TEI Work Group for Medieval Manuscript Description:

Minutes, Union Theological Seminary, NY NY

3-4 July 1998

Table of Contents

TEI Present:

1 A. Administrative Session (morning, 3 July):

1. Introductions: --each person presented himself, explaining interest and past work in similar fields of endeavor

2. Tax forms: --agreed to put off discussion until following day, since Michael Sperberg-McQueen would not be present then: no reason to waste time while he was there

3. Receipts/reimbursements: --idem; mention of extreme difficulties by Eva Nylander and Ambrogio Piazzoni to get the putatively necessary visas from American embassies in their respective countries and then to obtain the supposedly correct stamps at Passport Control at JFK; Piazzoni's ITIN can be obtained Monday morning from appropriate Columbia office but Eva leaves over the weekend

4. Communications: --TEI-L (to which we should all subscribe): received by all TEI-interested parties --EAMMS (to which we should all subscribe): some 45 receipients, interested in standards for the cataloguing of medieval manuscripts --TEI-MMSS (to which we are all subscribed automatically): 8 receipients, to serve as working, note-form communications to advance work more speedily --agreed for all listservs, as per P. Kidd's suggestion, to limit each message to one item of concern

5. Handouts at meeting: --Rules and Recommendations for TEI Work Group Procedures (http://www-tei.uic.edu/orgs/tei/ed/edw54.html) --TEI Reimbursements Rules for TEI Work Groups, 1998-1999 (http://www-tei.uic.edu/orgs/tei/wg/reimburse.html) --"Drama" (from P3, 1:269-295) as sample deliverable --L. Burnard, R. Gartner, P. Kidd, "The Cataloguing of Western Medieval Manuscripts in the Bodleian Library: a TEI approach with an appendix describing a TEI-conformant manuscript description," in the version dated August 1997 (http://users.ox.ac.uk/~lou/wip/MS/)

6. Handouts sent as study materials prior to meeting:

7. Outlines of other known groups working in related areas:

8. Outline of TEI:

2 B. Philosophy of approach (afternoon, 3 July):

1. Accusations of disadvantage to SGML/TEI (to Know The Enemy) --too regimented for scholarly use --insufficient number of tools --too difficult to learn

2. For our purposes --"punt" on problems that are of general import (i.e. bring to TRC)

3. Triple use for TEI-Med MS Descript DTD envisioned: A. as description (secondary goal) to accompany edition of text (primary goal) B. as unit internal to larger grouping of the same units (e.g. catalogue) C. as independent unit, free-floating, to plug into any other function (e.g., edition of text; catalogue of medieval manuscripts of a given institution; study of one scribe's work surviving in three extant mss all in different institutions, and in one ms destroyed in WWII, and in one ms of unknown present location); database record" N.B. This discussion was crucial, spirited and lengthy with segments of confusion and of clarification, since acceptance, especially implicit, of one use as paramount over the others affects the relationship of header to body, and thus all work that lies ahead. In outline form, some of the turning points of the conversation:

--Generally: The work already accomplished at Oxford (L. Burnard; R. Gartner; P. Kidd) follows situation A; how to best incorporate its achievements --PK: Feasible to consider placing descriptions under umbrellas? e.g. internal to situation B, group the mss of a given collection? e.g. internal to catalogue of the Bodleian Library, provide the umbrella of the history of the Douce Collection, with its own mss? --AP: Each description needs its own header with its own revision history and specific sources of information, bibliography; if a library wants to maintain the best available information on a ms, it would need to revise individual descriptions accordingly; in fact, responsibility for a given piece of information often needs to be assigned at the field level. --MP: Are we thus disallowing situation B? --CF and CD in unison, and respectively: "no"; "yes" (indicative of how much we all care about this stuff!): CF because we need to provide something that will work in all contexts; CD because we need to stay as simple as possible --CD: we are one arm of the two-part effort of EAMMS: keep in mind work currently under way by Gregory Pass and Larry Creider --S-McQ: remember information factoring: if repository information is present in the header, you can assume that it is true for the body of the work --S-McQ: current header should not be a straight-jacket but modify it with caution (since it is clear that our work will require us to add to the header)

3 C. Administrative session (morning, 4 July):

1. Tax and receipt business that had been put off from preceding day

4 D. Philosophy of approach (morning, 4 July):

1. Agreement that we seem to be heading towards situation C, but that all situations need to be addressed; agreement that in our limited time left we should move on to the issues of the tags themselves.

2. PK on EAD --EAD already exists --allows for hierarchical movement into a library's entire holdings --work to date with TEI for medieval ms description has produced a DTD daunting in its complexity

3. Templates --as a means for simplifying use of TEI in actual practice

5 E. Tags vs. Attributes (morning, 4 July):

[The following is somewhat expanded, with help from P. Kidd and M. Proffitt, from notes taken at the session, to be sure of representing in these minutes the nature of the meeting's discussion.]

1. Examples follow, in which the first -extreme-- markup treats everything as metadata (verbose), while the second (more concise) treats "50 mm" as the data. The second (and the fourth) markup also serve to pose the question of whether, for example, a Dimension should be an attribute (or a subset) of a Leaf or vice versa (P. Kidd favors the former situation). The third and four markup samples demonstrate a minimized reliance on attributes.

<leaf><dimension TYPE=Height EXTENT=50
<dimension TYPE=LeafHeight>50 mm</dimension>

2. The issue of "tag vs. attribute" pulled in questions of controlled vocabulary: do we provide a closed list of tags, a fixed list of attributes? (problematic due to differing vocabularies, and even due to differing perceptions on the cataloguer's part); all this is related as well to issues of encoding new material directly or dealing with legacy data. P. Kidd brought up examples in the area of description of decorated initials:

--simplest way, especially for legacy data:

<initial>With lots of fancy initials</initial>

--a two-pronged, but still fairly simple method:

<initial><decorative>Plain red
<initial><figurative>Initials with animals and

--a more complex system could continue nesting elements (as in the first below) or could use attributes (as in the second where a pre-defined list of legal values would be called for):

<initial><figurative<inhabited>Initials with animals
amid foliage</inhabited></figurative></initial>
<initial><figurative TYPE=inhabited>Initials with animals
amid foliage</figurative></initial>

3. CF: issue of repeated data elements vs. repeated and associated data elements

4. MP: warnings on structuring our thinking in terms of divs.

6 F. Tags (afternoon, 4 July):

1. Agreement to stay with work division as outlined before meeting, i.e.:

2. Discussion conducted using EAMMS first-level list (determined at Nov. 1997 meeting in NY) as thread; for results, see appendix

7 G. Date Targets (afternoon, 4 July):

1. Suggested target of 1 September 1998 to coincide with MASTER's planned first meeting [N.B. now revised by MASTER to 1 November] in order to follow this sequence:

2. 1 October 1998: each group member to have his assigned segment of elements to Merrilee Proffitt and Michael Sperberg-McQueen (with recognition that Eva Nylander's section on codicology is the most work, inspite of the excellent documentation in this area provided by Belinda Egan; therefore when a member completes a shorter section, such as Consuelo Dutschke on paleography and provenance, that member should go pick up a subsection from Eva and keep working)

3. 1 November 1998: draft DTD prepared by Proffitt and Sperberg-McQueen

4. March 1999 in Rome for second work group meeting (Don Raffaele Farina, Prefect of the Vatican Library, has already extended an invitation) --necessarily at least 15 days before Easter in order to secure accomodations

5. June 1999, next meeting of TRC, ergo not possible for us to supply completed work that has had at least three months' prior public airing for commentary/critique (i.e. by March 1999); will have to submit work to following TRC meeting

Respectfully submitted, on this 28th day of August, 1998,

Consuelo Dutschke, co-chair

Ambrogio Piazzoni, co-chair

A H. Appendix: tags

....... to be continued