- From: Nigel Megitt <nigel.megitt@bbc.co.uk>
- Date: Thu, 23 Nov 2023 16:57:25 +0000
- To: TTWG <public-tt@w3.org>
- Message-ID: <ACDC3A2B-2596-4CF1-B731-C60F324C4D96@bbc.co.uk>
Thanks all for attending today’s TTWG call. Minutes can be found in HTML format at https://www.w3.org/2023/11/23-tt-minutes.html
In text format:
[1]W3C
[1] https://www.w3.org/
Timed Text Working Group Teleconference
23 November 2023
[2]Previous meeting. [3]Agenda. [4]IRC log.
[2] https://www.w3.org/2023/11/09-tt-minutes.html
[3] https://github.com/w3c/ttwg/issues/269
[4] https://www.w3.org/2023/11/23-tt-irc
Attendees
Present
Atsushi, Gary, Matt, Nigel, Pierre
Regrets
Andreas, Cyril
Chair
Gary, Nigel
Scribe
nigel
Contents
1. [5]This meeting
2. [6]IMSC-HRM
1. [7]IMSC-HRM Tests
3. [8]DAPT
4. [9]Meeting Close
Meeting minutes
This meeting
Nigel: On the agenda today we have:
… Some issues and pull requests relating to IMSC-HRM
… The Registry pull request for DAPT
… There may be some other details about those two specs to look
at too
… Is there any other business, or points to make sure we cover
within those topics?
IMSC-HRM
Nigel: I'm a bit sheepish that I'm just raising issues at the
moment when we said we are about
… to request exit from CR.
… The background is I only just got around to doing
implementation work for it and that helped
… me spot some things.
… One of the changes was a clarification about glyph processing
in IMSC-HRM itself.
Pierre: I think it's a good clarification
[10]Clarify the behavior for duplicate glyphs w3c/imsc-hrm#71
[10] https://github.com/w3c/imsc-hrm/pull/71
Nigel: I think we could discuss if that's purely editorial or
needs 2 week review
Pierre: I think the risk of closing today is minimal given
where we are.
… I would support merging it today.
Nigel: [shares diff on zoom]
… Any objections to merging today?
group: no objections
Nigel: I think we have adequate consensus for that, please go
ahead and merge it.
Pierre: doing that now
Pierre: On issue 69 I think it's worth adding a note and would
like your opinion
[11]Note on document changes at every frame of the related
video w3c/imsc-hrm#69
[11] https://github.com/w3c/imsc-hrm/issues/69
Pierre: This is related to converting from 608. The HRM
document references 608
… in the context of the glyph cache.
… At first I thought this could be the subject of a separate
conversion document,
… but since there's a reference I plan to add an informative
note where the spec talks
… about 608 and 708.
Gary: That makes sense to me. I've seen lots of related issues
where naive implementations
… make a change on every character.
Pierre: I tried to write the note a couple of times yesterday.
… Rather than broadly discussing conversion, I think it's best
to say that the HRM does not
… support one document per field.
… At some point we could write a WG Note with guidelines for
conversion.
Gary: There is an old document for 608 and 708 into WebVTT.
… It's probably worth brushing it off and updating it or using
it as inspiration for a new document,
… and including both IMSC and WebVTT.
Pierre: My other favourite problem is exact region sizing for
the number of characters.
… I'll propose a PR with that note.
… The other thing we need to address is 2 outstanding issues in
the CR of IMSC-HRM.
… The first is image profile being at risk.
… I'm pretty comfortable with removing it before we move
forward.
… I've seen zero interest.
… We can always add it back in the future.
… If someone has a significant objection it would be good to
know.
… Unless there's new information I recommend removing support
for image profile from this version of IMSC-HRM.
… The second is, in section 5 relating to the complexity of
drawing span elements with background colour.
… The way IMSC-HRM is currently specified is the way it's been
forever.
… I had concerns that this would fail documents that contain a
large number of spans with background colour.
… So far, in all the tests that I've seen, it's never been an
issue.
… I'm tempting to err on the side of no change and closing #5
as "will not fix".
[12]span elements are included in NBG(R_i) w3c/imsc-hrm#5
[12] https://github.com/w3c/imsc-hrm/issues/5
Pierre: I think we can note in the implementation report that
we have seen no documents that
… conform to image profile, in support of removing that
feature.
… Secondly, that outstanding issue #5, where the HRM as
currently written has not resulted in issues
… with background colour, so I'm comfortable with closing as
will not fix.
Nigel: I want to spend time reminding myself about issue 5.
… I have no objection to removing the at-risk image profile
feature.
… I think there are people using image profile, but they
haven't been engaged with the HRM.
Pierre: I agree, I think they use it in closed systems and the
HRM is not relevant for them.
Nigel: It would be nice to have a positive statement about
that;
… working on lack of feedback is a weaker position to be in.
Pierre: It's really easy to plug back in if we need to, without
affecting anyone else that was happy.
… That's a different situation for issue 5.
Nigel: I see that issue 51 is assigned to me, I think I need to
double check that and hopefully close with no change.
Pierre: We should try to plan on requesting transition before
the end of the year.
Nigel: Then we need a resolution in our call in two weeks'
time, so that the Decision review period is
… over by the time of the following meeting.
IMSC-HRM Tests
Nigel: I raised two issues, and we have pull requests for both.
… The first was #3 where there were p elements with no
associated region.
… I wondered if it was deliberate.
Pierre: Remembering my process for creating them, I'm pretty
sure it was a copy/paste error.
Nigel: Ok, we have an open pull request for that.
… Then the other one was about having the tests be conformant
with more sub-profiles.
… In particular EBU-TT-D, but 2 of the tests also aren't
conformant IMSC 1.0.1 because they
… use prohibited features e.g. textShadow that are permitted in
v1.1.
… I opened a pull request for that also.
[13]Make tests EBU-TT-D compatible w3c/imsc-hrm-tests#6
[13] https://github.com/w3c/imsc-hrm-tests/pull/6
Nigel: I take it from your approval that you agree they don't
make any substantive difference, these changes.
Pierre: Correct.
Nigel: For tests, I think we can probably go ahead and merge
soon.
… Anyone want extra review time for these?
group: no
Nigel: For the test documents that cannot be IMSC 1.0.1 or
EBU-TT-D conformant I just added an XML comment
… to explain that.
Pierre: I'm happy to merge both of these.
… I'd prefer to do them both in order
Nigel: That's fine I'll rebase the second one after the first
is merged, after the call.
Pierre: OK thanks.
DAPT
Nigel: We got a positive HR review from TAG.
… We are still waiting for APA and Security
… APA has almost completed, they are just doing a CfC to check
if they want to ask for an
… increased emphasis on client side players offering audio
description mixing options to users,
… for example to change the levels.
… I've told them we might do that anyway, because it's a good
idea.
Atsushi: For Security review, usually they almost never provide
review comments, so usually
… we can pass without comment from security to go to CR.
Nigel: Oh, that's annoying, ok. At least a "we have finished"
response would be nice.
Atsushi: I plan to open a CR transition request issue when
Editors and Chairs are satisfied to go to CR
… and we have positive HR and WR responses.
Nigel: Good, yes, thank you, that would be great.
Atsushi: Don't wait for the Security review in order to
proceed.
Nigel: Ok, we won't, thank you.
… There's one pull request to look at, which I hope to merge
today.
[14]Add inline Registry Section w3c/dapt#196
[14] https://github.com/w3c/dapt/pull/196
Nigel: Thank you for the review comments.
… The one late change is in the requesting a change section,
thanks to Atsushi's comments.
… [shows commit ef4ac7e8293fb83891e5717ed7599cea734c8515 on
zoom]
… I'm getting "that's an improvement nods" from Matt and Gary.
Atsushi?
Atsushi: I believe that's a great change, makes things clearer.
Nigel: In that case I will go ahead and merge this following
enough review time, unless anyone
… wants to ask for more time now.
group: no request for more time.
Nigel: That's all on DAPT for today.
Meeting Close
Nigel: Thanks all, we've completed our agenda for today. See
you in 2 weeks. [adjourns meeting]
Minutes manually created (not a transcript), formatted by
[15]scribe.perl version 221 (Fri Jul 21 14:01:30 2023 UTC).
[15] https://w3c.github.io/scribe2/scribedoc.html
Received on Thursday, 23 November 2023 16:58:13 UTC