TEI Technical Council Teleconference, 2024-05-15

10:30am PDT / 1:30pm EDT / 5:30pm UTC / 7:30pm CEST

Transfer of the TEI Listservs

Release techs: HBS, GR, TR
Dates (postponed by one week):
- Refrigeration: Fri 14 Jun → Fri 21 Jun 2024 at Council meeting
- Freeze: Sat 22 Jun → Sat 29 Jun 2024
- Release: Mon 24 Jun → Mon 01 Jul 2024

When: this summer up to TEI Conference
Council Tentative Plan to meet on Mondays: 10:00–11:30 PDT | 13:00–14:30 EDT | 17:00–18:30 UTC | 19:00–20:30 CEST
HBS will use our preferred date / time from the when2meet for Stylesheets Meetings.
Next Stylesheets Meeting is Monday May 27 at this new time.
Council decision to move June release dates forward by one week and June Council Meeting on 21st June (earliest available time)
Meetings in July and August at designated new times:
- July Council Meeting: Monday July 8
- August Council meeting: Tues Aug 13

Travel planning notes / updates
Conference proposals were due May 10; review stage beginning
Separate Slack channels for
- #f2f-planning
- #tei-conference (more general to share with board / non-Council members in our Slack)

Time to find something better than Slack for preserving our messages?
See Slack anti-Slack discussion
Check out Matrix
HBS: Huma-Num has a Mattermost instance: https://documentation.huma-num.fr/mattermost/
TR: Form a Task Force to investigate alternatives
- EBB, JT

SVF2F April 27 Council decision: intention to include the CMC chapter in the next release
#2554 Introduction of the new attribute class (planned for a mentoring session: May 17th)
- Where is this class to be available? Should not be limited to CMC module–put it in the tei module.
Action SB and EBB: Update the cmc branch before the introduction of the new attribute: update before Fri 17 May 24.

JT working on #678, #680, and the PR #681 [But cf. #645, which may also be related]
Stylesheets are currently breaking customizations that weren’t broken before.
JT’s PR is a simple fix, we hope
Lou Burnard has some chained ODDs to test.
Action on JT to post sample chained ODDs by 2024-05-16 — DONE
Action on SB to test (on cmdline) by 2024-05-22 — DONE
Action on JT to check with Martin Holmes on how to test in oXygen (without the published plugin) by 2024-05-22

Trouble for several Council members, especially when setting up on new computers
What do we need to make this better?
- Mentoring session can't solve all problems in a 90-minute call.
- What's going wrong for us
  - Windows OS
  - Problem is often for Docker to work in Windows OS
  - Docker container difficult to setup
- Variety of tests / builds we run
Who has good setup for tests just now?
- - MT (on Mac + Docker)
    - MS on Windows (Linux under Windows with WSL2, following Hugh’s instructions on Building and Testing the TEI Guidelines and Stylesheets)
    - HBS (Ubuntu + Docker)
    - EBB (Mac + Docker)
    - SB (Ubuntu, but no XeTeX, so no PDF testing)
- Peter Stadler: Jenkins tests set on branches
- Peter Stadler can walk us through the Docker setup
  - Action on HBS to invite Peter Stadler to May 27 Stylesheets Meeting
  - Suggestion: have GitHub testing do more, i.e. entire test suite and building artifacts.
  - Make sure this meeting also addresses the GitHub testing / GitHub Actions environment. RV can assist with this but not until July. JT can also help out.

Guidelines PRs
- #2509 address #1981 with SchXslt (aka mausatron) instead of skeleton
- #2540 add missing <sch:rule> elements
- #2552 constrain attributes of elementRef
- #1996 align teidata.version with Semantic Versioning Specification
Stylesheets PRs
- #682 styling of new attr lists, fixes #657
- #681 Fix regression: duplicate idents in *Specs

Q: Why is <idno> (rather than <name>, <head>, or <label>) a child of <moduleSpec>?
- A: Because that is what we changed <altIdent> to, because <idno> made much more sense than <name> for an FPI, and this portion of the specification of a module is used in its FPI.
tei_simplePrint.odd still has two IDs duplicated (both “page1” and “OTASH” occur twice)