7 DECEMBER 1999
OXFORD, UNITED KINGDOM
National Council on Orientalist Library Resources
The 1999 Conference
The 1999 Conference of the National Council on Orientalist Library Resources was held at Rhodes House, South Parks Road, Oxford, on 7th December 1999. The title of the Conference was 'Old wine in new bottles; electronic access to Asian language resources' reflecting the major theme of the Conference which was a review of on-line resources and an investigation of software management systems and their functionality in relation to Oriental collections. The Conference was attended by 34 members of the NCOLR representing a wide range of academic and professional institutions throughout the UK.
By C.A. ANSORGE
The first of the presentations was given by Eleanor Robson, of the Oxford Oriental Institute, and described the basic structure of Unicode and its implications to cataloguers and librarians working in the Oriental field. She outlined the difficulties met with in attempts to display non-Roman script in a consistent way in automated catalogues and databases and the problems this had raised in relation to searching and retrieval. This has led to a need for a universal standard coding to include all scripts which could then be used in all software systems. The Unicode system was first developed in 1991 and is still under development. It was originally devised to look at scripts used in modern languages using non-Roman scripts, but has more recently developed into processing historical scripts used in classical languages. A Unicode website with full details of the system is available at www.unicode.org which gives details of the set up for all the character sets so far developed. Font design companies do not usually have off-the-shelf packages for non-Roman scripts or for any scripts which are of interest only to a very small group of specialists. Unicode has been accepted, in principle, by Microsoft for inclusion in further developments and programmes. The complexities posed by sorting in non-Roman character sets are still to be investigated.The second presentation was given by Michael Popham who is Head of the Oxford Text Archive in Oxford. He first described the establishment of the OTA within the Oxford University Computing Service under the auspices of the National Arts and Humanities Data Service in 1976. This was designed as a repository for electronic texts and to provide a free distribution to the scholarly community. This was a JISC-funded establishment of NAHDS which in January 1997 launched the web-site http://ahds:ac.uk with a web-site for the OTA at http://ota.ahds.ac.uk The OTA contains more than two and a half thousand texts in around twenty-six languages and includes electronic editions of individual authors, a variety of language corpora, and standard reference works such as the Bible. It also contains single language dictionaries, bilingual dictionaries, and newspapers.
Also outlined in some detail were the problems posed by issues of preservation of electronic resources and the problems of the migration of data to updated technology. Methods of preserving the readability of texts in an ever-changing world of new electronic equipment poses many problems yet to be solved. Some of these issues are addressed at www.bodley.ox.ac.uk/scoping/report.html. So far there are rather few examples of databases in Asian languages and the ones present at the moment are mainly in transliteration. The number of texts are growing and there is a Japanese and Chinese text initiative specifically to deal with texts in these areas.
Diacritics
Following the break for lunch there was a panel discussion focusing on three commercial library management systems and their capabilities in relation to Oriental language material. The complexity of the problems facing Oriental language cataloguers concerning the display of diacritics in transliterated records, and problems to do with the display of non-Roman scripts, were to be described by people with expertise in various commercial software packages. The first presentation given by Margaret Savage-Jones from the Wellcome Institute was of the Innopac system. This system has been chosen by the Wellcome for its multilingual support capability and additional customization had also been carried out since its installation. To assist the cataloguing process keyboards had been customized to input special characters. Catalogue records in transliterated form were displayed but the display of Oriental scripts was not yet possible.
Shirley Perry from Manchester University then described the Talis system which is in use in their library OPAC system. The Talis system can produce catalogue records for Oriental material in transliteration only, and has, at present, no plans to develop original script cataloguing. Special script facilities existed only as add-ons from the Web. Diacritics could occur only in ASCII character sets where the special characters can be combined with the letters. Diacritics displayed in catalogue records appeared as splodges which were of a very unsatisfactory for the catalogue-user.
The third system to be considered was GEAC, described by Peter Burnett of the Bodlean Library, Oxford. He outlined the history of the Oxford on-line catalogue including the migration of the catalogue data from a previous system to GEAC in 1997. Cataloguers in Oriental language are able to use the cataloguing module, Geocat, which includes input and display facilities for diacritics. This facility was a later development and the inputting of diacritics is by keymap, which is composed on the screen. The character string for each diacritic can be stored and used when needed. Catalogue display of records with diacritics was much more successful than in the other two systems. At the moment the system stores but does not display the 880 fields in RLIN records. GEAC cannot, at the moment handle CJK script.
The system demonstrations were followed by a general discussion among Conference participants. The Aleph system, not demonstrated, was also mentioned as a system for possible consideration. This system, developed in Israel, has a UK version, but the British examples of Aleph are unable to display non-Roman scripts. The Hebrew Aleph system with Hebrew script display could not be integrated with the Roman script version. All panel members emphasized the need to be very clear about the precise specification needed when purchasing a system and also to be clear exactly what was on offer from a supplier. It was also very important to have technical assistance from the supplier until all requirements have been met and are successfully working. It was important to know if special fonts came as add-ons or came automatically as part of the overall package. The need for searching and display in original script was considered an important feature for future development but also the need for transliterated records would be likely to continue. The incorporation of Cyrillic or Oriental scripts into the same databases as Roman scripts raised problems of indexing which had not yet been addressed. *
C.A. Ansorge, National Council on Orientalist Library Resources.
E-mail: caa1@cus.cam.ac.uk