W3C

Timed Text Working Group Teleconference

06 Dec 2018

See also: IRC log

Attendees

Present
Andreas, Gary, Matt, Pierre, Nigel, Cyril, Thierry
Regrets
Glenn
Chair
Nigel
Scribe
nigel

Contents


<scribe> scribe: nigel

This meeting

Nigel: Today, let's welcome Matt Simpson from Red Bee/Ericsson who has joined us.

Matt: Hi, I'm Head of Portfolio for Access Services at Red Bee, provider of live and prerecorded captioning, audio (video)
... description and sign language translation, particularly interested in the transmission of live captioning data in means
... that don't hark back to old linear broadcasting approaches like 608 or Teletext.

Nigel: Thank you!
... [round-the-tables of who is on the call]
... For today we have, in principle, TTML Profile Registry but I haven't seen any movement there yet, and neither Glenn
... nor Mike are on the call, so we probably can't cover that.
... Then TTWG future requirements, joint f2f.

Pierre: Even if we can't make progress we should spend 2 minutes on the registry - we need to get the pull requests done
... somehow.

Nigel: Okay let's do that.
... Back to the list, I don't think there's anything on CSS actions. Is there anything for WebVTT today?

Gary: I don't think so.

Thierry: Not for me either.

Nigel: Any other business or points to make sure we cover?

group: [silence]

Nigel: Before we move on a very quick look ahead to the rest of the year. Next week I will be at an ITU/EC event in Vienna
... but hope to be able to run the call as usual. Then Dec 20th is our last call of the year. There will be no call on 27th
... or 3rd January, so our first call in 2019 will be 10th January.
... The other thing to remind everyone about is our window for gathering and describing requirements for our work
... in 2019 is open until Dec 20th - things added after that may not have time for proper consideration. The plan is to
... have materials submitted for review and discussion prior to our face to face meeting on 31st January in Geneva.
... I'll add to the AOB list that ITU/EC meeting in case there's any input from this group.

TTML Profile Registry

Nigel: As Pierre pointed out we have 4 pull requests open.
... As a point of normal process, we generally allow a 2 week review period after which we assume that if there is no
... objection then we can merge the pull request.
... 3 out of the 4 open pulls here need a review before GitHub will allow me to merge them.

github: https://github.com/w3c/tt-profile-registry/pull/47

Add reference to IMSC 1.0.1 tt-profile-registry#47

github: https://github.com/w3c/tt-profile-registry/pull/47

Nigel: This has no objections, one approval, so we just need to ask the Editors to merge it.
... Any objections to doing that?

group: [silence]

RESOLUTION: Editors to merge this Pull Request please.

Fix the media type in the abstract tt-profile-registry#48

<github-bot> nigel, Sorry, I don't understand that command. Try 'help'.

github: https://github.com/w3c/tt-profile-registry/pull/48

Nigel: This pull request makes one small fix for correctness - if someone can please approve it we can go ahead and
... request that it be merged.

Pierre: I'm looking at it now.
... [just double checking it is the correct value] Yes.
... That one should be non-controversial.
... [approves]

Nigel: Thank you
... Any objections to requesting that the Editors merge this?

group: [silence]

RESOLUTION: Editors to merge this Pull Request please.

Fix #42 and #43 tt-profile-registry#50

github: https://github.com/w3c/tt-profile-registry/pull/50

Andreas: I have limited availability to review this over the next week or so.

Pierre: The note about the ebuttm: prefix seems to bind that prefix forever.

Nigel: The intent is merely to explain its usage in this document.
... The ebuttm:conformsToStandard element is no longer defined by any of the referenced specifications in the table,
... because EBU moved its definition to a different document, hence the explanatory note.

Pierre: I'll propose some alternative wording.

Nigel: Thank you.
... If everyone with an interest in this could please look at this pull request and review it that would be very helpful.

SUMMARY: @palemieux to propose a change, all to review

Reorder tables alphabetically (#51). tt-profile-registry#54

github: https://github.com/w3c/tt-profile-registry/pull/54

Nigel: It looks like this one will need to be revisited after the other pull requests have been merged, and may need to
... be updated further.

SUMMARY: Check this again after the other pull requests have been merged.

TTWG Future Requirements

Nigel: I see one new requirement issue raised, thanks to IRT (Peter tho Pesch in this case), and he also kindly sent a
... covering email describing it.
... Unless anyone especially wants to discuss it right now we don't need to, but thank you for raising it.
... For the CSS property requirement I plan to document the range of possibilities as I see them and if there is a BBC view
... about which option to choose then I will document that too. I expect this to lead to a significant amount of discussion
... and thought for how we get to a good end result.
... That's because of the overlap between TTML and CSS and the huge range of CSS properties, and also the potential
... for clashes.

Pierre: Yes, as mentioned last time, it could be that for any CSS property not all of the values can be straightforwardly
... applied. There may need to be interop or testing in some cases. I really like the idea of converging CSS and TTML, so
... no objection there, but importing in bulk is not the right way to get to interop.
... Things we could consider: we could have a document and folk interested in properties could write them down and we
... can iterate over that document. Whether that is on the Rec track or not is a decision we can make later. At first we
... should identify which properties are relevant and which are not, and write them down and tests for them. For example
... anything relating to writing direction is a non-trivial mapping.
... I can clarify that on the issue - my concern is not so much with adding functionality to TTML especially where there's
... a desire to do so, it's doing it in bulk without the proper mapping that will ensure interop and specifying it and testing it.

Nigel: Of course one of the notable things about CSS is that in some cases there are CSS properties which are considered
... to be stable and usable in the annual CSS snapshots, but the specifications are still in Candidate Recommendation,
... sometimes the reason being a lack of tests. So we should think carefully about our approach to adopting CSS properties
... that are in the snapshot, therefore widely deployed and used, but CSS tests may not be fully available.
... It's good to add this to the pile of considerations.
... Anything else on any of the TTWG requirements or our process or anything else relating to them?

group: [silence]

F2F meeting and joint F2F meeting with EBU

Nigel: This is a placeholder topic at the moment. Does anyone have any questions or points to raise on this topic?
... (I don't think there is any development since last week)

group: [nothing to discuss]

CSS

Nigel: CSS published a WD of CSS Text Level 3 today

CSS Text Level 3 Working Draft

Nigel: I thought from memory this would have our line alignment requirement in, but I can't see it right now.

AOB

Nigel: I've been invited to present and participate in a panel at the ITU/EC Accessible Europe: ICT for ALL event next week.
... If anyone wants me to raise or mention anything in my role as Chair here then please let me know.

Accessible Europe: ICT for ALL event

scribe: I plan to talk briefly about subtitles and captions standards convergence and future standards for audio description.
... This is mainly for info but if you do want to push anything my way then please let me know.

Meeting close

Nigel: We've hit the end of our agenda for today so let's adjourn. Thank you everyone. [adjourns meeting]

Summary of Action Items

Summary of Resolutions

  1. Editors to merge this Pull Request please.
  2. Editors to merge this Pull Request please.
[End of minutes]

Minutes manually created (not a transcript), formatted by David Booth's scribe.perl version 1.154 (CVS log)
$Date: 2018/12/06 16:03:24 $