TEI Council Meeting

The TEI Council teleconference meeting took place on Tuesday 15 October 2009 from 14:00 to 15:45 UTC.

Participants: Gabriel Bodard (GB), Peter Boot (PB), Lou Burnard (LB), James Cummings (JC), Elena Pierazzo (EP), Dot Porter (DP) Sebastian Rahtz (SR), Laurent Romary (LR, Chair), Paul Schaffner (PS), David Sewell (DS, minutes).

Council members unable to participate: Dan O'Donnell.

Change in Council terms?

LR introduces a proposal from TEI Board to increase TEI Council terms from 2 to 3 years. Board is concerned that there is too much turnover with the 2-year terms, and that a large election slate is required at each election. Consensus of Council is that the status quo is working; we just need to encourage good candidates to stand for election and to continue in office so long as they are willing.

LR will report to Board that we recommend no change

Member resignation

LR reports that Arianna Ciula has recently tendered her resignation from Council, citing the demands of her new position which remove her from active involvement with TEI. Council accepts with sympathetic regrets, appreciation for Arianna's past service, and best wishes.

Discussion of Sourceforge bugs and feature requests

We discussed open Sourceforge bugs and feature requests, in that order, categorized using our color-code conventions ("green" for clear action, "amber" needing moderate discussion, "red" complex or problematic issues).

Bugs

Open items assigned to SR were not discussed as he is addressing them independently. See also the comments for each item, linked from the item number.

2812295, graphic should allow desc child* Accept LB's proposal 2714682, biblScope in Imprint?* LB wants to leave for more consideration. LR says essential for next release. Agree to leave as amber; needs more work.

Open "green" bugs: LB is clear on how to implement fixes. LB: fix and close green bugs.

Feature Requests

2830815, vesta: convert multiple schemata* SB: Vesta is still experimental, not a core deliverable. It doesn't affect next release, hold off on action. 2859355, subst should permit textual data* EP explains rationale for the request. Should be classed as amber, needing more discussion. EP: supply use case examples 2859183, Make all milestoneLike elements spanning* Needs more discussion. Amber. 2242434, element for text suppressed by the editor* LB wants specification w/examples of use and some text. SR + EP: draft something for further discussion 2728061, Add target to relatedItem* PB says we did decide to implement this; we just need examples. Main question is to decide whether target and element content are mutually exclusive. PB feels that target is for simple cases, content used for complex elements. LR has found precise use cases. LB agrees to greenlight, under mild protest. 2728068, suggested values for relatedItem/@type* We need examples. PS: supply examples 2783323, add from and to to choice* LB sees problems with the proposal, we agree to defer discussion. 2794512, move space to core module* LB doesn't think good idea; not convinced it's common enough in texts to warrant the change. JC: why not just include transcription module in customizations? LR: any bad side effects of moving it? LB: core module is already bloated. LR: we need to discuss the general nature of core modules. PS: we need more discussion of the semantics of textual spaces. Keep this amber. DS + PS prepare a short discussion of issue for our next FTF meeting. 2798963, add role to author* Previously, role was added to name, so there is some precedent. If author were member of att.naming, it would get role. LR: for scientific publishing, biblStruct is OK except for a few things, this is one. It's important to be able to identify corresponding author. We keep this amber, will discuss further. 2811234, add ref to material* LB has noted that there's nothing obvious to point to from ref her. GB: it is useful to be able to point to a description of a material from individual

material elements. LB: we don't have clear consensus, defer to FTF. Agreed. 2811239, new element object* Merge with preceding and consider jointly. 2812634, docStatus on edition* Request for this arises mostly from born-digital documents, to indicate the state of a document (edition). LR thinks we should use attributes the same way in revisionDesc. Keep this amber, needs description and examples. SR will work up a suggested implementation 2834505, cert on choice and model.choicePart* JC: the point would be to allow nested choice elements, assigning relatively probablities to each. LB has concerns and requests amber status. LB suggests that this discussion incorporate 2862161 (Allow certainty etc. in "empty" elements)

In connection with the preceding ticket, we need to revisit the use of match and related attributes to indicate certainty. GB: create a new ticket.

2411994, Define canonical way of referencing TEI element definitions* Awaiting implementation of proposed Vault archiving. 2507305, Alignment and documentation of sound files* Presents some of same issues as facsimile. Needs further discussion. Should be merged with 2724997, "Cater for audio/video facsimile" 1934586, Chapter Revision: Critical Apparatus* LB wonders why this is a Sourceforge ticket. Essentially, it's a reminder of needed work. EP: MS SIG wants to undertake a revision w/a working group. Need to identify a chair. We agree to close the ticket and ask the MS SIG to submit specific tickets as needed.

Of the Green items, per LB the only problem is 2531384, "Rationalise application of target". (The basic issue is that target is defined in several different ways for different elements, and differs in optionality and data type.) LR: it would be nice to have a single

target attribute with a defined syntax. It would clarify TEI to outside world to have single mechanism for pointing with target. SR: we can still use Schematron to constrain to a single token. Q: what do we do about the existing

targets, allowed on 3 elements? Consensus: we deprecate. Convert to Amber.

2009 Members Meeting: Council Report, etc.

LR ask that all Council members report to him via email their 2008–9 activities within the next two or three days, along with other items that should be added to the report. LR. will compile and send to Council.

All: send notes to Laurent. LR will be giving a paper on the future of ODD. We should also monitor discussion of TEI in publishing. LB raises the issue of the ACLS (North America) Humanities E-Book project, which uses a non-TEI DTD; should we get involved?LR notes that this is causing issues for e.g. European publishers like revues.org who are wondering why to use TEI if Americans are doing something different. One of our goals should be to compose a single recommendation for academic publishing. (DS and others pursued this on Council email following the telco.) We agree to mention this in our Meeting report.

All: send LR examples of TEI in publishing (especially scholarly / scientific).

Final Points; Next Face-to-Face Meeting

valList is a sibling of content. Any further changes will need more discussion.