Timed Text Working Group Teleconference

<trackbot> Date: 30 July 2015

<scribe> scribe: nigel

This Meeting

nigel: AOB? From me: the new W3 Process

group: no AOB

Action Items

action-409?

<trackbot> action-409 -- Pierre-Anthony Lemieux to Discuss liaison with unicode with richard ishida -- due 2015-07-09 -- PENDINGREVIEW

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/actions/409

nigel: Thanks pal for the email to the reflector earlier.

pal: I'm happy to draft that as a note for Unicode.

nigel: Your update addresses my feedback, and you've clarified that when localizedSets reference
... commonSets they are *including* by reference.

Action-409: [Meeting 2015-07-30] pal to draft a more formal text to send to Unicode, then check in with Richard Ishida, then Nigel (as chair) to send to Mark at Unicode when it's okay.

<trackbot> Notes added to Action-409 Discuss liaison with unicode with richard ishida.

nigel: I've mentioned the idea of this to at least one other group I'm working with and they wanted more detail but were in principle in favour, and could lend their support.

pal: I'll add this additional more concrete information to the draft liaison.

nigel: Sounds perfect.

Issues

issue-398?

<trackbot> issue-398 -- Add IMSC1 in title of IMSC1 -- pending review

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/issues/398

nigel: Pierre has done this - it's a change of title of IMSC 1 as discussed.

pal: That's correct.

nigel: So now it's "TTML Profiles for Internet Media Subtitles and Captions 1.0 (IMSC1)"

close issue-398

<trackbot> Closed issue-398.

glenn: So you added (IMSC1) at the end.

nigel: and subtracted "Text and Image"

glenn: Looks good.

nigel: I have another issue that may be worth raising. Consider a ttm:agent that contains a
... ttm:actor, which references that ttm:agent.
... A naive implementation could process all the actors, find one that references its own parent,
... follow that through, add an actor to the list, and loop forever. I don't think there's a substantive
... problem with the spec, and it's a function of the naivety of the implementation, but we could
... add an Informative Note to point out the problem at least.

glenn: I never thought of that possibility. I wonder if it makes sense semantically.
... Let's open an issue for it, and we may want to create an errata for TTML1 that excludes that
... usage, unless we can think of a usage.

nigel: Okay I'll do that. The other possible resolution is to add an informative note warning
... developers about the possibility. I think it's the only place in the spec where there is any chance
... of having a circular reference in that way.

<scribe> ACTION: nigel Raise issue regarding actor and agent references [recorded in http://www.w3.org/2015/07/30-tt-minutes.html#action01]

<trackbot> Created ACTION-410 - Raise issue regarding actor and agent references [on Nigel Megitt - due 2015-08-06].

nigel: Another resolution might be to remove the ability to contain an actor in agent.

glenn: That's a major structural change.

nigel: Yes it would be!
... Okay I'll go ahead and raise that issue against TTML1.

glenn: It'd be interesting to see what I did in the TTV verifier for this. I don't think I have a test case for it.

TPAC 2015

nigel: We have a wiki page, registration is open, we have a schedule.
... For the agenda, I think we need something about HTMLCue and meeting with HTML WG,
... something about charter renewal, presumably some TTML2 review, hopefully some TTML-WebVTT
... mapping review work also.
... I may add to the Charter discussion some work BBC has been doing on responsive subtitles
... where work would be needed on the presentation behaviour, which would likely be
... interesting to many people. Also on a similar topic, APIs for modifying presentation dependent
... on other on-screen user interface elements that may clash with subtitles or captions.

glenn: That seems like a very generic question. It seems like it's out of the scope of TTML but
... in scope of how TTML is embedded in some processing environment.

nigel: Well since condition was introduced to TTML2, if you're prepared to permit on the fly
... processing of conditions then that could be used to do this, by setting parameters that
... result in different presentation.

glenn: That's true, there are a lot of ways to conditionalise the content. Certainly you could
... author the content based on defined states.
... Do they have specific scenarios in mind?

nigel: Yes, they do, such as "controls are visible", "advert is visible" etc. but the more general
... mechanism might be to signal priorities and layers and areas of the screen to avoid or prefer.
... Or another way would be to expose as an API some of the processing details of the TTML and
... allow e.g. javascript events to modify positions as a result.

glenn: You could have a generic accessor for environment variables that could be used to affect
... presentation. That content wouldn't be truly portable unless we define the names ourselves,
... but if we're doing that then we don't need the generic getEnv('xxx') approach.

IMSC CR2, Test Suite and Implementation Report

nigel: Frans pointed out recently that https://dvcs.w3.org/hg/ttml/file/tip/ttml-ww-profiles/testsuite/multiRowAlign/multiRowAlign1.ttml
... isn't actually EBU-TT-D compatible.

Frans: Yes, that uses named colours - it would be nice if they could be hex colours. There are one
... or two other changes too. I can summarise them in an issue.

nigel: The wider point for me is: would it cause any problems to change tests at this stage?

pal: I can't think of any reason not to do it.

Frans: Shall I raise this as an issue then?

<scribe> ACTION: Frans Raise an issue requesting that IMSC 1 tests for multiRowAlign and linePadding be made fully EBU-TT-D compatible. [recorded in http://www.w3.org/2015/07/30-tt-minutes.html#action02]

<trackbot> Created ACTION-411 - Raise an issue requesting that imsc 1 tests for multirowalign and linepadding be made fully ebu-tt-d compatible. [on Frans de Jong - due 2015-08-06].

AOB

nigel: I wanted to mention the new W3 Process document.
... It's at http://www.w3.org/2015/Process-20150901/
... Changes are summarised at the end. This will apply to us from 1 September with no
... further action required for us to adopt it.
... The only thing I can see that could affect us is §6.3.2 that recommends us to publish new
... WDs of WDs that haven't changed in the last 6 months. That might apply to WebVTT and TTML2
... if we haven't published a new version by 1 September, but it's a SHOULD not a SHALL.
... We've hit the end of our agenda, so I'll adjourn. Have a good week everyone! [adjourns meeting]