1 Introduction

There are three main series of published catalogues of the western manuscripts at the Bodleian Library: the so-called `Quarto' catalogues, published between 1845 and 1900, in quarto format, which cover the major collections acquired (for the most part) in the seventeenth and eighteenth centuries (Coxe 1845-1900); the Summary Catalogue , published between 1895 and 1953, which covers the manuscripts acquired from 1602 to 1915, except those already described in the Quarto catalogues (Madan 1895-1953); and the Summary Catalogue of Post-Medieval Western Manuscripts , published in 1991, which covers most post-medieval manuscripts acquired between 1916 and 1975 (Clapinson and Rogers 1991). In January 1996 the Library began a four-year project, funded by the Higher Education Funding Council for England (HEFCE), under the Non-Formula Funding Specialised Research Collections initiative, the purpose of which is to make available descriptions of the medieval western manuscripts acquired by the Library since 1916, for which no full published catalogue yet exists. For more detailed bibliographical information on the catalogues of western manuscripts at the Bodleian, see http://www.bodley.ox.ac.uk/guides/wmss/wmss02.htm.

Broadly speaking, this project focusses on two rather different tasks: firstly, the cataloguing of the manuscripts themselves, and secondly, the publishing of the resultant catalogue descriptions in both printed and electronic form. Each of these two main strands of work presents its own challenges, while the second, in particular, involves breaking new ground at the Library.

This paper reports on some of the problems addressed by the project, primarily from the bibliographic point of view, together with the technical approaches we have adopted for their resolution. Our basic approach has been to build on existing work as far as possible, while at the same time seeking to develop a system adequate to the Bodleian's arguably rather specialist needs. In that spirit, we have developed a set of extensions to the Text Encoding Initiative (TEI) proposals for general purpose text encoding (Sperberg-McQueen 1994), tailored to the needs of manuscript cataloguers. A detailed appendix to the paper documents this set of extensions, as they are currently formulated.

It should be emphasized that the proposals here presented are very much `work in progress' intended to promote discussion and review by other similar projects, and to clarify our own thinking on the issues. They should not be regarded in any sense as a final or fully articulated set of recommendations. The Bodleian authorities have not yet made a final decision about which SGML solution to its cataloguing needs will be adopted; and in addition to the exploration of the TEI, experiments are underway to explore the uses to which the Encoded Archival Description (EAD) DTD can be put.