The next version of the TEI Guidelines, P5, will contain a
major new chapter on manuscript description. Although
developed with the needs of manuscript scholars working in
the European tradition, it is hoped that the scheme
presented there is general enough so that it can also be
extended to other kinds of materials and other traditions.
The tagset documented in the chapter is based chiefly on
that developed by MASTER (1999-2001), an EU-funded project
headed by Peter Robinson, and the work of the TEI Medieval
Manuscripts Description Work Group (1998-2000), headed by
Consuelo Dutschke and Ambrogio Piazzoni. Although the work
of these two groups proceeded in tandem (members of each
attending the other's meetings and so on), and despite an
avowed intention that a single set of recommendations
should emerge from them, there were, in the end,
substantial discrepancies between the two. In 2002 the TEI
Council appointed a special task-force whose job was to
review both sets of proposals and identify and document a
common subset of those recommendations adequate to the
needs of the TEI community, taking into account the actual
experience of the many projects using MASTER as well as
complementary work done in this area by other agencies,
notably the Repertorium of Old Bulgarian Literature and
Letters. The result, we believe, is not simply a "common
subset" of the two schemes, but rather represents a
significant improvement on both, which at the same time
lays the foundation for future work.
A draft of the chapter is now available for comment on the
Specific suggestions, corrections, enhancements (i.e.
"feature requests") should be sent to
http://tei.sourceforge.net/, while general comments can be
posted to the TEI-L list or to the undersigned.
We ask for volunteers interested in working on conversion
of legacy data to the new format, and would be keen to
receive examples which could be used in the chapter.
Finally, although the major strength of the scheme we
propose is that it offers the choice between loosely
structured and highly structured data at every level, it is
clear that there are still areas (chiefly specific aspects
of physical bibliography) for which more richly structured
content models need to be developed and we would therefore
be keen to hear from people who might be interested in
proposing such models.
M. J. Driscoll