Timed Text Working Group Teleconference -- 11 Sep 2014

<trackbot> Date: 11 September 2014

<scribe> scribeNick: nigel

This meeting

Geneva F2F September

https://www.w3.org/wiki/TimedText/geneva2014

courtney: My goal is to have sample code that will map from TTML to WebVTT and back the other way and also a paper that
... describes how to do the mapping both ways. I've made a lot of progress - it won't be finished next week. I want to go through the TTML attributes.
... Andreas has also been working on a mapping and we've been coordinating our efforts. He's focussed on the EBU-TT-D profile primarily. He also wants to present his work.
... I've identified some areas in both specs where there is no equivalent, and it would be good to highlight those.

nigel: +1
... Is there a good version of WebVTT we should be referencing?

courtney: I'm using the most recent draft, which I think will become 1.0.

<courtney> http://dev.w3.org/html5/webvtt/

nigel: Great, we'll use that as our baseline dated on Tuesday.
... Without wishing to steal Frans's thunder, he tells me EBU has offered to sponsor lunch on both days. Thanks EBU!
... I'll ask for it at 12:30.
... On day 2 we have a write-a-test hackathon - any thoughts on the test format?

courtney: There's no test suite for WebVTT yet - my team is working on that.
... The TTML test suite covers a lot of cases, so one thought I had is to convert them all to WebVTT.
... We're also looking at it from the WebVTT spec perspective.

nigel: I understood that Wednesday's work will be for testing the mappings rather than each spec.
... How much time do we need to set aside for IMSC 1 blocking issues on Wednesday?

pal: I think there's only one real issue for discussion, which is #overflow because there's a difference in understanding of the impact.
... On the other ones, largely from the comments so far I don't expect much discussion.
... On Change Proposal 28 that's something I'm not sure we need any more. We may be able to remove that.

nigel: I'll re-examine CP28 and see if I can withdraw it.

pal: Without that I think that leaves Issue-339, and I don't expect us to need more than 1 hour on that.
... I'll be happy to drive the IMSC 1 issues on Wednesday morning.

nigel: We also have an agenda item on our TTWG process.

pal: Did we resolve that?

nigel: The consensus call had an end date of Wednesday. I intend to raise it on Wednesday and conclude the topic with a resolution. So far I've heard no negative comments.

Santa Clara F2F October (TPAC)

http://www.w3.org/wiki/TimedText/tpac2014

nigel: The structure of each day is a little different than last year - WG meetings will run 0830-1100 and 1500-1800 with lunch and ad hoc meetings from 1100-1500.
... Please let me know if you need a telco facility.

pal: I suggest planning for one, and not using it if we don't need it.

tmichel: I suggest using it for specific slots rather than the entire day.
... There's no difference in terms of cost but we need to reserve the speaker phone.

nigel: I'll amend our wbs survey and reserve it.
... I expect the agenda to be TTML2 and the TTML<-->WebVTT mapping.

pal: I suggest we take 30 minutes next week to discuss the CR process and next steps and timeline for IMSC 1, based on whether we use the 2005 or 2014 process.

nigel: good point.

Action items

action-325?

<trackbot> action-325 -- Pierre-Anthony Lemieux to Check if respec has been updated for the 2014 w3c process -- due 2014-09-11 -- PENDINGREVIEW

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/actions/325

pal: Respec has been given an additional config option to select process, so that's all done as far as I can tell.

close action-325

<trackbot> Closed action-325.

action-322?

<trackbot> action-322 -- Jerry Smith to Indicate preference for updating sdp-us for ttml2 -- due 2014-08-21 -- PENDINGREVIEW

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/actions/322

jdsmith: I think focusing on IMSC rather than updating SDP-US is the better path. We're the only current implementor that I know of and I don't think we see value in revving the spec or the implementation.

RESOLUTION: We will not update SDP-US for TTML2.

close action-322

<trackbot> Closed action-322.

action-324?

<trackbot> action-324 -- Pierre-Anthony Lemieux to Draft a note for imsc 1 progressivelydecodable to make concrete what authors should take into account -- due 2014-08-21 -- PENDINGREVIEW

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/actions/324

close action-324

<trackbot> Closed action-324.

Issues

issue-344?

<trackbot> issue-344 -- Add a clarification note expressing intent behind unqualified attribute names -- raised

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/issues/344

reopen issue-344

<trackbot> Re-opened issue-344.

IMSC 1 Issues

issue-340?

<trackbot> issue-340 -- Make clear that the use of the ttp:profile attribute is not required. -- pending review

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/issues/340

pal: I made two changes to implement this, both a note and a change of wording from 'profile SHALL be associated' to 'profile is associated'.

<scribe> ACTION: frans Review resolution to Issue-340 and add review comments to issue [recorded in http://www.w3.org/2014/09/11-tt-minutes.html#action01]

<trackbot> Created ACTION-326 - Review resolution to issue-340 and add review comments to issue [on Frans de Jong - due 2014-09-18].

issue-343?

<trackbot> issue-343 -- Processing of non-IMSC/TTML namespace elements -- pending review

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/issues/343

<scribe> ACTION: frans Review resolution to Issue-343 and add review comments to issue [recorded in http://www.w3.org/2014/09/11-tt-minutes.html#action02]

<trackbot> Created ACTION-327 - Review resolution to issue-343 and add review comments to issue [on Frans de Jong - due 2014-09-18].

pal: We have some issues making blanket statements about foreign namespaces here - the rule is simpler: if it's not forbidden then it's allowed.
... I couldn't find a way to word it relating to foreign namespaces without getting really convoluted.
... Because there are some things from specified foreign namespaces from SMPTE and EBU that are permitted.
... The next issues aren't captured yet in the tracker because I wanted to make sure that the reflector didn't generate further comments.
... Those are the three issues I emailed about yesterday: issue-332, issue-342, issue-339.

issue-332?

<trackbot> issue-332 -- #cellResolution support -- open

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/issues/332

pal: I don't expect much further concern on this. The background is simple: IMSC 1 forbids ttp:cellResolution. The request was for any permitted value to be used.
... That only affects the maths for the c metric, and there's no further impact. So the proposal is to lift the prohibition.

nigel: LGTM

pal: I plan to implement these changes if I don't hear any more in the next few days.

issue-342?

<trackbot> issue-342 -- Add ebutts:multiRowAlign and ebutts:linePadding attributes to the IMSC Text Profile -- open

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/issues/342

pal: These extensions are intended to allow timed text presentation of a style that's consistent with European practice, and perhaps elsewhere in the world.
... This does require adding a feature to the IMSC layout engine, but not to the HRM. It looks like a processor can ignore them and still yield reasonable results. My suggestion is to add those features,
... even as a SHALL, and move forward with that. The only caveat is these features may appear in TTML2.

nigel: This is analogous to the IMSC 1 situation with forcedDisplay, in that TTML2 may do it differently and IMSC 2 would adopt the TTML2 way.

Issue-342: Make sure to include a note along the lines of the forcedDisplay note that TTML2 and IMSC 2 may do it differently.

<trackbot> Notes added to Issue-342 Add ebutts:multiRowAlign and ebutts:linePadding attributes to the IMSC Text Profile.

issue-339?

<trackbot> issue-339 -- Allow the use of #overflow -- open

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/issues/339

pal: F2F discussion may be needed here. IMSC 1 forbids tts:overflow="visible" because that mutes the specified extent of the region.
... This has profound implications for the HRM and because from an authorial standpoint, to me and others, the author has no idea how big the region will eventually be.
... I'm also trying to understand what the use cases are for tts:overflow="visible" provides a benefit.

nigel: There's a tension here between region size and font size, where rendered font sizes can not be known at authoring time.
... Overflow allows for some safety there.

pal: That's what reference fonts are designed to address.

courtney: What about when implementations allow users to override settings?

pal: In that case the implementation would need to know how to expand the regions to compensate, if, say, the font size is doubled.
... I'm happy to add a note to the spec to explain this.

nigel: At first a reflector response along those lines (i.e. about reference fonts) would be helpful.

pal: Notwithstanding CP28 I think this concludes all the outstanding notes on IMSC, so we have a realistic shot at moving along on our proposed timeline - thank you all.

nigel: adjourns meeting - thanks all, and see you at 0900 at the EBU in Geneva!

- DRAFT -

Timed Text Working Group Teleconference

11 Sep 2014

Attendees

Contents