What makes a good Web index? (Lang.: eng).- In: Indexer, 21(1999)4, p. 182-183.- Refs.

0316 879; 752.3Zins, C.- Models for classifying Internet resources (Lang.: eng).- In: Knowledge Organization, 29(2002)1, p. 20-28.- 25 refs.

0317 879; 757Assadi, H., Beauvisage, T.- A comparative study of six French-language web directories (Lang.: eng). - In: Challenges inKnowledge Organization ...(See 0188), p. 271-278.- 18 refs.

0318 879; 757Kwasnik, B.H.- Commercial Websites and the use of classifica-tion schemes : the case of Amazon.com (Lang.: eng). - In:Challenges in Knowledge Organization ...(See 0188), p. 279-285.- 1 ref.

0319 879; 757Serrano Cobos, J., Quintero Orta, A.- Design, developmentand management of an information recovery system for anInternet Website : from documentary theory to practice (Lang.:eng). - In: Challenges in Knowledge Organization ...(See0188), p. 286-290.- 6 refs.

88 Classing and Indexing in Subject Fields

0320 88-71/2Van der Walt, M.S.- An integrated model for the organizationof electronic information/knowledge in small, medium andmicro enterprises (SMMEs)in South Africa (Lang.: eng). - In:Challenges in Knowledge Organization ...(See 0188), p. 211-217.- 9 refs.

0321 89JAPOgawa, Y., Matsuda, T.- Overlapping statistical segmentationfor effective indexing of Japanese text (Lang.: eng). - In: In-formation Processing and Managament, 35(1999)4, p. 463-480.- Refs.

0322 89KORLee, J.H., Cho, H.Y., Park, H.R.- n-Gram-based indexing forKorean text retrieval (Lang.: eng). - In: Information Process-ing and Managament, 35(1999)4, p. 427-441.- Refs.

9 Knowledge Organization Environment

91 Professional and Organizational Problems in GeneralSee also: 0213, 0307

0323 912Fernández-Molina, J.C., Guimaraes, J.A.C.- Ethical aspectsof knowledge organization and representation in the digitalenvironment : their articulation in professional codes of ethics(Lang.: eng). - In: Challenges in Knowledge Organization...(See 0188), p. 487-492.- 9 refs.


119

0324 918Polanco, X.- Clusters, graphs, and networks for analysingInternet-Web supported communication within a virtualcommunity (Lang.: eng). - In: Challenges in Knowledge Or-ganization ...(See 0188), p. 364-371.- 14 refs.

0325 918Smith, M.- DFAS : the distributed finding aid search system(Lang.: eng). - In: Ariadne, (1999)22 [electr]. - Refs.* http://www.ariadne.ac.uk.issue22/metadata/

94 Bibliographic Control. Bibliographic Records

941 Bibliographic Control. Bibliography as Discipline

0326 941Grimaldi, T.- The object of cataloguing (Lang.: eng; ita). - In:Seminario FRBR: Functional requirements for biblio-graphic records : reguisiti funzionali per record bibliografici: Florence, 27-28 January 2000 / proceedings ed.byM.Guerrini.- Roma: Associazione italiana biblioteche,2000.- ISBN 88-7812-067-7, p. 64-77.- Refs.* Also: http://aib.it/aib/sezioni/toscana/conf/frbr/

0327 941Smiraglia, R.P.- Crossing cultural boundaries : perspectives onthe popularity of works (Lang.: eng). - In: Challenges inKnowledge Organization ...(See 0188), p. 530-538.- 5 refs.

0328 941O'Neill, E.T., Lavoie, B.F.- Bibliographic control for the Web(Lang.: eng). - In: Serials Librarian, 37(2000)3, p. 53-69.-Refs.

944 Bibliographic records. Functions of catalogues andbibliographical databases

0329 944Byrum, J.D., Madison, O.M.A.- Reflections on the goals, con-cepts and recommendations of the IFLA study on FunctionalRequirements of Bibliographic Records (Lang.: eng; ita). - In:Seminario FRBR: Functional requirements for biblio-graphic records : reguisiti funzionali per record bibliografici: Florence, 27-28 January 2000 / proceedings ed.byM.Guerrini.- Roma: Associazione italiana biblioteche,2000.- ISBN 88-7812-067-7, p. 12-51.- Refs.* Also: http://aib.it/aib/sezioni/toscana/conf/frbr/byrma-dis.htm

0330 944Guerrini, M.- The functions of the catalogue from ICCP toFRBR (Lang.: eng; ita). - In: Seminario FRBR: Functionalrequirements for bibliographic records : reguisiti funzionaliper record bibliografici : Florence, 27-28 January 2000 /proceedings ed.by M.Guerrini.- Roma: Associazione ital-iana biblioteche, 2000.- ISBN 88-7812-067-7, p. 52-63.- Refs.* Also: http://aib.it/aib/sezioni/toscana/conf/frbr/

0331 944Le Boeuf, P.- The impact of the FRBR model on the future re-visions of the ISBDs : a challenge for the IFLA Section on Cata-loging (Lang.: eng). - In: International Cataloguing and Bib-liographic Control, 31(2002)1, p. 3-6.- 16 refs.

0332 944Le Boeuf, P.- Towards a common conceptual model for (A)LM/ report prep. (Lang.: eng; fre). - In: Archive, Libraries andMuseums Convergence : 24th Library Systems SeminarParis, 12 - 14 April 2000 / ed.by M.Witt and M.Ihadjadene.-Paris, Cité des sciences et de l'industrie, 2001, p159-162[eng], 163-167 [fre]. - Refs.

0333 944Van der Starre, J.- Walking over the ALM (Lang.: eng; fre). -In: Archive, Libraries and Museums Convergence : 24thLibrary Systems Seminar Paris, 12 - 14 April 2000 / ed.byM.Witt and M.Ihadjadene.- Paris, Cité des sciences et del'industrie, 2001, p. 17-29 [eng], 30-42 [fre]. - Refs.

0334 944Crofts, N.- Introduction to the ICOM / CIDOC ConceptualReference Model (Lang.: eng; fre). - In: Archive, Libraries andMuseums Convergence : 24th Library Systems SeminarParis, 12 - 14 April 2000 / ed.by M.Witt and M.Ihadjadene.-Paris, Cité des sciences et de l'industrie, 2001, p168-177[eng], 178-189 [fre]. - Refs.

0335 944Fattahi, R., Parirokh, M.- Restructuring the bibliographic re-cord for better organization, management, and representationof knowledge in the global online environment : a new ap-proach (Lang.: eng). - In: Challenges in Knowledge Organi-zation ...(See 0188), p. 107-112.- 12 refs.

0336 944Riesthuis, G.J.A., Žumer, M.- The Functional Requirementsfor Bibliographic Records and knowledge organization (Lang.:eng). - In: Challenges in Knowledge Organization ...(See0188), p. 165-172.- 7 refs.

0337 944Carlyle, A., Fusco, L.M.- Equivalence in Tillett's biblio-graphic relationships taxonomy : a revision (Lang.: eng). - In:Challenges in Knowledge Organization ...(See 0188), p. 258-263.- 5 refs.

0338 944Frías, J.A., Ríos Hilario, A.B.- Visibility and invisibility ofthe kinship relationship in bibliographic families of the librarycatalogue (Lang.: eng). - In: Challenges in Knowledge Or-ganization ...(See 0188), p. 264-270.- 18 refs.


120

0339 944Smiraglia, R.P.- Works as signs, symbols, and canons : the epis-temology of the work (Lang.: eng). - In: Knowledge Organi-zation, 28(2001)4, p. 192-202.- 45 refs.

947 Interface and displays for Bibliographic or ArchivalRecords

0340 947Lee, H.L., Carlyle, A.- Academic library gateways to onlineinformation : a taxonomy of organizational structures (Lang.:eng). - In: Challenges in Knowledge Organization ...(See0188), p. 158-164.- 4 refs.

949 Authority control

0341 949Bourdon, F.- Functional requirements and numbering ofauthority records (FRANAR): to what extent can authoritycontrol be supported by technical means? (Lang.: eng). - In: In-ternational Cataloguing and Bibliographic Control,31(2002)1, p. 6-9.- 16 refs.

95 Education and Training in Knowledge OrganizationSee also: 0347

0342 958Slavič, A.- Teaching classification to fit a modern and sustain-able LIS curriculum : the case of Croatia (Lang.: eng). - In: In-ternational Cataloguing and Bibliographic Control,31(2002)1, p. 13-15.- 7 refs.

0343 958Oyler, P.G.- Teaching classification in the 21st century(Lang.: eng). - In: International Cataloguing and Biblio-graphic Control, 31(2002)1, p. 16-17.- Refs.

97 Economic Aspects of Knowledge Organization

0344 977; 485Becht, M.- Der Aufwand für die kooperative Sacherschliessungnach der RSWK in der Universitätsbibliothek Tübingen 1995-1998 [The work on cooperative subject cataloguing withRSWK at the University of Tübingen 1995-1998] (Lang.:ger). - In: Zeitschrift für Bibliothekswesen und Bibliog-raphie, 46(1999)5, p. 375-388.- Refs.

98 User Studies

0345 982King, C., Marwick, D.H., Williams, M.H.- The importanceof context in resolving of conflichts when sharing user profiles(Lang.: eng). - In: Challenges in Knowledge Organization...(See 0188), p. 523-529.- 8 refs.

0346 982; 321López-Pujalte, C., Guerrero Bote, V.P., Moya-Anegón, F.de- Evaluation of the application of genetic algorithms to rele-vance feedback (Lang.: eng). - In: Challenges in KnowledgeOrganization ...(See 0188), p. 422-428.- 16 refs.

0347 986; 959Obermeier, F.- Schlagwortsuche in einem lokalen OPAC amBeispiel de Universitätsbibliothek Eichstatt [Subject headingsearch in a local OPAC as exemplified by the UniversityLibrary of Eichstatt] (Lang.: ger). - In: BibliotheksforumBayern, 27(1999)3, p. 296-319.- Refs.

Knowl. Org. 29(2002)No.2Personal Author Index 29(2002)

121

Personal Author Index

Alba Quiñnones, F.0292

Albrechtsen, H. 0314Aldana, J.F. 0213Andersen, J. 0186,

0197, 0211Angós Ullate, J.M.

0270Arsenault, C. 0276Assadi, H. 0317Banwell, L. 0294Barité Roqueta, M.G.

0284Barrière, C. 0223Barrueco, J.M. 0309aBean, C.A. 0215Beauvisage, T. 0317Becht, M. 0344Beghtol, C. 0198,

0248Bell, H. K. 0308Bertrand-Gastaldy, S.

0224Binding, C. 0217Blair, D.C. 0258, 0259Blanco, I. 0295Blocks, D. 0217Boll, J.J. 0190Bourdon, F. 0341Bowker, L. 0297Broughton, V. 0218Bruza, P.D. 0312Byrd, D. 0293Byrum, J.D. 0329Campbell, G. 0249Carlyle, A. 0337, 0340Caro Castro, C. 0268Chachra, V. 0225Chaudiron, S. 0237Chinenyanga, T.T.

0290Cho, H.Y. 0322Chowdhury, G. 0267,

0280Ciocca, G. 0311Cohen, S. 0291Coleman, A. 0245Collier, N. 0298Cordeiro, M.I. 0234

Cordón, O. 0274,0282

Crawford, T. 0293Crofts, N. 0334Cuesta, P. 0283Cullen, R. 0263Cunliffe, D. 0217De Campos, L.M.

0285Dextre Clarke, S.G.

0244Diaz Esteban, A. 0302Díaz, M.C. 0203Dippie, S.R. 0220Downey, L.L. 0230Dunning, A. 0288Eíto Brun, R. 0204Elichrigoity, F. 0250Faba Pérez, C. 0202Fattahi, R. 0335Fernández Falero, M.

0271Fernández Ruiz, M.J.

0270Fernández-Luna, J.M.

0285Fernández-Molina,

J.C. 0323Fong, A.C.M. 0289Frâncu, V. 0300Frías, J.A. 0338Fuertes Olivera, P.A.

0200Fuhr, N. 0275Furner, J. 0281Fusco, L.M. 0337Garciá de Quesada, M.

0200García Gutiérrez, A.

0195Garcia Jiménez, A.

0302García, L. 0229Gervás, P. 0302Giret, F. 0201Gjerluf Eslau, A. 0221Gómez, A.C. 0213Gómez, A.M. 0283González de Gómez,

M.N. 0193

Gramsbergen, E. 0231Green, R. 0199, 0215Grimaldi, T. 0326Grygierczyk, N. 0278Gu, H. 0222Guerrero-Bote, V.P.

0202, 0346Guerrini, M. 0330Guimaraes, J.A.C.

0323Haas, S.W. 0264Hasting, S.K. 0269Henninger, M. 0315Herrera Morillas, J.L.

0271Herrera, J.C. 0235,

0282Herrera-Solana, V.

0246Herrera-Viedma, E.

0235, 0274, 0282Hert, C.A. 0264Hipola, P. 0286, 0287Hirsch, S.G. 0272Hjørland, B. 0219,

0279Holm, L.A. 0265Hope, H.A. 0190Hudon, M. 0215,

0301Huete, J.F. 0285Huisman, F. 0277Hurtado, M.V. 0229Hutchinson, T. 0306Ihanjadene, M. 0237Inglada, V.J. 0309aIntner, S.S. 0242Iyer, H. 0299, 0313Jacob, E.K. 0192Jones, K.S. 0255Justice, A. 0187Kanza, Y. 0291Kedar, R. 0303Keefe, J.M. 0313Khammaci, T. 0201Khoo, C.S.G. 0273King, C. 0345Knott Malone, C.

0250Kobashi, N.Y. 0194

Kogan, Y. 0291Konings, E. 0231Kravchyna, V. 0269Kushmerick, N. 0290Kwasnik, B.H. 0318Kyllesbech Nielsen, L.

0279Lavoie, B.F. 0328Le Boeuf, P. 0331,

0332Lee, H.L. 0340Lee, J.H. 0322Leont'eva, N.N. 0232Lopez-Huertas, M.J.

0188, 0284López-Pujalte, C.

0202, 0346Lovis, C. 0247Luque, M. 0274, 0282Lykke Nielsen, M.

0221Madalli, D.P. 0226Madison, O.M.A.Mai, J.E. 0210Marshman, E. 0257Martin, M.T. 0203Martin-Bautista, M.J.

0295Martínez, F. 0203Marwick, D.H. 0345Mastenbroek, O. 0278Matsuda, T. 0321McIlwaine, I.C. 0212Meo-Evoli, L. 0228Meyer, I. 0259Miller, N. 0296Moneda Corrochano,

M. 0201aMontero, S. 0200Montes, A. 0286Moreno, N. 0213Morgan, T. 0258Moya-Anegón, F.de.

0246, 0274, 0346Mulvany, N.C. 0309Muñoz-Fernández,

F.J. 0188, 0292Mustafa el Hadi, W.

0254Naumis Peña, C. 0310

Nebro, A.J. 0213Neelameghan, A.

0299Negrini, G. 0228,

0251Ng, K. 0273Nielsen, J. 0220Nobata, C. 0298Nutt, W. 0291Obermeier, F. 0347Ogawa, Y. 0321Oliveira, M.O.E.de

0305Olson, H.A. 0220O'Neill, E.T. 0328Orrico, E.G.D. 0216Ou, S. 0273Oussalah, M. 0201Oyler, P.G. 0343Parets, J. 0229Parirokh, M. 0335Park, H.R. 0323Péis, E. 0235, 0292Pejtersen, A.M. 0314Pérez López, A. 0201aPinfield, S. 0262Polanco, X. 0324Poli, R. 0207Prasad, A.R.D. 0226Priss, U. 0256Pritchard, E. 0236Qin, J. 0209Quintero Orta, A.

0319Rafferty, P. 0240Redalen, A. 0296Revie, C. 0267, 0280Reyes Barragán, M.J.

0202Riesthuis, G.J.A.

0239, 0336Riggs, K.R. 0238Riley, B. 0261Ríos Hilario, A.B.

0338Rivas, V.M. 0203Rocha, J. 0227Rodriguez, F.J. 0283Roldán, M.M. 0213Role, F. 0237

Knowl. Org. 29(2002)No.2Personal Author Index 29(2002)

122

Ruiz, A. 0292Sa'ari, C.Z. 0241Saarti, J. 0252, 0253Sagiv, Y. 0291Salvador Oliván, J.A.

0270Sánchez, D. 0295Satija, M.P. 0189,

0243Schettini, R. 0311Scolari, A. 0266Serebrenik, A. 0291

Serrano Cobos, J.0319

Shapiro, J.J. 0205Shiri, A.A. 0267, 0280Shoman, S. 0303Sieverts, E. 0278Silva, N. 0227Slavic, A. 0234, 0342Smiraglia, R.P. 0327Smiraglia, R.P. 0339Smit, J.W. 0194Smith, M. 0307, 0325

Smith, T.R. 0214Spitters, M. 0233Tálamo, M.F.G.M.

0194Tennis, J.T. 0206Tice, D.M. 0231Travieso Rodriguez,

C. 0268Tsujii, J. 0298Tudhope, D. 0217Ureña, L.A. 0203Van Bommel, P. 0260

Van der Starre, J.0333

Van der Walt, M.S.0320

Van der Weide, T.0260

Van Gent, J. 0233Vargas-Quesada, B.

0286, 0287Vila, M.A. 0295Williams, M.H. 0345

Williamson, N.J.0196, 0208

Wondergem, B. 0260Zapico Alonso, .F.

0202Zarco, C. 0274Zarri, G.P. 0304Zeng, M.L. 0214Zins, C. 0316Zozi, P. 0251Žumer, M. 0336