Timed Text Working Group Teleconference

08 Jan 2015

See also: IRC log


glenn, nigel, tmichel, pal, Andreas, jdsmith


<trackbot> Date: 08 January 2015

<scribe> scribeNick: nigel

This meeting and future planning

nigel: Any other business?

pal: ITU-R liaison
... Also the comment from Simon Hailes on IMSC1

nigel: also a F2F possibility
... Looking at the TTML2 timeline, we had said that we'd freeze the ED on 4th Jan prior to the FPWD.

glenn: I'm trying to get into place spec text for all proposed new features. There's one more that
... I've been contemplating, which is adding support for an anchor link functionality.
... I wanted to get all the 'new syntax' issues knocked off. The other issues that don't require
... new syntax I will address either via an Editorial Note or by editing directly.
... I'm trying to finish that all up this week and then get a document ready for FPWD review.

nigel: We will have a 2 week review period for the group before requesting publication.

glenn: We should be able to target the review period starting on Monday.

nigel: Will that have addressed all the P1 change proposals?

glenn: I believe so, in terms of new features, but not necessarily all the new material, e.g.
... the HTML mapping, which doesn't require new syntax and can be done later.

nigel: Okay it looks like we need to push all the dates back by 1 week.

tmichel: After the 2 week review the group will okay FPWD publication?

nigel: correct.
... Thank you Pierre for putting together a graphical view of our publications timeline:


nigel: When I looked at this it seemed to me that a F2F might be needed around April to
... help stick to these dates.

group: discusses potential locations too - preference for west coast US due to other commitments

nigel: Are there any events that we would need to avoid?

pal: We could do it around NAB (April 11-16) in Las Vegas.

group: concludes that a west coast US location just before or after NAB is the most promising option.

<inserted> tmichel: I would largely prefer before NAB, not after.

Action Items


<trackbot> action-355 -- Glenn Adams to Resolve duplication between issue-357 and issue-229 -- due 2014-12-04 -- PENDINGREVIEW

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/actions/355

close action-355

<trackbot> Closed action-355.


<trackbot> action-362 -- Glenn Adams to Raise an issue for font-kerning in ttml2 -- due 2014-12-25 -- PENDINGREVIEW

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/actions/362

glenn: The requirement for font kerning came from Japan. This was raised as Issue-359.

close action-362

<trackbot> Closed action-362.


<trackbot> action-358 -- Pierre-Anthony Lemieux to Draft response to itu-r liaison re imsc 1 questions. -- due 2014-12-11 -- OPEN

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/actions/358

nigel: Thanks Pierre for drafting this and including a response to the ITU-R question on IMSC1 and WebVTT.

pal: ITU-R Working Party 6B came back to us with a set of specific questions. The liaison response
... tries to answer them factually.


pal: The first question asks about the relationship between IMSC 1 and WebVTT.
... The second question was about progressivelyDecodable and its possible usage.
... The third question was about support for high frame rate video.

glenn: I would edit the last sentence on that to begin with 'if' so that it doesn't get missed.

pal: I'll edit it now.
... Point 4 is about text profile and image profile and why they are mutually exclusive.
... I'll post the revised document.

nigel: I'll take it from there and post the response.

close action-358

<trackbot> Closed action-358.

<scribe> ACTION: nigel Format and send ITU-R liaison response [recorded in http://www.w3.org/2015/01/08-tt-minutes.html#action01]

<trackbot> Created ACTION-363 - Format and send itu-r liaison response [on Nigel Megitt - due 2015-01-15].


<trackbot> action-360 -- Nigel Megitt to Send liaison request on behalf of w3c ttwg to arib -- due 2014-12-18 -- OPEN

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/actions/360

nigel: I've been held up on this partly due to the holidays but also because they want a piece of
... paper to arrive, with W3C branding and my signature on it - I'll follow up with Daniel from
... W3 staff on this to see how we can complete this action.



<trackbot> issue-360 -- Use of content encoding -- raised

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/issues/360

glenn: I would suggest we recommend gzip encoding for transfers if the client supports it.

jdsmith: Is this for data compression?

glenn: yes

nigel: We don't really mention transfer at all at the moment.

glenn: We do have a section on concrete encoding - I thought this could go in the same place.

reopen issue-360

<trackbot> Re-opened issue-360.


<trackbot> issue-361 -- Add ttm:mediaTimestamp (or equivalent) attribute. -- raised

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/issues/361

glenn: In HLS there's a way to specify a timestamp map to associate MPEG-2 presentation stamps
... with document times. I welcome feedback on this one.
... I thought we could add a ttm: metadata attribute.
... If it's a bad idea I wouldn't mind dropping it.

nigel: I know that this is likely to be a problem that needs solving, and people working on it,
... but I'm not sure when we'll have the solution for it.

glenn: We can open this issue and possibly add an extension spec for something like this.

nigel: Doing that as a Rec would need a new Charter but we could certainly do it as a Note.
... I suggest we open this and defer until those folk putting TTML into MPEG-2 can define
... the problem space better.

glenn: I'd like to associate this with TTML2 for now and possibly move to v.next as part of a
... deferral process.

reopen issue-361

<trackbot> Re-opened issue-361.


<trackbot> issue-362 -- Handling forward interoperability of attribute extensions in TT namespaces. -- raised

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/issues/362

glenn: In TTML1 under the definition of an abstract document instance there's some language
... about what constitutes a valid abstract document. It handles new vocab that is unknown to
... the processor that is in one of the TT namespaces, like TTML2 adding a new element. But the
... language didn't handle attributes in TTML namespaces, e.g. a new style attribute like
... fontKerning. The language in TTML1 under that would not correctly address it but I think the
... intent was to do that.

reopen issue-362

<trackbot> Re-opened issue-362.


<trackbot> issue-363 -- Add support for font shear -- pending review

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/issues/363

glenn: This one came from an industry standard called Lambda Cap. I've already moved that to
... Pending Review by the way, having implemented it.

nigel: I'd encourage review of the solution.


<trackbot> issue-364 -- Relax constraint between the extents of the root container and the dimensions of the related video object frame -- raised

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/issues/364

pal: This was my attempt to capture Simon Hailes comment and resolve it, based on my
... discussions with him and others. My next step would be to implement in the IMSC 1 ED.

glenn: We should coordinate on that because I'm drafting some new language based on mapping
... to the related media object's viewport from the document's viewport.

pal: Okay. In a nutshell IMSC 1 currently forces the author to specify a root container extent
... size in pixels when images are present, and forces that extent to be the same as the related
... media object. The feedback is that this prevents usage of one document with multiple
... resolutions of related video objects. The proposal is to remove the constraint. It forces the
... renderer to be able to scale PNGs to match the related video object.

glenn: It forces renderers to be able to scale the composited root container region content into
... the related video object as opposed to scaling the individual images.

pal: You said it better!
... The feedback I've received is that this is a good idea in the multi-resolution world. I intend
... to implement it.

reopen issue-364

<trackbot> Re-opened issue-364.

jdsmith: So this affects IMSC only and not TTML in a broader way?

pal: Yes, because image is a SMPTE-TT extension. It affects IMSC 1 image profile documents but
... not other profiles.

glenn: The general issue is certainly going to be present in TTML2 because it will have image
... support. The general issue of mapping the root container region into some final region
... associated with the related video extent does apply to TTML1 too.

nigel: Looking at the remaining issues, did we resolve the ttm:item points?

glenn: I implemented the namespace resolution for ttm:item values in TTML2.

tai: I think we can go ahead with this from my perspective.
... I think a way out of this is to avoid listing predefined names or items but list the concepts
... that may be included.

glenn: I know that Pierre had that concern about not including smpte namespace metadata items.
... I have already added an editorial note on the CEA608/708 metadata items. It's not my
... intent to diverge from the existing syntax.

pal: It's hard to have a final opinion if the intent is to provide further information. The decision
... is on us to go ahead with a FPWD with incomplete information. I'm open minded about this.

glenn: There are other parts of the spec that are incomplete too, but that's okay for a FPWD.

nigel: I don't hear any objections to closing the ttm:item issues so I'll go ahead after this meeting
... and do that by email.
... Thanks everyone, we're 7 minutes over so I'll adjourn the meeting - meet at the usual time next week, for 1 hour.

group: wishes each other a Happy New Year.

Summary of Action Items

[NEW] ACTION: nigel Format and send ITU-R liaison response [recorded in http://www.w3.org/2015/01/08-tt-minutes.html#action01]
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.140 (CVS log)
$Date: 2015/01/08 16:25:13 $