W3C

Timed Text Working Group Teleconference

12 March 2026

Attendees

Present
Andreas, Atsushi, Gary, Nigel, Pierre
Regrets
Chris_Needham, Cyril
Chair
Gary, Nigel
Scribe
nigel

Meeting minutes

This meeting

Nigel: Agenda today quite full.
… DAPT Implementation Report,
… IMSC 1.3
… WebVTT
… Any other business?

no other business

IMSC 1.3

Nigel: There's a CfC as discussed last week.

CfC email to public-tt, 2026-02-27

Nigel: This ends tomorrow.

Namespace document

Add IMSC Text Profile 1.3 namespace document

Nigel: I see that Pierre just merged the pull request, thank you.
… Next step is to submit to the namespace repo, I'm not sure how to do that.

Atsushi: It's to open a PR on the ns repository.
… Filenames or some other details might need to be adjusted.
… I need to submit the pull request to w3c/ns repository, I believe.
… Let me take care of that.

Nigel: Thank you. Do you want a GitHub issue to track that?

Atsushi: Yes please, so I don't forget.
… There may be others for other specs in TTWG so I might need to add something there.

Nigel: I will do that after this call

imsc-tests fontVariant=sub|sup

Nigel: I noticed when reviewing my own CfC that the TTML2 test suite has some
… tests for tts:fontVariant but not for the sub or sup values.
… So even though formally for the Implementation Report we do not need implementations that
… pass such tests, it seems like good practice to have them.

Add IMSC Test Profile 1.3 tests

Nigel: I think this looks good, would be good for others to review too if they wish.
… My question here is if we should add an informative section to the Implementation Report
… saying "we don't need this, but useful information is that this test exists",
… and pointing to implementations that pass it.

Pierre: imscJS support for subscript and superscript was added, which was used to generate the reference render
… in the test suite.
… I've also implemented it in the ttconv data model.

Nigel: Since this isn't a formal requirement I'm not worried about the number of independent implementations

Pierre: There's a catch-22 though - I want to wait until Rec before merging the functionality into imscJS,
… so it's only in a branch. Can I point to that?

Nigel: Yes that's fine, it's still an implementation even if it's only on a branch.

Pierre: Other than that happy for it to be added to the Implementation Report.
… I can do it now in fact.

Nigel: Great, thank you.

Issues and pull requests for discussion

Nigel: Nothing more to discuss here that I can see.
… I plan to respond myself to the CfC, positively.
… Please others do so too.

Pierre: Thanks for the reminder. Just need to reply to public-tt?

Nigel: Yes, that works, or direct to me works too.

Pierre: I just did it.

Nigel: Any queries on this, please raise them now

DAPT

Implementation Report

DAPT Implementation Report

Nigel: The IR has changed quite a bit recently.
… You might remember discussions about the Charter text, which is relevant for DAPT.
… [describes updates, shares browser screen showing the updated IR]
… Regarding at-risk features to do with audio resource embedding or referencing,
… I've discussed this with technical architects in BBC and
… the favoured approach is to maintain external audio resources,
… which then opens up an interesting question about bundling together a DAPT document
… and its associated audio resources.
… I'd be interested in anyone's experiences doing similar resource bundling, or if there are other
… specifications we should look at. I'm aware of IMF.

Gary: You could base64 encode and use a data URL

Nigel: I'm not sure that's an improvement!

Gary: It's an option

Nigel: There are already embedded audio features in DAPT so that's not needed,
… but the preference seems to be not to do that.
… There are schemes with tar files or zip files and hashes and that kind of thing.
… I don't know if W3C specifies any like that already
… I also don't know that we need to specify it here at all, either.
… It could well be in the application space to define downstream.

Gary: That definitely seems like a problem this spec doesn't need to solve right now.

Nigel: +1

Nigel: Any questions?

no questions

WebVTT

Transition ownership to TTWG (no longer TTCG)?

Atsushi: Background is that recently the W3C Process developed in a formal way so that there
… is a clear boundary between CG and WG and their deliverables.
… Some discussion like draft WG deliverables could be listed and provided in w3.org URI space
… instead of github.io so these kind of options are now in the Process and Guide.
… Strong recommendation that for deliverables there is a distinction between CG and WG.
… Better to reduce joint deliverables.
… Joint between CGs or between WGs should be fine but across the boundary is tricky now.
… Also as TTWG published WebVTT as CRS in 2010s, the WG needs to maintain the CR stage documents -
… it is recommended to update snapshots in the 6-12 month period, so there is now pressure on the WG.
… Following that and general discussion about closing CGs in 2-3 years.
… So I would want to ask about the current status of the CG and what the Editors want to do
… on this front including maintaining the spec being joint with the CG or closing the CG or something else.
… I would want to hear some update from Editors.

Gary: Two questions here:
… 1. Transferring ownership to be solely under the TTWG
… 2. Should the TTCG be closed

Atsushi: Yes, of course I understand that closing the CG should be their decision.

Andreas: I haven't followed this in the last years but how active is the CG?

Gary: It's not

Andreas: That's what I expected

Gary: When Atsushi sent an email I think almost 2 years ago to the mailing list there were some questions
… around that.
… Silvia asked about the document about 608/708 to WebVTT conversion, what will happen to that document
… if the CG closes?
… I know we discussed the WG taking it over but we haven't had the time or energy for that.
… It could still be nice but I think it's a long term process.
… Another question: If there is work happening in CCSubs would it makes sense for folk outside W3C to
… provide feedback through the CG.
… There hasn't really been much activity since then.

Off-minutes conversation about status of resourcing for WebVTT development work, and ongoing Interop work

Andreas: You raised two questions about the consequences of closing the CG,
… one about the conversion document, the other about contributions and comments.
… Is it possible for a CG document to stay online when the CG closes?
… If people want to submit comments to WebVTT couldn't they go through the public-tt reflector?

Atsushi: If there are any other documents in the CG then TTWG could take them as non-normative
… deliverables like a WG Note or Draft Note, if co-chairs and the WG agrees. TTWG could take everything
… to maintain continuity.
… For the second question, comments to the WebVTT spec - all work is open in GitHub.
… We have a public mailing list that anyone can post to, as well as filing a GitHub issue.

<gkatsev_cloud> 608/708 to webvtt conversion doc

Atsushi: Of course commits to spec text will raise IPR issues but there is an external agreement to integrate
… updates from external parties.
… Also if there is an external group, such as in a liaison, then I believe we can have a relationship between
… groups for closed communications including liaison letters to request reviews or get comments.
… (when we have a new stage of maturity)

Andreas: Thank you. To the first point, does that mean it is not possible to maintain a CG document
… even if it is not taken by any other W3C Group?

Atsushi: There is no formal place for CGs to publish anything.
… All CG documents, under the current Process, have no official W3C status.
… But there is another GitHub organisation to cover every CG, w3c-cg or something like that.

<gkatsev_cloud> https://github.com/w3c-cg

Atsushi: Any CG could have a repository there as well as the CG itself has no strong process to suspend.
… If someone is keen to work as a CG then the group could exist.
… You may notice some pros and cons like inactive CG will not be closed anymore.
… There's a new process where if there's no activity over 6 months they will be asked if they want to be closed.
… There is no solid rule for closing a CG, only for opening one. Everything is fuzzy, so you may continue.

<Zakim> atsushi-bot-no-user, you wanted to discuss external feedback could be provided directly into repository, or if there is solid group/community we could have liaison?

Andreas: My assumption would be that even if the CG is closed there would still be a home.
… I didn't check the license agreement but possibly it is okay for someone to take over the draft and
… continue working on it outside of W3C.

Atsushi: I may need to check - in any case TTWG has an "other deliverables" section for maintaining
… non Rec track documents, so any other document could be taken into TTWG and maintained.

Gary: I looked up the URL of that document and its on a previously used version control system, mercurial,
… so they didn't close that so the links would still work. The URL will work for the foreseeable future.
… It might make sense to move it to that CG org on GitHub.
… Or published as-is as a WG Note.
… The copyright is with the contributors now and it is under the W3C Contributor License agreement.

Atsushi: To be honest my personal opinion is there should be some way to host within TTWG or even
… someone within TTWG could host the Text Track Cue community group to contribute something.
… There should be something.
… For WebVTT specifically I would want to have some progress.

Pierre: A key proponent and stakeholder of WebVTT is Apple, and their input would be really
… useful to this discussion as members of TTWG.
… I suggest we try to follow up with Apple offline, and I'm happy to be part of those discussions
… to figure out what the plan is. I can't speak on behalf of CCSubs, but if we needed a place to
… put a document and talk about it once in a while I'm sure we can make that happen.
… In terms of the CG and TTWG specifically it would be good to understand Apple's plans.

Gary: Even if the CG is closed I can't imagine it couldn't be reopened later.

Pierre: It depends on how much work is needed on WebVTT, I don't have a handle on that.

Gary: I can't speak for Apple but I imagine they wouldn't be too worried about the CG.
… In terms of the spec itself it sounds like it could be transitioned to be solely owned by the WG and it
… would be good to see if Apple has any thoughts on that.
… I'll follow up with Apple and Nigel and get their thoughts on it.

Atsushi: To be honest our WG has several related CGs, like ADCG and Text Track Cue CG, but every time
… I get pinged about if ADCG is active I answer "please don't close it".

Nigel: Thank you!
… As Chair of that group, I appreciate it.

Atsushi: Someone needs to manage CGs because if noone manages anything then they exist but there's
… nothing inside.

Gary: I think the bulk of the work related to WebVTT has been happening in the WG.

Atsushi: If there are active participants in the CG then there is no need to close it.
… Someone should take care of the CG while it exists though.
… For TTWG's point of view, I want to make clear the status of WebVTT spec.
… If it is handled by TTWG then WG needs to publish a CR Snapshot in the near future.

Streamlined publication, and at-risk features

Gary: The question is what to do with the outstanding snapshot pull request with at-risk features.
… There's agreement to close it, update bikeshed etc and start work on a new snapshot pull request.

Atsushi: You can have at-risk features in the CR, marked up as such.

Gary: Yes we can use the new bikeshed features for that.

Atsushi: What we need to do to publish the CRS is asking for Horizontal Review and Wide Review, as well
… as a WG decision for streamlined publication, which I would propose.
… The last WebVTT publication to /TR was 2019 I believe.
… Updating the CR Draft should be the most front matter for me.

Gary: That makes sense.

Atsushi: Chairs, is it possible to raise a CfC about using streamlined publication for WebVTT?

Gary: Just for how to publish the snapshot whenever its made?

Atsushi: For several years there has been a new stage, CR Draft, which can be published at any time between
… CR Snapshots without any formality, it's like an Editor's Draft.

Gary: Publish what we have now on /TR

Atsushi: Yes, also auto-publish on PR merge every time.

Gary: Sounds like a good idea. I need to update my affiliation, but aside from that sounds like a good
… idea to raise the CfC.

Nigel: I have no objection - Gary, will you do that then?

Gary: Just an email?

Nigel: Yes, for example see the one I did for IMSC recently.

Gary: Two week decision period?

Nigel: Yes. That's the minimum according to our Decision Policy.

Meeting close

Nigel: Thanks everyone. Next meeting in 2 weeks, same time UTC 1600.

2026-03-26 meeting agenda

Nigel: [adjourns meeting]

Minutes manually created (not a transcript), formatted by scribe.perl version 248 (Mon Oct 27 20:04:16 2025 UTC).