Telco 20251118

Date

Tuesday, 18th Nov, 15:00 UTC

Connect

Agenda

  • NIAC status

  • Release

Present

SB, AB, HB, BW, PC, RB, FdA, CZ, ZM, HG, FS - MarkusK, LukasP (FAIRmat)

Minutes

  • Release v2025.11

    • PC: release went fine, quite some definitions are still left in contributed

    • MK: drafting the NeXus paper is also in its finish.

    • LP: NeXus Ontology needs to be also updated

    • AB: our release cycle should also include the update of the ontology

    • PC: CI/CD could be added to generate/update the ontology automatically

    • Some PRs still needs to be merged and a patched release could include them

    • LP: can we release only regularly (e.g. once per year)

    • AB/BW: No, we could release at any time. RO should be also asked.

  • PRs:

    • NXparameter - PR #1560

      • MK: CI/CD was failing which required an Attribute ‘model’ to be moved to be a Field because ‘model was also defined as a Field. (Sphinx script handles Attribute names and Field names under the same namespaces)

      • CP: this fix is problematic, because it can always happen that a Field has the same name as an Attribute or a filename matches an Attribute name pattern (e.g. nameType partial). Rather, the sphinx script needs to be fixed.

      • MK volunteered to make a separate issue on this.

      • AB: model could be removed from the base class and can be defined by whoever needs it.

      • SB: let us check it also with RO as it was his PR.

    • Voting process and NXgoniometer - PR #1561

      • Vote did not pass, and a cleaner voting process may be needed.

      • AB: a possibility is: to define a 1 week commenting time, and then start a 1 week vote where no more comments are expected.

      • BW: specifically this PR could handle conventions differently. More discussion could be useful.

      • FdA: why not stay as now, voting starts, but comments could come in still

      • AB: this does not solve the process issue

      • HB: Let us go back to definitions and agree on: What is NeXus?

      • AB: 1) self-descriptive machine-readable standard? 2) community agreement? - In general (2) for App Defs, and more towards (1) for base classes.

      • BW: process may not be wrong, but more discussion is needed.

      • LP: while 1+1 is an option, actual experience shows that long-planned/discussed topics are still bringing issues to reviewers.

      • AB: Contributed definitions could/should be considered as a discussion period.

      • LP: Especially when a PRs is brought up to bring it to the standard.

      • AB: Let us make a suggestion for improving the constitution. What about 1+1w?

      • PC: 1w is enough for comments

      • BW: it would not make any difference

      • SB: a) it works already like this. b) 1w would not be long enough for actual vote

      • AB: actually, we did make modifications as a result of comments and merged not what the vote actually started on.

      • MK: we may not need to try stop commenting, but voting should be considered seriously; 1w is too short for vote.

      • BW: instead of 1+1w, let us call for comments and suggest a specific date for starting a vote.

      • SB: allow shortening the vote time if the vote is already certain.

      • HG: not everyone can follow all PRs and check if it really fits to local beamline needs. Not to vote, “Obstain” or “No” : not always clear what to do.

      • AB: more and more obtains are arriving. It is normal. We should all add opinions together. We can do it actually by voting.

      • HB: we need a strong agreement on the constitution and it shall be followed, or alternative processes could be proposed.

      • AB: consensus was and shall always be a priority.

      • RO: if comments make it obvious voting could be cancelled.

      • FdA: good to note that obstaining is not the same as simply not voting.

      • AB: No, obtaining is an important contribution.

      • HG: Thank you for the clarifications. it is convincing.

      • BW: a failing vote and restarting is a vote is actually is the same

      • RO: important to understand well the value of Obstaining compared to simply not voting.

    • Remaining PRs moved to milestone ‘’NXDL 2026’ and reviewed if they can go to a patch release ‘NXDL 2026.01’ by Jan 9, 2026.

Dec Telco

Please help to choose the date by responding to the poll by Nov 28.

We are planning to hold the telco in the regular slot of UTC 15:00. Check your local time to avoid scheduling surprises!