TCM05: TEI Council Conference Call 18 Mar 03

The following were present: Syd Bauman (SB), David Birnbaum (DB) in part, Lou Burnard (LB), Matthew Driscoll (MD), David Durand (DD), Laurent Romary (LR), Chris Ruotolo (CR) in part, Susan Schreibman (SS), John Unsworth (JU), Perry Willett (PW), Christian Wittern (CW).

Actions from previous meeting: all actions not deemed already complete are on today's agenda. It was agreed to defer to the next meeting the question of who should succeed JU as chair.

Council approved the procedure defined in EDW78, and actioned the editors to begin implementing them.

• CR reported progress from the Migration Work Group. The expert group has finished its draft and repository representatives had begun testing. There had been a useful joint meeting in College Park. The biggest problem so far identified was what should replace SDATA entities in XML DTDs. A structure has been defined for writing up the final reports, which included a template for individual case studies. Plans for surveying existing TEI projects had been scaled back somewhat but would continue. It was planned to hold a further meeting (of the whole group if funds permit) June 16/17th in Alicante, and the reports should be ready in time for TEI AGM. Council thanked CR for her report and for the Group's good progress.

• Reporting on the Character Encoding workgroup, CW said that

  • he had drafted a working paper on SDATA entities which had been discussed in some detail, but which the workgroup had not been able to endorse; work on this had now been suspended.
  • the group was instead making substantial progress towards recommendations for a markup scheme for character set extensions (see CE W06). Although only a preliminary draft, this was now undergoing test implementation
  • the group had requested the new ML workgroup to consider whether or not it was realistic to disallow any kind of CDATA content in attribute values. Without such a prohibition, representation of non-Unicode characters in TEI documents by elements would be at best problematic.SR reported that the ML Workgroup has briefly considered this question. It seemed advisable to get a sense of the scale of the problem, and he suggested the editors should review all existing TEI attributes, characterising each as tokenizable, numeric, or textual. In DTD world, the textual ones should be left unchanged; in schema world, however, equivalent child elements could be generated.

Council agreed that further work was needed to decide whether this proposal was necessarily restricted to schemaland, how the content models and attributes for generated elements should be determined, and whether such a scheme might generate name clashes. Proposals should be sent to the ML list for further discussion.

Editors1 May to produce categorised list of attributes and investigate possible name clashes CW1 May to propose content models for generated sub-elements

DD reported that the Standoff workgroup continued to make slow progress. Some papers were now available on the web site, (SO W01 discusses use of Xpointer and gives an overview of the groups major concerns; SO W02 is a discussion of the canonical reference scheme described in the Guidelines; SO W03 is a working paper on graph description); several others are in progress and should be available by end of April in time for d1scussion at the Council's May meeting. Council was requested to review SO W01 as soon as possible.

SB1 Aprilto update SO website with details of papers as they become available DD30 Aprilto make remaining work group drafts available to Council

• MP reported that Consuelo Dutschke had now delivered extensive documentation of the original TEI MS Description Workgroup's proposals for mss description, which were now available in MS Word format from MD's website. In line with the Manuscript Taskforce's charge, Council now requested DB, MD, and MP to review these drafts and identify an achievable programme of work for harmonizing them and the MASTER scheme. This should fall somewhere between simply producing a list of differences between the two sets of proposals and merging them to a single coherent scheme. LB agreed to set up a web-site area for this Task force (Now done: </activities/MS/>) MP, MD, DBasap to review available drafts and initiate task force activity

• SR reported that the Metalanguage Taskforce had been set up, as per its charge, with himself as chair, and membership of CW, DD, Norm Walsh and Michael Sperberg-McQueen and the editors. In discussion it has endorsed the existing literate programming model of the Guidelines, and has produced a proof-of-concept protype for P5, in which schema definitions using RelaxNG are generated from an extended version of the current ODD model. SR is currently working on a replacement for the pizzachef (called Roma) which will deliver TEI conformant schemas as DTDs or in RelaxNg. Council was suitably impressed. Taskforce working documents are maintained at </activities/META/>.

• Council endorsed the charge to set up a joint wg on Feature Structures as a new work item for ISO TC37-SC4. LR's proposal has now been progressed within ISO: an official letter requesting status of liaison was to be sent. ISO member bodies have three months to comment on the new work item proposal; there will be a short meeting on 6/7 July in Sapporo which should compile the comments received and instruct an editorial workgroup to work on the proposed document. JU will be in Japan and may be able to attend the meeting. SR noted that producing a free-standing feature-structure validation module would be an interesting exercise for the new metalanguage taskforce.

LR? to send current draft documents to LB for the website. LB/JUimmediate to progress letter requesting liaison status SR? to investigate feasibility of embedding Feature Structures eg in Docbook

• Council discussed documents submitted by Patrick Durusau (PD) proposing to set up a work group to redraft current chapter on graphs networks and trees. It was felt that the chapter was more concerned with graphs as an abstract formalism than with graphs as lines, and that the basis on which graphic formats were selected for discussion in PD's drafts was not clear. The immediate need for the TEI was to decide whether or not a chapter on this topic should be retained in the Guidelines.

JU and LRimmediate to produce detailed feedback which JU would pass to PD

• PW reported that the IMLS Grant application had been submitted on time, and that a response should be expected before September. Some

pre-grant activity would be expected before then. JU suggested setting up an e-mail discussion list to include Martha Crowley (the program officer).

PW1 May to set up discussion list, to include all council members who did not explicitly opt out

• LR reported that there would be a European call for proposals for "e-content creation" in December for which he was working with Alois Pichler on a project called Lyric that would involve use of FS and standoff.

• LB reported that optimism about Marie Curie funding had proved unfounded. The TEI was named in at least two competing EU FP6 bids (ECHO and Hyper). Outcomes would be known in May.