Email from Geoffrey Rockwell, as amended by Lou Burnard
Attending: Christian Wittern (CW), Perry Willett (PW), Geoffrey Rockwell (GW), Sebastian Rahtz (SR, board representative), Merrilee Proffitt (MP), Syd Bauman (SB, editor), Matthew Driscoll (MD), Lou Burnard (LB, editor), John Unsworth (JU, chair), David Durand (DD), Laurent Romary (LR)
It was agreed that Geoffrey Rockwell would take notes.
John Unsworth provided an update on the Consortium. The 2002 budget (handed out in Pisa) predicted 57 members and 50 subscribers. We now have 49 members so we are on track. Total anticipated income $357K. $16K budgeted for workgroups (separate from monies spelled out in NEH grant).
There was a suggestion that the list of institutional members
be moved to a more prominent spot in the web site as that is what
impresses administrators when one is making the case that your
institution should be a member. Lou Burnard said he would look
into that.
We reviewed the TEI-C Council Terms of Reference. Our priorities are to oversee changes and long range planning.
Lou Burnard provided an update
148 changes from the sublime to the enormous, 6 outstanding
issues. It is still the intention of the changes to not break any
existing documents. Lou and Syd will produce a list of
outstanding problems.
We talked about how workgroups submit their work and how that
integrates with ODD. There is a document (EDW55) that
describes how draft chapters should be submitted which this group
can look at, but there is no urgency to reviewing this (unlike
the urgency to reviewing EDW54
– see below.)
One of the issues that arose in Pisa was whether or not to
include for each element information about the element’s parents
and children. Various people at Pisa felt this was useful when
using the paper guidelines. Lou felt this could be misleading
since the relatives listed are not necessarily available. It was
agreed that the parents and children should be included in the P4
in the manner they appear in P3.
- 7 votes for P3-like text
- 1 vote for dividing the information by tag set
- 1 vote for leaving relatives out
In the course of our discussion we discussed the possibility
of providing a custom documentation tool that would produce a
version of the Guidelines tailored to the project which could
have the relevant relatives.
The result of a call for changes to P4 didn’t turn up much.
Julia Flander’s questionnaire had a question that may have turned
up some P4 issues. John U should check with her.
The Ibsen suggestions seem more like P5 issues, since the
intent of P4 was mostly to bring P3 into line with XML. Ibsen
suggestions should be deferred to a group that will review
We reviewed what needs doing to finish the P4. There are three major tasks:
- Including the information about parents and children back in (see 2.2)
- A bunch of little things
- Revising two chapters – the XML chapter and the Character Set chapter (see below for discussion of both.)
The Council was asked for input on chapter 2, esp. 2.11 which
Lou felt gets dated quickly. We felt that 2.11 should stay and
be updated with reference to the W3C, stylesheets, and pointers
as to where to get more updated information.
We had a discussion about URLs and how to include them. Some
of the hacks mentioned included adding a URL attribute to xptr,
xref, and figure (Sebastian’s hack). We wondered if we should
have a web page of hacks captured in the wild? We agreed that
the editors will look at this and that a note will be included in
chapter 14 that mentions Sepastian’s hack. A proper solution will
have to wait for P5 and solutions from W3C.
Christian summarized the work on chapter 4. The working group
found it difficult to deal with both the SGML and XML issues
around character sets. It was agreed that the chapter should
concentrate on XML with notes about SGML where it differs. It was
noted that the xml:lang attribute and the TEI lang attribute
overlap but do different things so we need to keep the TEI lang
attribute. The title of the chapter might be changed to
“Languages and Character Sets”. Issues with the WSD also need to
be dealt with. An appendix of all the acronyms would be useful
along with examples of the differences between character and
glyph. In short the chapter needs work. The editors will return
an edited version of the chapter quickly (end of Feb.) to the
working group which will then try to get it back to the editors
by the end of March.
We discussed when P4 will go to print and how it will be
printed. Lou estimated 3 months to finish the P4. John is going
to try to get an estimate for print-on-demand costs so that we
can compare that method to massive print job of 1000
copies.
- End of Feb. – Draft sent out (with Character Set chapter rewrite)
- End of March – Comments solicited for no later than end of March.
We had a general discussion of P5. We will not be constrained by trying to avoid breaking existing TEI-conformant documents in development of P5. This led to question as to what should happen with P4 when P5 comes out and what we should say about the life span of P4. Do we freeze P4 or accept only corrections for a fixed amount of time? We need to provide guidance to people who want to send suggestions.
Sebastian felt we should keep P4 around for 5 years (after P5 comes out), support software (Pizza Chef) and fix things that are actually broken for that period.
The question was raised as to whether we should say that P4 was the last to support SGML. Lou and Sebastian didn’t want to go that far but it was felt we could suggest that new users use XML.
We agreed that we don’t need to make announcements about the future status of P4 until P5 is out.
We had a discussion about whether TEI-C Council members who
were not TEI members could have access to the member’s
area. John is going to e-mail the board about this but believes
they need access in order to do their work.
Lou reported on the areas where people had suggestions for
P5. His list is based on a talk he gave in Pisa (see
- Character Encoding
- Core Tags and Schemas
- Metadata and Miniheaders
- Source Description
- Transcription of Primary Sources
- Stand-Off Markup, Xlink and Xptr
- Term Banks, Dictionaries and Ontologies
- Tags for Authoring
- Multimodal Communicative Acts
- Metrical Annotation
- Image and Multimedia Annotation
- Core Tags and Schemas
We first agreed that 3.3.2 was the business of the whole Council so we removed it. We then went through a process of selecting our top 2 issues to be dealt with by a workgroup. The two that got the most votes were Character Encoding and Stand-off Markup, Xlink and Xpntr.
In the second round of voting the two issues that got the most votes of the remaining issues were Metadata and Miniheaders and Transcription of Primary Sources. Thus we have the following priority sequence:
Core Tags and Schemas – For the entire council Character Encoding Stand-Off Markup, Xlink and Xptr Metadata and Miniheaders Transcription of Primary Sources Source Description
The general strategy for P5 is:
- Look at what the W3C is doing first
- Don’t do things that others have developed standards for (like mathematics)
- Commission workgroups for what is left over
We agreed to commission two workgroups (given the money we have).
Christian Wittern was voted chair of the Character Encoding workgroup. The workgroup gets a budget of US $8,000. The workgroup has the immediate task of working with the editors to rewrite chapters 4 and 25 and this will involve the existing members and can be considered a continuation of their current work. (See above [AI11].)
Once done with the P4 chapters the workgroup has is renewed
for a term of one year and should meet at least once. The new
membership is up to the chair. The workgroup should report at the
Oct. 11th meeting. Their task is to review the proposals in
extenso and propose revisions to the way character encoding is
handled in general. This will involve a rewrite of chapters 4 and
25 – to reconsider the subject of character encoding and
writing system declarations. A workplan is called for by the end
of April, preliminary report in October, and final report in
January of 2003.
There is a document (EDW 54) on
how to make workgroups work. We need to all review this and send
suggestions to editors.
We commissioned a workgroup on Stand-Off Markup, Xlink and
Xptr with David Durand as chair. David has the same schedule with
the additional request that he inform us by the end of February
as to who the members are. This workgroup will also have a budget
of US $8,000. We do not have a formal charge for this group; one
should be drawn up.
For the record it was mentioned that one editor should attend each workgroup meeting.
John updated us on the NEH grant. The funding came in May. There was a problem with the TEI tax status so it is the University of Virginia that administers the grant. The expenses have to be billed over a 2 year period. The main thing the Council had to do was select people for task 2 (SGML to XML TEI texts conversion.)
We came up with a list of names for the 6 experts:
John Price-Wilkin (John PW), Wendell Piez, Michael Popham, Jessica Perry Hekman, Sue Fisher, David McKelvie, Frank Tompa
We decided that John should contact John Price-Wilkin and see
if he would chair the group. If he agrees he should then be
provided with our suggestions for names, but allowed to select his
team.
We also came up with a list of possible people for the 10 who have experience with SGML texts that need converting. This list will be passed to John PW.
Julia Flanders, Lealle Fredland, Perry Willett, Louis Barch, Perseus Project, Nancy Kushigian, Tomaz Erjavec, Peter Flynn, Slave Narratives Project, Sebastian Rahtz
We would recommend to the chair of this project that he/she
make a open call for groups with significant TEI SGML holdings who
want to participate. The chair, who we hope will be John PW,
should be provided with a schedule that includes a report to the
board in May.
We discussed the request from the ESTATE project to write a
letter of support for their grant proposal and approved it. The
proposal is to develop TEI training, deliver it in Europe, and
support two workgroups. We discussed the issues around the
protocol of other groups running TEI workgroups. John U will
contact the leader of Estate and clarify the language in the
grant. We did feel this grant proposal was worth supporting as it
will not only develop and deliver TEI training, but it will also
pay for workgroups we cannot afford. Anyone who is interested in
the grant should look at the proposal and get comments back to
John as soon as possible.
We discussed training and whether TEI should certify training
or develop training. We decided to form a committee made up of
Geoffrey Rockwell, Julia Flanders, Perry Willett, and Sebastian
Rahtz. The committee should look at existing training and develop
a training strategy for the TEI. The committee should consider
the question of what people are willing to pay?
We discussed certifying Tools and Texts. To be able to certify
things we may need to trademark a name and/or logo. John will
look into that.
We decided that a committee should look generally into the
issue of the TEI providing consulting. The members of the
committee will be Matthew Driscoll, Merrilee Proffitt, and
Laurent Romary. They will develop a consulting strategy for the
TEI which could include consulting on software or texts.
We discussed getting copyright clearance from all workgroups
and Council members. John will work with lawyers to develop such
a document and we will all have to sign it. This is to ensure
that the TEI-C has a license to use all the materials developed
for it.
The next meeting may be Oct. 11 in Chicago depending on funding. Rockwell was asked when these minutes would be ready and promised them for Friday.