Bodleian Library
Elizabeth Solopova presented an overview of the Bodleian’s own project to recatalogue its medieval manuscripts collections, which has been carried on since 1997 in several stages, beginning with basic records and adding detailed descriptions to classes of material, including, recently, the creation of separate records for illuminations in manuscripts in partnership with ARTstor.

Dr Solopova described the Bodleian’s current plans for continuing to enrich descriptions of medieval manuscripts; to add digital images, where possible, linked to the records; and to improve electronic resources by linking databases of citations to the records.

The format for descriptive data used by the Bodleian is currently XML in an EAD format; the Bodleian hopes to switch to using TEI P5. It is also intended that all digital objects, such as records and images, will be held within the METS format.

Bodleian Library electronic catalogue of medieval manuscripts, http://www.bodley.ox.ac.uk/dept/scwmss/wmss/online/medieval/

Parker Library, Corpus Christi College Cambridge
Prof. Nigel Morgan presented the Parker-on-the-web project, that will present digital facsimiles of 530 medieval and early modern manuscripts from the Parker Library, Corpus Christi College, Cambridge. The project is managed by Prof. Morgan and his colleague Rebecca Rushforth, and the photography is conducted by by a team of photographers who work in Corpus Christi College, funded by the Andrew Mellon Foundation. Stanford University, the other partner in the project, are responsible for the website and the presentation of metadata.

The entries are prepared by the Parker Library team taking an XML mark-up of the M.R. James catalogue of manuscripts in the library as the core of the record, with author and title forms updated. The mark-up requires expert supervision, because of the format of the James entries and the lack of consistency in their layout and in the transcription from original texts. Longer entries of several pages could take days to mark up in XML. To the entries, the team adds bibliographical references.

Answering a question from Richard Ovenden about the way this digital version of the library would be presented to users over the web, Prof. Morgan replied that users would be able to go directly to images of manuscripts, or choose to search through records and enter the image database via the catalogue records.

Parker-on-the-web http://www.corpus.cam.ac.uk/parker/parkerweb.php

Digital Scriptorium
Consuelo Dutschke presented the Digital Scriptorium database. This is compiled according to TEI-MS and descriptions are accepted from a number of institutions. To coordinate the descriptions, institutions are enabled to enter records in DS-Access, which is then exported to an XML encoding. The input schema has been simplified to allow libraries to contribute minimal records. Dr Dutschke described the apparently conflicting desires for DS: while consortial partners want ease of contribution, preferably with interoperability between their own systems and DS, researchers want more content, more images, and fuller descriptions of what they find, while also seeking simplified discovery and access.

Digital Scriptorium http://www.scriptorium.columbia.edu/

Houghton Library, Harvard University
Data formats: the Harvard experience: As Harvard was one of the libraries contributing to the DS, Bill Stoneman described the experience of delivering records suitable for the DS format, while retaining MARC records as required by the Harvard library (HOLLIS) system. Records sent to DS were converted from MARC to DS-Access. As this was a straightforward process, Harvard spent the bulk of the grant it received to join DS on digital imaging.

Subject and name headings in all the Harvard HOLLIS records are entered according to Library of Congress authorities. These headings enable users to search the collections for all “Manuscripts – France – 1350-1400”.

Bill Stoneman also confirmed that the Harvard policy on scanning of manuscripts was now to approve only cover-to-cover digitisation, creating a complete digital facsimile.

http://hcl.harvard.edu/libraries/houghton/collections/early_manuscripts/index.html

CERL
Ivan Boserup presented the CERL portal: Manuscripts and the European Printed Heritage, and described the Simple and Advanced search options, remarking especially on the CERL thesaurus that enables assisted searching for personal and place names. Dr Boserup commented on the need for a specialist review of the thesaurus, for instance by reviewing the names of medieval authors, and adding biographical data if possible to distinguish similar names.

CERL manuscripts and early printed books portal http://cerl.epc.ub.uu.se/sportal/

British Library
British Library Catalogue of Illuminated Manuscripts Kathleen Doyle presented the Catalogue of Illuminated Manuscripts from the British Library. This had been initially made on an Access database, but the library was considering migrating the data to XML. A potential further refinement would standardize the iconographical searching, and participants discussed the merits of ICONCLASS as a tool for this purpose. While limited to figurative decoration, and to description of the iconographic content, rather than the material elements of decoration, such as medium and size, the ICONCLASS system has the advantage of providing a browser available in several languages.

British Library Digital Catalogue of Illuminated Manuscripts, http://www.bl.uk/catalogues/illuminatedmanuscripts/welcome.htm

Also noted are the union catalogue of German medieval manuscripts, Manuscripta Mediaevalia : http://www.manuscripta-mediaevalia.de/

http://hcl.harvard.edu/libraries/houghton/collections/early_manuscripts/index.html

And the project for short-title descriptions of 7500 medieval manuscripts in Dutch collections http://www.nwo.nl/projecten.nsf/pages/1900120317 http://www.kb.nl/hrd/samenwerking/parchment/index.html

Others are described on the CERL webpages: http://www.cerl.org/Manuscripts/ms%20websites%201_Dec%2004%20update.htm

Discussion
The strongest theme in discussions during the rest of the day was standardization to enable cross-searching of records online. Richard Ovenden asked participants to consider:

1) the minimum set of fields that would be required to share or cross-index records, as is done through the DS and the CERL portal.

2) contributions to a thesaurus of names or terms.

The basic record: list of fields Some candidates for such a list were discussed: Author Title Identification of text through incipit Whether containing figurative decoration or not (Yes/No) Genre term (as a supplied title or as a genre-subject field) This list can be compared to the longer list of required fields as outlined in the Digital Scriptorium data dictionary: https://www1.columbia.edu/sec/cu/libraries/bts/img/assets/7895/DS-Access_Data_Dictionary_v8.0.0.pdf

Sources for thesauri of names and terms
Names
Most discussion centered on the idea of an authority file for names. Participants noted the lists already used in a number of online library catalogues:

See a list of source files used in the CERL thesaurus: http://cerl.sub.uni-goettingen.de/ct/

Personennamen des Mittelalters; available via PND: the German national authority file for personal names, http://www.ddb.de/standardisierung/normdateien/pnd.htm
The BSB uses PMA for author names but does not feed in new names from the manuscript catalogue.

Richard Sharpe, A Handlist of the Latin Writers of Great Britain and Ireland before 1540- (1997). There is a brief update (2001) on-line at http://www.history.ox.ac.uk/sharpe/lw.pdf
And the cumulative list of works identified in medieval catalogues from British libraries (AD 1123--1556), of which the most recent public version has been available since 2003 from http://www.history.ox.ac.uk/sharpe/index.htm (This link leads to the list itself, an explanatory introduction, and keys to the catalogues on which it is based). This shows around 7,500 works from antiquity to the renaissance; with multiple copies of many popular works, the total number of copies index and identified is now around 30,000.

For Anglo-Norman, Ruth Dean & Maureen Bolton’s guide to Anglo-Norman literature is used by the Parker project

For Middle English, no comprehensive list was named.

Titles
In Germany, Manuscripta Medievalia has begun an authority file for titles of works. In Principio (from Brepols) links incipits to identified texts

Other wordlists
Following from the discussion of ICONCLASS, some thought was given to the question of a list of terms for the type of decoration (along the lines of Getty’s Art and Architecture Thesaurus).

Models of authority file-building These could be found in LCNAF, CERL, PND (mentioned above) and André Bouwman mentioned the LEAF project to link authority files: http://www.crxnet.com/leaf/

Conclusion

Richard Ovenden proposed that the group might agree a set of standards for key elements of the medieval manuscript record: - names, entered into a thesaurus so that different forms would be cross-referenced - titles - terms, or the creation of a thesaurus of terms, for such elements of physical description as illumination and figurative illustration - expressions of dates; to reconcile different forms in which dates are entered and to reconcile the different calendars in use

The group endorsed the creation of an enhanced name authority file for medieval names (with biographical information), creating new online forms of currently used namelists and incorporating these into the CERL thesaurus + PMA. Ivan Boserup was asked whether a special CERL thesaurus of medieval names could be extracted, to be incorporated into current online cataloguing projects.

Participants were willing to share information about the namelists used in creating their own online catalogues and to seek institutional support or other funding for converting these into machine-readable lists as required.