TEI Technical Council P5 Teleconference, 2026-06-22

09:00 PDT | 12:00 EDT | 17:00 BST | 18:00 CEST

Present:

• Syd Bauman (SB)
• Elisa Beshero-Bondar (EBB, Chair)
• Ulrike Henny-Krahmer (UHK)
• Martin Holmes (MH)
• Patricia O'Connor (TOC)
• Martina Scholger (MS)
• Joey Takeda (JT)
• Raff Viglianti (RV)
• Torsten Roeder (TR)

Apologies:

• Helena Bermúdez Sabel (HBS)
• Elli Bleeker (EB)

Next meetings

• P6 meeting in two days: Wed 24 Jun 26 (P6 meeting): 10:00 PDT | 13:00 EDT | 18:00 BST | 19:00 CEST
• Calendar polling discussion for the two upcoming meetings in July.

Infrastructure

• Huma-Num migration last week
  • TEIC was (inappropriately?) caught by surprise when, on Wed 17 Jun 26, sites went down.
  • Huma-Num had sent notifications a month or so ago, but communications never reached the right people.
• Infrastructure Group chaired by MS (thank you!) 🪑
  • Planning to hold regular quarterly meetings
  • Next meeting Fri 17 Jul 26 @ 08:30 PDT, 11:30 EDT, 16:30 BST, 17:30 pm CEST
  • This meeting was scheduled before our upcoming release on purpose so that details can be hammered out and TCW 22 updated, etc.
  • Peter Stadler now looking after tei-c.org and interactions w/ Huma-Num
    • Council extends a heartfelt “thank you” to Peter!
• MS reports that per Peter Stadler, not a lot will be different for release process
  • Questions to be resolved in this next month:
    • How will migrations & server changes affect our Guidelines Release process?
    • How will our process of updating the https://tei-c.org website change?

July Release

• Scheduled for 2026-07-28 (Tuesday).
• Action on release techs: Immediately following the release, review changes needed for TCW 22.
• Release techs: MH, MS, EBB
• Refridge dates (changes to prose only; any change to schema or code can only be merged into dev or a release branch by a release tech): Wed 22 Jul – Sat 25 Jul 26, inclusive
• Freeze (only release tech can merge anything to dev or a release branch): Sun 26 Jul 26 until release

TEI Conference / F2F Meeting updates

• Tentative schedule : https://tei2026.tei-c.org/schedule.html
• Discussion / request to move Council's panel ("It's Time for P6") to a later moment in the conference to make sure we can all attend and are in good shape to present. [Panel was rescheduled later thanks to JT and conference organizers!]
• Council F2F arrangements / locations at SFU discussed.

ATOP presentations + discussion:

Attribute override problem — by MH on behalf of ATOP

• ODD chaining can accumulate attribute overrides at multiple stages
• Attribute class membership is complicated: if an ancestor class is referenced to add attributes, that can override the local class that member is in
• Downstream customizations can delete a class. If a root class is deleted, does it delete all the downstream customizations?
• If an attribute is overridden in a subclass, then overridden again in an element which is a member of that subclass, should the latter customization be applied to the customization in the subclass? Logic says yes, but Stylesheets don't do that; they try to apply it to the original root class definition.
• Discussion:
  • RV and EBB agree with the logic of dropping attribute if its root class has been deleted upstream.
  • RV: Changing should be seen as a state machine – each further change only changes the previous one, there is no jumping to previous state (@source may allow to override that)
  • What's being referenced by pointers should be resolved before integration
  • RV: Similar to the state-machine principle, whatever is referenced should be fully resolved (i.e. its own chain resolved and compiled) before being applied to the caller’s customization chain.

Grouping element auto-deletion and repeatability — by SB on behalf of ATOP

• We converged on a consensus that it should actually be fine to have a sequence, and alternate, or an interleave which contains only a single element; if we remove the constraints that disallow that, then we can simply do the arithmetic in the way we normally would, and our intermediate ODD will actually be valid.
• JT: A single <elementRef> child of <content> could be understood as having an implicit <sequence> of one with minOccurr = 1 and maxOccur = 1.
• Decision: Change content model of grouping elements (<alternate>, <interleave>, and <sequence>) from <classRef key="model.contentPart" minOccurs="2" maxOccurs="unbounded"/> to <classRef key="model.contentPart" minOccurs="1" maxOccurs="unbounded"/>. This will have the effect of allowing a singleton child.

Updates from Board: community survey & funds outlook

• Diane Jakacki is assuming Treasurer responsibilities from Hugh Cayless.
• Handover meeting is happening this month.
• Board is planning a major fundraising effort in 2026–2027 to support future P6 F2F meetings
• Arrange a task force between Board and Council members to work on fundraising / grant writing for P6.

Council Homework for the P6 meeting on Wed 06-24

Each of us will make an abstract model for bibliography in UML, and (if we feel up to it), take to blueprint expression for encoding a bibliographic record.