Many scholars have made suggestions concerning the making of computer-readable manuscript descriptions over the last twenty years or more, and many different computer systems have been designed to support them. Notable pioneering initiatives include those of the IRHT in the seventies (resulting in the formation of the MEDIUM database) and of the Italian Censimento in the early nineties; a convenient summary of several such systems can be found in the volume Bibliographic Access to Medieval and Renaissance Manuscripts: A Survey of Computerized Data Bases and Information Services , ed. Wesley Stevens, Primary Sources and Original Works, Volume 1, Numbers 3/4, 1991. Such systems have tended to reflect a particular research context, and thus to be rather narrow in scope of application; they have also almost all chosen to use relational database technology of some kind. By contrast, the goal of MASTER is to create a generic system, sufficiently flexible and robust for application in many different domains of manuscript description, and the technology chosen to support this more ambitious goal is based on the international standards SGML and XML.
The advent of the world wide web and advances in digital imaging of manuscripts gave a new impetus to the work of computer based manuscript description during the mid-nineties. Almost at the same time that the libraries of Europe realised how digital imaging could offer an unprecedented expansion in the accessibility of manuscripts, so they also realised the pressing need for online cataloguing of such materials. Only with this in place, would the provision of unified access to the new digital repositories springing up across Europe make the difficult transition from the technically feasible to the practically realized. And such unified access could only be achieved by prior standardization work -- by agreement about how manuscripts should be described.
In North America, a key event was the Mellon foundation's decision in 1996 to fund two collaborative projects: the Electronic Access to Medieval Manuscripts project (EAMMS) and Digital Scriptorium. In Europe a seminal meeting at Studley Priory, near Oxford, brought together nearly thirty European and American experts on computer systems and manuscript cataloguing in November 1996. This meeting revealed both a widespread concern about the need for a standard, and perhaps more surprisingly, a broad consensus about the appropriate technical means to be adopted to reach one. After a long process of negotiation with the European Union, MASTER began (with funding from the Fourth Framework program) on 1st January 1999.
We noted above that most previous approaches to this problem had reflected the prevailing orthodoxy in attempting to use relational database technology as a means of creating, managing, and searching the highly complex textual objects which make up the traditional manuscript description. Inevitably these attempts had tended to diverge considerably, both in the scope of the information contained, and in the portability or generality of the software systems in which records were stored and managed. By 1996 however, an alternative technology had emerged in the shape of the Standard Generalized Markup Language (SGML). This International Standard had already proved its worth as a basic enabling technology for the World Wide Web; by this date its relevance to the basic encoding of the materials of textual scholarship had also become apparent, largely as a result of the work of the Text Encoding Initiative (TEI) whose Recommendations , first published in 1994, had gained very wide acceptance within the international scholarly community.
While the consensus amongst the various scholars, archivists and computer experts involved in the EAMMS, Digital Scriptorium and MASTER projects indicated that that SGML encoding offered at the least a very promising basis for the evolution of a general purpose computer-readable manuscript description system, there was also a realistic concern to maintain compatibility with more traditional database technology. Many records have already been created using such systems, and if the chief virtue claimed for the new approach was application independence, then there existed a large constituency of powerful and useful legacy application systems to retain compatibility with. Consequently, both Master and EAMMS have committed themselves to supporting both database (for example using MARC) and XML -based systems. It was hoped that an XML-based system might help us realise the dream of a universal interchange format into which many manuscript records from many databases might be usefully combined.
Several factors contributed to the choice of the TEI as the basis for the proposals here outlined, some of which are discussed further below. Firstly, there was the perceived success of the TEI guidelines in addressing many of the issues which would have to be resolved in any scheme of computer readable manuscript descriptions. Such issues as the encoding of transcription of primary textual material, of various kinds of textual division, of metadata in general, of hypertextual links, of multilingual materials, and many others had all been addressed by the TEI Recommendations, and moreover have for the most part already been widely tested. Secondly, the TEI workgroup system seemed offered a well-proven means of bringing together domain experts and encoding experts, and harnessing their various skills towards a consensus which could form the basis of a widely-accepted standard.
The Studley Priory meeting which effectively inaugurated MASTER was followed by meetings of the EAMMS group at Hill Monastic Manuscript Library in December, 1996, and a year later, in November 1997, by a meeting at Columbia University which brought together many of the participants in the EAMMS, Digital Scriptorium and MASTER projects with other manuscript experts. Following this meeting, the TEI workgroup was established, with Consuelo Dutschke and Ambrogio Piazzoni as co-chairs. This workgroup first met in July 1998. The guidelines here described are the result of this meeting and of a series of further meetings involving various members of EAMMS, MASTER and the TEI workgroups: held most recently in New York in January 1999; in Paris in February 1999; in Rome in March 1999; and in Oxford in June 1999. The first draft of the DTD described here was prepared following the Rome meeting, and refined further over the next two months by means of vigorous electronic mail discussion. A preliminary draft was thus available for consideration in May 1999 by the MASTER Experts Group.
This Group brings together a small group of distinguished manuscript scholars to provide an independent evaluation of the project's proposals. Its members are: Dr Ian Doyle, Durham University Library (chair); Professor Peter Gumbert, Leiden; and Dr Gilbert Ouy, Paris. Following a meeting of these Experts, a `green paper' describing the draft and summarizing the discussions concerning key points within it, was posted for further semi-public comment through the members of the EAMMS discussion group. In June 1999 a technical subcommittee of the MASTER group met in Oxford to discuss further modification of the draft. It is this modified draft which is described in this paper. We confidently expect that publication of this draft and its associated documentation will provoke further discussion and redrafting. It should be emphasized that what is here presented has not yet been formally adopted by any of the working groups concerned, and should be understood only as the current recommendations of the authors.