Bibliographic Research Bibliographic ResearchThere has been a long history of bibliograhic research at Bradford undertaken in the Department of Computing with collaboration from the University Library, much of the work has revolved around the USBC (Universal Standard Bibliographic Code). This work is ongoing and we welcome proposals of collaborative work.
Recent Projects include:
We have been investigating the development of some of the ideas raised in the Bradford OPAC 1 project in conjunction with Z39.50 access to multiple databases. This is an attempt to tackle the problems of dealing with large and complex retrieval sets possibly from multiple databases. This work is now continuing as BOPAC2 (Bradford OPAC 2).
Also known as the Manifestation Project, this was a one year feasibilty study into developing an OPAC based on the concept of grouping together manifestations of the same work. Our intention was to create a catalogue that was not restricted to automating old concepts that originated with the card catalogue. An aim here was also to eliminate features like the short title listings that appear on many OPACs which are often too brief to contain useful information that might distinguish between different versions of the same work.
The prototype OPAC was developed on a PC with a Windows interface and used a small database of computing books at Bradford University and some known large and complex manifestation sets such as Handbuch der Physik and Smollett's Humphrey Clinker.
More information on this project and the manifestation concept can be found in FH Ayres, 'Bibliograhic Control at the Crossroads' and FH Ayres, LPS Nielsen, MJ Ridley, 'Bibliographic Management: A new approach using the Manifestation concept and the Bradford OPAC.'
This work was funded by the British Library R&D Dept (later known as the Research and Innovation Centre) and the full report is available electronically with their permission. ( Note this is a 50 page Postscript document that contains a large number of screen dumps and may be slow to down load or print)
This was a project investigating Greek language transliteration problems under the EU Telematics for Libraries programme. There are many problems with transliterated material in library catalogues, different forms of transliteration are often used making it hard or impossible to locate Greek language material in library catalogues and other databases. Many of the problems tackled are also applicable to other languages that dont use the Roman script. Developments in computer character sets such as Unicode are making the storage of non Roman script material unnecessary. It is now possible to store this material in its correct form and transliterate in software when this is required. A range of documents and software, including programs to convert MARC records to and from different Greek transliteration schemes, is available by ftp at Bradford or Crete which also houses the Project Helen home page.
We have recently been working with the British Library Document Supply Centre on systems for automating the processing of Inter Library Loan requests. This work has used USBCs to match ILL requests to DSC holdings. Problem areas here are the varied and sometimes incorrect ways in which citations are quoted.
This project used the USBC to cluster together duplicate and possible duplicate records in a number of bibliographic databases, looking for duplication both within a database and across databases. An expert system was then used to provide a detailed examination of the records to determine if they were really duplicates and if so which record was best. There are a number of problems in this area that the project tackled. Some records are clearly duplicates since they are identical, in many other cases small differences may or may not be significant. Small differences between records can be due to spelling errors or different editions. Although cataloging rules may specify locations for information in the bibliographic record, in practice these rules may be interpreted in different ways or ignored.
In addition to a full BL R&D report information can be found in MJ Ridley, 'An expert system for quality control and duplicate detection in bibliographic databases'
The Universal Standard Bibliographic Code is a code used to identify bibliographic items, monographs, serials, articles in serials etc. that is derived from parts of the bibliographic record. In contrast to other identifiers such as the ISBN or ISSN no part of it is allocated. Therefore a USBC can always be created from a reference, and can also be used for material that hasnt had an ISBN allocated. Various forms of the code have been developed for different uses and it has been used to match between bibliographic databases to find duplication and to match citations to full text databases. A history of the development of the USBC can be found in Ayres, Nielsen, Ridley and Torsun, 'USBC (Universal Standard Bibliograhic Code): its origin and evolution.
Z39.50 is a NISO standard for searching remote databases and retrieving information from them. There are comparable ISO standards for SR (Search and Retrieve)(10162 and 10163). Z39.50, which is still under development is more widely used. Using the Z39.50 protocol a remote library catalogue can be queried, possibly using the more familiar interface of your local catalogue.
Z39.50 resources can be found at a number of places on the Net, in the UK a good starting point is UKOLN's Z39.50 in Europe page.
Publications in this area are listed with the Database publications.