Newsletter of the Section on Classification and Indexing

No. 19
June 1999

To all members of the Section

The Chair, Secretary/Treasurer and Standing Committee members wish to express their appreciation of your support for the furthering of subject access to information in all its forms. In addition, I would like to say a personal thank you to outgoing members of the Committee, Poori Sultani (National Library of the Islamic Republic of Iran, Tehran, Iran), Pilar Benedito Castellote, (Biblioteca Naçional, Madrid, Spain), Elisa Grignani (Università degli Studi di Parma, Italy), and welcome the two newcomers, Jesus Jimenez Pelayo (Biblioteca de Andalucia, Granada, Spain) and P. Vidanapathirana (University of Sri Jayawardenepura, Negeguda, Sri Lanka) whom we look forward to meeting in Bangkok.

Ia McIlwaine
Chairperson
IFLA Section on Classification and Indexing

Standing Committee and Open Programme meetings for the Bangkok Conference

Saturday, August 21st 1999, 12.00-2.50, Standing Committee I
Monday, August 23rd 1999, 9.00-10.25, Division of Bibliographic Control Open Forum
Tuesday, August 24th, 1999, 12.30-3.00, Section Open Programme
Friday, August 27th, 1999, 8.30-10.20, Standing Committee II

Standing Committee of the Section on Classification and Indexing

The changes noted above will take place in Bangkok, when the new committee takes over.

The Standing Committee has one corresponding member: Mandana Sadigh-Behzadi (National Library of Iran, Tehran, Iran) and one observer: Marie-France Plassard (IFLA UBCIM Programme, Die Deutsche Bibliothek, Frankfurt am Main, Germany)

Abstracts of papers for the open programme of the Section at Bangkok

Lois Mai CHAN (School of Library and Information Science, University of Kentucky, USA) Xia LIN (College of Information Science and Technology, Drexel University, Philadelphia, USA)
Marcia Lei ZENG (School of Library and Information Science, Kent State University, USA)

This is a three-part presentation on recent efforts in imposing structure on, and in improving multilingual subject access to, Web resources.

1. Classificatory Structures on the Web (Lois Mai Chan)
   As Web resources multiply at a breathtaking rate, there has been an urgent need for effective mechanisms to organize them for efficient retrieval. With the increasing volume of Internet resources, we see a renewed interest in using classificatory structures to organize directories for more efficient knowledge discovery and retrieval. Such uses include directories based on existing classification schemes, e.g., NetFirst based on DDC and CyberStacks based on LCC; subject hierarchies used by popular search engines and directories such as Yahoo! and Lycos; and various classificatory structures used by many academic and research libraries to organize selected Web resources for assisted access by their users. There are also efforts in using hierarchical structures to organize Web sites on specific topics or for specific purposes and even for individual users. As an example of the last category, we will present a new device for personalized knowledge organization and access to Web resources.

2. Knowledge Class - An Example of Structural Access to the Web (Xia Lin and Lois Mai Chan)
   Knowledge Class was developed as an attempt to provide a mechanism by which information professionals can assist users in organizing, retrieving, and maintaining links to Web resources. It contains two basic components: an organizing framework and an interface for access and retrieval. The organizing framework is a classified mini-thesaurus, consisting of a hierarchically structured collection of terms on a specific topic or a particular discipline, such as investment, the solar system, diabetes, information science, etc. The emphasis of this framework is on the structure of information and the semantic relationships among terms, topics, branches of subject areas, etc. The interface serves as an interactive mechanism between the user and the terms in the organized framework as well as between the user and the Web resources. Through this device, the user can initiate searches by selecting pre-stored terms or search strategies or connect to specific sites previously discovered and stored. In the authoring mode, the mechanism will also provide functions to let the user add controlled vocabulary or free-text terms to the hierarchical framework, collect and store new links found during the retrieval process, and expand or modify the hierarchical structure to accommodate additional levels and branches.

   Improving subject browsing and the precision of retrieval are the two main goals of our research on Knowledge Class. In the first stage of our research¹, we introduced the mini-thesaurus-like knowledge class. We emphasized that (a) a knowledge structure can be built on principles of classification and knowledge organizing, (b) the knowledge structure should be seamlessly integrated with search engines for access to Web resources, and (c) an easy-to-use graphical interface should be constructed to support user interactions with the structure and the relevant resources discovered and retrieved through search engines.

   In this presentation, we will discuss our continuous efforts in the development of Knowledge Class. More examples of Knowledge Class will be demonstrated, including different ways to handle synonyms and homonyms, etc. General guidelines that help users construct their own knowledge class will be presented. A new Java interface that improves data handling and user interaction will be described. With the guidelines and the new interface, most information professionals will be able to develop knowledge classes tailored to the needs of the patrons they serve. Such efforts will help provide structural access to the unstructured web. The audience will be encouraged to develop their own knowledge classes by using free software necessary to build the knowledge class.

   References: 1. Lin, X. & Chan, L. M. (1997). Knowledge Class -- A dynamic structure for subject access on the web. Proceedings of the 8th ASIS SIG/CR Classification Research Workshop. (November 1, Washington, D.C.). pp. 31-40.

3. WWW Search Engines' Subject Directories in the Globalization and Localization Process (Marcia Lei Zeng) The World Wide Web search engines in 1998 have been involved in a competition of globalization and localization. To serve multi-lingual and multi-cultural populations all over the world, major search engines, such as AltaVista, Excite, HotBot, InfoSeek, and Yahoo! have developed new services including regional editions, search-by-language/region features, translation of search query or webpages, etc. This presentation will provide an overview of the approaches used by major search engines to ensure both compatibility and localization of the regional subject directories. The author will analyze the impact of society and culture on the setup of subject directories such as the order of subject categories, adjustment of hierarchical relationships, and the display interface. The presentation will also demonstrate the differences between Web-based subject directories and conventional subject catalogs and directories.

PROBLEMS OF DEVELOPMENT OF THE NATIONAL SUBJECT AUTHORITY FILE

Irina TSVETKOVA (Deputy Director, National Library of Russia, St Petersburg)
Julia SELIVANOVA (National Library of Russia, St Petersburg)

In recent years Russian libraries have taken an active part in various regional, national and international projects and programs on corporate cataloguing. The effectiveness of these projects and programmes is determined both by technical and software support, on the one hand, and by linguistic support, on the other hand. By linguistic support is meant, above all, common or compatible formats for data representation (bibliographic, authority, classification, etc.), as well as a compatible means for subject access to the information. In Russia problems of representation and exchange of bibliographic and authority data are basically solved: Russian communication formats for bibliographic and authority data have been developed and are being inculcated. With regard to the second group of problems, i.e., development of compatible means for subject access, these problems are yet to be resolved. There is a variety of types and kinds of subject access languages used by Russian libraries: UDC, LBC, different versions of subject headings languages (descriptor languages, etc.).

The most important problem for the Russian libraries is that of development of a common subject authority file (SAF) and common subject indexing rules. The significance of this problem is determined by the following:

1. Subject headings language is most close to natural language and so it is most convenient for the user

2. Subject headings language permits reaction to changes in terminology and vocabulary of natural language more efficiently and effectively.

A national file of subject headings could give all Russian libraries the possibility:

• To distribute their bibliographic records and re-use information accumulated by other libraries

• To keep union catalogues on different levels

• To publish joint bibliographies together with other libraries

• To provide their users with unified access to the information resources both in union catalogues and in other libraries.

An analysis of the practices of leading libraries in the world shows that there are two main approaches to development of national subject headings vocabularies and rules for subject indexing:

• Development of original national subject authority files and rules for subject indexing (Germany, Finland, Poland);

• Creation of national subject authority files on the basis of LCSH and adaptation of the Subject cataloging manual: subject headings of the Library of Congress.

Russian libraries have accumulated a rich experience in the field of theory and practice of subject indexing and subject catalogues. The first subject catalogues appeared in Russia at the end of the XVII century.

This experience could serve as the basis for the development of an original national subject authority file; such an approach would allow:

• The maintenance of national traditions

• More efficient and easy development as well as being less time- and effort-consuming than the adaptation of foreign lists of subject headings.

The problems of the development of a national subject authority file (NSAF) have been discussed at numerous conferences and seminars; in particular, the National Library of Russia (NLR) holds annual seminars on the problems of development of NSAF. These conferences have already become traditional. The resolution of the last seminar (Saint Petersburg, 1998) emphasizes the role and significance of NLR as a centre in the field of subject indexing problems. The decision was made to use NLR SAF as the basis for creating an NSAF.

The first subject catalogue in NLR appeared in 1894 in the philosophical division of the library. The history of the existing subject catalogue goes back to the end of the 1920s. At the same time the list of subject headings began to be created (in card form), and this became the ancestor of the contemporary NLR SAF. At the end of 1990s this list has evolved into a unique universal list of subject headings established and used at National Library of Russia (former - State Public Library), covering the main collections of NLR. It is significant to mention that the list of subject headings was created for use in current cataloguing, and it is the library collection which served as literary warrant for the headings.

In 1998 the list of subject headings was converted into machine-readable form. The current SAF includes over a million headings covering the following categories: personal and family names, corporate body names, geographical names, uniform titles, topical headings. In addition to subject headings, the authority file includes "see" and "see also" references (the latter reflecting both hierarchical and associative relationships between headings), as well as notes, application instructions, associated LBC class numbers (first division) and other appropriate information.

SAF is currently used in NLR for subject indexing various types of material: monographs, collections, reference and bibliographic publications, periodicals and serials.

NLR is universal, its scope is the whole world of knowledge, various disciplines and fields of learning: natural sciences (biology, medicine, geology, geography, etc.); social sciences, economics, mathematics, physics, chemistry, astronomy, technology, culture and arts.

SAF is dynamic, it is updated regularly. About 12,000 new records are created annually. While processing new acquisitions cataloguers submit proposals for creating new headings or changing the existing ones. The Authority Control team on a regular basis considers these proposals. It is planned that later, while creating National Subject Authority File proposals for new subject headings and references, the information will be contributed not only by NLR cataloguers but also by other bibliographic institutions which carry out subject indexing as well (similar to the practice of Library of Congress).

The National Subject Authority File is conducted in the local format for authority data and distributed in the Russian communication format for authorities. Both the local and the Russian communication formats are based on UNIMARC/Authorities and are fully compatible with it.

The project for the development of NSAF is connected with a number of other significant projects currently underway. First of all, there is the project for creating a machine-readable current national bibliography. The project will be realized under the agreement between three leading Russian universal libraries - National Library of Russia, Russian State Library and Russian Book Chamber.

Another project worth mentioning here is the project aimed at establishing correspondence between two systems of subject headings, namely NLR subject headings and LCSH. Successful realization of this project will permit the creation of some kind of tables of correspondence or matching headings between the two systems. The advantages of such tables are as follows:

1. In conditions of shared cataloguing, libraries could abandon their own indexing, which would allow them to decrease labour-intensiveness and to increase efficiency of their work

2. Cataloguers while processing acquisitions could use various databases to obtain missing data

3. It would be possible to provide subject access to multilingual databases.

The report deals with the necessity for developing a National Subject Authority File in Russia. Basic approaches to such a file creation are introduced. The history and main features of the National Library of Russia Subject Authority File are described. The report includes information on a number of projects being undertaken both in NLR and in Russia.

THE PREPARATION OF AN INDEX FOR THE CHINESE DDC 21: ISSUES AND APPROACHES (HOU Hanqing, WANG Dongbo, Editorial Board, Chinese Library Classification)

Dewey Decimal Classification (DDC) is an influential classification in the world. The editorial board of the Chinese Library Classification (CLC) is in charge of the translation of the DDC 21st edition Chinese version (Chinese DDC). Considering that the Relative Index is an important part of the DDC, the CLC editorial board decided to compile an index for the classification scheme of the Chinese DDC while translating the whole schedule. This paper will discuss the issues we have encountered and our approaches in the preparation of an index for the Chinese DDC.

SUBJECT GATEWAYS IN SOUTHEAST ASIA: AN EXAMINATION OF CLASSIFICATION PRACTICES Edward LIM HUCK TEE (Professor and Librarian, Monash University, Victoria, Australia)

Traditional libraries have always provided access to information resources designed to meet the needs of their clientèle; the Internet contrasts starkly with this focused approach and poses serious navigational problems for researchers. The majority of search engines are not effective for information control, due to a range of reasons, which will be outlined. The problems relate to locating resources and to the framing of requests by searchers. Solutions to the situation fall into three broad categories:

• Assisting users to make more effective use of search engines

• Improvement of the effectiveness of the search engines themselves

• The use of metadata and the creation of subject or information gateways

An examination of practices of a number of Web sites shows that the majority arrange subjects in alphabetical order, using subject descriptors, while some employ a classification scheme.

Use of the Internet is quite widespread in Southeast Asia, but it in no way reaches the proportions of usage in the developed world. Basically, Southeast Asian subject gateways fall into three categories:

• Those created outside the region

• Those created by commercial organizations

• Those created by Southeast Asian libraries

The largest such gateway is situated at the University of Leiden - Southeast Asian Studies WWW Virtual Library. It uses a home-grown classification. Others exist in Malaysia and Singapore and will be compared.

Commercial organizations maintain most of the major sites and these will be surveyed. Libraries, principally in universities or colleges, or national libraries also provide subject gateways and three, those at Universiti Sains Malaysia, The National University of Singapore and the Asian Institute of Technology, Thailand, are examined.

Although there are many techniques and methodologies which can be used to improve access to information resources on the Internet, many Southeast Asian libraries and organisations, unfortunately, still do not have the expertise, knowledge or resources to make use of these. This situation is not unexpected, as many Southeast Asian libraries continue to work within very tight financial constraints and suffer from a lack of trained staff; and so their ability to make use of the more sophisticated techniques and mechanisms available is limited.

If there is an area that IFLA wishes to make a contribution in, perhaps this is the one. There is an obvious need to conduct workshops to train Southeast Asian librarians to use many of the classification techniques to improve and enhance access to their Web subject gateways.

Developments since the Amsterdam Conference

Working Group on Principles Underlying Subject Heading Language (SHLs)
The Working Group on Principles Underlying Subject Heading Languages (SHLs) chaired by Maria Inês Lopes (Fundação Calouste Gulbenkian, Lisbon, Portugal) has completed its work, the report has been reviewed and it should be with the publisher by mid 1999.

Requirements for a Format for Classification Data
The Section has monitored and supported implementation of the recommendations of the Joint Working Group on a Classification Format of the IFLA Sections on Classification and Indexing and Information Technology. The Chair is a member of the Working Group and has also attended meetings of the Permanent UNIMARC Committee by invitation. Implementation involves modification of the USMARC Format for Classification Data and development of a UNIMARC format for classification data. It is hoped that work will be completed by August 1999.

The State of the Art Survey of Subject Heading Systems
A working group has been established to follow up on the State of the Art Survey of Subject Heading Systems. The survey sought to learn what countries provide subject access, how long they have done so, what systems they use, and whether they have manuals to provide guidance in application; 45 out of 123 national libraries responded. A follow-up survey brought results from an additional 38 countries. The Working Group, under the chairmanship of Magda Heiner-Freiling met twice during the Amsterdam conference to discuss the next steps and resolved that when the findings were complete, analysed and written up, they should be published as a journal article. The intention is to offer the article to International Cataloguing and Bibliographic Control.

International Conference on National Bibliographic Services
The Committee was invited to prepare a 3-4 page position paper from the Section to be given to attendees before the International Conference on National Bibliographic Services in 1998. This was prepared by Ia McIlwaine and Lois Mai Chan, in consultation with committee members and forwarded in December 1997. The Chair and several members of the section attended the conference and a report appeared in the last issue of this Newsletter.

Division project on OPAC displays
Representatives of the Committee attended meetings of the Working Group at the Amsterdam IFLA and the Section intends to maintain this involvement.

Dewey Decimal Classification news

Dewey for Windows
Dewey for Windows Version 1.20 has just been released. The database contains new terminology and changes previously announced on the Dewey Web site (http://www.oclc.org/fp/) plus several major additions. The additions include revised area tables for Great Britain and Republic of South Africa; comprehensive numbers for native American peoples; and built numbers and new index terms for popular topics in literature, computer science, and the life sciences. The database also incorporates the changes found in Decimal Classification Additions, Notes and Decisions (DC&), vol. 6, nos. 1-2.

Revised Area Tables for Great Britain and Republic of South Africa
Revisions of Table 2 Great Britain (-41-41495; -42-42998) and Republic of South Africa (-682-68792) are now available in a new Forest Press publication, Table 2. Geographic Areas: Great Britain and Republic of South Africa. The revised tables are also included in the new release of Dewey for Windows. The revision of Great Britain reflects the changes in the administrative units of Great Britain. The updated Table 2 notation for Republic of South Africa reflects the creation of new provinces and the elimination of homelands in South Africa. The revisions include updated Relative Index entries, as well as a new Manual note for Great Britain.

Decimal Classification Additions, Notes and Decisions (DC&)
Decimal Classification Additions, Notes and Decisions (DC&), vol. 6, no. 2, was published on the Dewey Web site in early 1999. Beginning with this issue, DC& has changed to electronic publication only. Dewey users are reminded that new and changed entries appear monthly on the Dewey home page, and are implemented by the Library of Congress upon posting.

DDC: Francophone Perspectives/CDD: Perspectives Francophones
Forest Press has just published Dewey Decimal Classification: Francophone Perspectives/Classification décimale Dewey: Perspectives francophones, the proceedings of the workshop on the French translation of the DDC held at IFLA 1998 under sponsorship of the IFLA Section on Classification and Indexing. The book contains five papers, in English and French, that discuss the experiences of the editors and translators responsible for the French translation of DDC 21. Three of the five papers focus on the translation process; the remaining papers describe how translations affect the development of the DDC, and the experience of teaching Dewey in French. Contributors to the publication include: Louis Cabral, executive director, ASTED; Raymonde Couture-Lafleur, translation editor; Bruno Béguet, Suzanne Jouguelet, and Max Naudi, of the Bibliothèque nationale de France; Julianne Beall, assistant editor, Dewey Decimal Classification; and Annie Béthery, former professor of library science, Université de Paris X-Nanterre.

French Area Table
The Dewey editorial staff is working with the Bibliothèque nationale de France on a revised area table for France. Review copies will be available in August on the Dewey Web site, and by request from the Dewey editorial office: Dewey Editorial Office, Library of Congress, Decimal Classification Division, 101 Independence Ave., S.E. Washington, DC 20540-4330, USA; phone: +1-202-707-5265; fax: +1-202-707-0279; e-mail: dewey@loc.gov.

540 Chemistry
A revision of 540 Chemistry is under way. Review copies are available by request from the Dewey editorial office (see address above). In February 1999, DDC and UDC principals met at University College London to discuss a proposed research project on a model chemistry development.

Translations of the DDC
ASTED (Association pour l'avancement des sciences et des techniques de la documentation, Canada) published the French translation of Edition 21 in 1998. Work is underway on translations in Arabic, Chinese, Greek, Hebrew, Icelandic, Italian, Korean, Norwegian, Russian, and Spanish. The Icelandic, Italian, Russian, and Spanish translations are expected to be published this year.

OCLC Forest Press
Peter J. Paulson, executive director of Forest Press since 1985, retired on 31 December 1998. Joan S. Mitchell, editor of the Dewey Decimal Classification (DDC) since 1993, has been appointed executive director of OCLC Forest Press, and editor in chief of the DDC. The OCLC Forest Press office has moved from Albany, New York, to OCLC's headquarters in Dublin, Ohio. The Dewey editorial office remains at the Library of Congress.

Universal Decimal Classification news

Staff changes
A new Director has now been appointed to take care of the UDC operation in The Hague. She is Ms Caren Apers, the former Office Manager, who attended the Amsterdam Conference and met a number of those who are interested in UDC there. There are currently two part-time Research Assistants appointed to University College London, funded by the UDCC, to assist the editor in chief. This has greatly improved the co-ordination of revision work and the regular monthly meetings between the editor and Research Assistants and the staff of the British Standards Institution have continued. This team will do the preparation of Extensions and corrections for 1999. There will be an opportunity during the IFLA Conference for interested people to meet the Editor to discuss the scheme and hear up to date news. The time and place will be announced in IFLA Express. Printed updates will also be available at the Division of Bibliographic Control meeting on Monday, August 23rd, at 9.00.

Revisions in progress
Work continues on revising Physics and parts of Chemistry, and it is hoped that this revision will be completed by 2000. Applications have been made to several sources to gain funding to support the project to develop a common switching language for Chemistry that will enable easy access via DDC or UDC, as mentioned in the report on DDC. The new class for Medicine should be completed within the next year, but it is not anticipated that it will be part of this year's amendments, as it is unlikely to meet the deadline of August 31st since it has yet to be approved by subject specialists. The draft schedule will be published this year. A revision for Cinema, for Tourism and for the Environment will be issued, and a revised and expanded Area Table for the USA and Canada, together with the usual number of corrections to update concepts or terminology throughout all sections of the scheme.

Extensions and corrections 1999
In addition to the revisions described above, this year's issue of Extensions and corrections will contain reports from the Editor in chief, the Chairman of the Consortium and users of the classification.

Guide to the UDC
Translations of the Guide to the UDC are presently being prepared in Czech and Portuguese (by Brazil). It has already been translated into French, Japanese and Spanish. A new edition is shortly to be commissioned by FID for publication within the next twelve months.

News from members

Participation in a European project on multilingual access to subject authority files and data (MACS - Multilingual Access to Subject headings) led to a close cooperation between Die Deutsche Bibliothek and the Schweizerische Landesbibliothek together with the Bibliothèque nationale de France and the British Library. The three indexing systems used by the partners: LCSH, RAMEAU and SWD/RSWK were compared, links between the subject entries in two fields, theatre and sports, were established and a practical study tested the possibilities of a linking system for the retrieval of indexed titles. The results were promising enough to plan the development of a prototype for the linking of descriptors from LCSH, RAMEAU and SWD later in 1999.

Classification data as a service in the national bibliography and for the needs of electronic publications and retrieval via internet have played an important role in the discussions of German specialists of indexing and classification in the past few years. An expert Klassifikationen für wissenschaftliche Bibliotheken published by the Deutsche Bibliotheksinstitut in 1998 evaluated a number of widely used international and German classification systems, a nation-wide working group is investigating the pros and cons of the DDC for the German-speaking countries, inspired by a workshop on DDC held at Die Deutsche Bibliothek by Joan Mitchell. There is a strong interest in a German translation of DDC and its application in the German national bibliography, and eventually in Die Deutsche Bibliothek as well as in the regional library networks and in Austria and Switzerland, too. But a lot of problems have to be solved first, beginning with the adaptation of DDC to the needs of German culture, administration and scholarly tradition, but also as far as finance, staff, professional training in the new system and effective organisation and information technology are concerned.

Magda Heiner-Freiling
Die Deutsche Bibliothek Frankfurt

Miscellaneous

The present officers of the Section are:

Secretary/Treasurer: Edward Swanson 1065 Portland Ave., Saint Paul, MN 55104-7011 USA (Tel + Fax: +1 651 291 1261; e-mail: eswanson@uswest.net)

Both are eligible for re-election in Bangkok for a further two-year period.

At the Copenhagen meeting Edward Swanson was also elected as Secretary to the Co-ordinating Board of the Division of Bibliographic Control.

Nominations for office holders should be sent to either the Chair or the Secretary not later than August 1st 1999, so as to ensure that those not attending the Bangkok Conference have the opportunity to nominate candidates.