- From: Nigel Megitt <nigel.megitt@bbc.co.uk>
- Date: Thu, 21 May 2020 16:32:18 +0000
- To: "public-tt@w3.org" <public-tt@w3.org>
- Message-ID: <C224DB40-79AE-46F3-BBB8-F34250C72663@bbc.co.uk>
Thanks all for attending today's TTWG meeting. Minutes can be found in HTML format at https://www.w3.org/2020/05/21-tt-minutes.html
In text format:
[1]W3C
[1] https://www.w3.org/
Timed Text Working Group Teleconference
21 May 2020
[2]Previous meeting. [3]Agenda. [4]IRC log.
[2] https://www.w3.org/2020/05/14-tt-minutes.html
[3] https://www.w3.org/2020/05/21-tt-irc
[4] https://www.w3.org/2020/05/21-tt-irc
Attendees
Present
Atsushi, Cyril, Gary, Nigel, Pierre
Regrets
Andreas
Chair
Gary, Nigel
Scribe
nigel
Contents
1. [5]This meeting
2. [6]IMSC 1.2 Introduction
3. [7]IMSC 1.2 PR
4. [8]TTML Profile Registry - addition of RTP profile short
code
5. [9]TTML2 2nd Edition Implementation Report
6. [10]IMSC ARIB issues
7. [11][WR/ARIB] Character Sets w3c/imsc#544
8. [12][WR/ARIB] (meta) Compatibility with ARIB-TTML
w3c/imsc#545
9. [13]IMSC 1.2 PR (continued)
10. [14]Meeting close
11. [15]Summary of resolutions
Meeting minutes
This meeting
Nigel: Agenda today: IMSC 1.2 intro, ARIB issues, TTML2 IR,
TTML Profile Registry.
… AOB? Or points to make sure we get to?
group: [no other business]
Nigel: I propose we cover the ARIB topics last if that's ok
Pierre: I think the priority should be to close all the issues
on IMSC. I'm not really
… ready to talk about the ARIB topics but we can go over them
if we have time.
IMSC 1.2 Introduction
github: [16]https://github.com/w3c/imsc/pull/552
[16] https://github.com/w3c/imsc/pull/552
Nigel: It feels like what's there now is probably good enough,
though I think the main
… remaining comments are from me.
Pierre: I want to make sure that Atsushi's comment got
resolved.
Nigel: Atushi's comment was about the Unicode text wording.
Atsushi: I assume this part wants to mention that this
specification should mention that
… Unicode code points should be used in the encoding but not
anything else.
Pierre: To answer that, Unicode is being used maybe not very
formally here. To the casual
… reader Unicode text means something.
Atsushi: I think I should point to some reference here but
sorry I haven't. I'm curious about
… using the word "encoding" here.
… The actual definition is "code point" in Unicode.
Nigel: Is there something misleading about the current wording
"encoded according to the Unicode standard"?
Atsushi: 3 encodings are defined. Encoding is a transformation
from code point identifier to byte stream.
… PUA has no meaning in encoding, it's within a code point of
Unicode.
Nigel: PUA is not mentioned, it's something that is understood
by experts.
<atsushi> > The Unicode Standard defines codes for characters
used in all the major languages written today.
Atsushi: [proposes to say that the document consists of Unicode
code points]
Pierre: That's fine by me
Nigel: Is PUA included in that set?
Atsushi: Included.
… PUA is defined by each party, not standardised with a match
between character and code point.
Pierre: I would be really happy to see the exact proposal on
the ticket, because that
… would also allow @vlevantovski to comment. Could I ask you to
make a proposal in
… the pull request for the exact text? That would be great.
Atsushi: Let me do that now.
Nigel: While Atsushi is doing that, I think it's safe to
mention that my comments that
… are still outstanding (thank you for addressing the others),
are all about adding an
… example. I think what we have already is good enough, and a
clear improvement,
… and crucially, satisfies the APA WG issue, so the best thing
seems to me to be to
… move addition of an example to a new issue, and I should try
to prepare a pull request
<atsushi> timed text is expressed exclusively using code for
characters defined in [[[Unicode]]] (just put into github PR)
Nigel: for that separately. It would be great to do it before
IMSC 1.2 PR, but not essential.
… In other words, it could go to a next version.
Pierre: Atsushi's change is fine with me.
<atsushi> [17]https://glyphwiki.org/wiki/u3110
[17] https://glyphwiki.org/wiki/u3110
Nigel: I might have used "character codes"
Pierre: Or "code points"
Atsushi: This U+3110 is a code point defined by Unicode.
… 3110 is the code point, and this will be transformed into
several formed, like in UTF
… it will be 3 bytes.
Pierre: Understood. How about my proposal "using code points
defined in Unicode"
Atsushi: Should be fine also.
Pierre: I will make the change now.
… I just want to point out that because the only representation
is UTF-8 it is true
… that the only representation is Unicode, right, but you're
saying that is too specific?
… In other words it is not wrong to say it is encoded according
to Unicode.
Atsushi: I actually wondered if people would think other
encodings would be valid
… like UTF-16, which is a Unicode encoding.
Pierre: Right, and it's forbidden in IMSC.
Atsushi: I just wanted to be clear about that.
Pierre: [makes the change]
… Nigel, you will resolve your review comment and open a new
issue?
Nigel: Yes.
Pierre: Then we can close this after our usual 2 week period.
Nigel: Yes.
… Any other comments on the introduction text before we move
on?
SUMMARY: @nigelmegitt to open new issue for example, normal PR
review to continue.
IMSC 1.2 PR
Pierre: What's our target for seeking transition to PR?
<atsushi> [18]https://www.w3.org/AudioVideo/TT/wip/
TTWG-2019-spec-timeline.html
[18] https://www.w3.org/AudioVideo/TT/wip/TTWG-2019-spec-timeline.html
Nigel: I think we deal with all the review comments and then
CfC transition to PR.
… I'm not aware of any other dependencies.
Pierre: So we could start the CfC based on pull #552?
… I would encourage that.
Atsushi: The Implementation Report is finished I think?
Nigel: Yes
Atsushi: Then it might be okay to begin the call for consensus
within this working group.
… If there are no other remaining issues for this spec.
Nigel: The only way this is going to work is if we decide to
make no changes based on
… the ARIB comments.
Pierre: We can wait until the end of the meeting.
TTML Profile Registry - addition of RTP profile short code
github: [19]https://github.com/w3c/tt-profile-registry/issues/
72
[19] https://github.com/w3c/tt-profile-registry/issues/72
Nigel: I prepared a pull request for this, #73
… Just wanted to check if anyone has any comments or might have
an objection.
… Otherwise we can do the 2 week review as normal.
… If anyone can have a look at the pull request and approve it,
that would be very
… helpful for merging.
… Any objections?
group: no objections
Resolution: Proceed with this pull request under normal 2 week
TTWG procedure
TTML2 2nd Edition Implementation Report
Cyril: I did a verification.
… I compared the number of files in the repo vs the number in
the JSON file.
… Just focusing on the 2nd Ed the numbers are the same so we
should be good.
… They're different for non-2nd Ed but I don't think it's a
concern.
… For more details, we have 57 files for validation and 14 for
presentation.
… The total is 71 and that's the number in the IR. I will check
that they're the right ones too.
… I added the 4th section for invalid presentation tests.
Nigel: Great, this is good enough for me to add the
implementation that I'm aware of.
… I will do that.
… We should assume this is now the correct set and obviously if
anyone finds any errors
… please fix them.
… Now we need to register implementations and verify that we
have enough to pass
… CR exit criteria.
… Any other actions/comments/questions on this?
group: [no]
IMSC ARIB issues
Nigel: I went through these and I think Pierre did too.
… The only one that might result in a change I think is the
character sets for "ja" language code.
Pierre: Yes, that's an easy one. Whether or not we do it now,
at this 11th hour, I'd rather
… not but it's an easy and a good addition.
[WR/ARIB] Character Sets w3c/imsc#544
github: [20]https://github.com/w3c/imsc/issues/544
[20] https://github.com/w3c/imsc/issues/544
Nigel: I think the first question to ask is if this is
normative/substantive.
Pierre: We should try to avoid making substantial changes this
far into the process, but
… we could formally because it is only informative.
Pierre: That section, regardless of the normative language
around it, is meant to inform
<cyril> " this section defines common character sets that
authors are encouraged to use."
Pierre: implementations. You could conclude that it affects
implementations.
Cyril: "encouraged to use"
Pierre: And the W3C Process definition.
[21]Process 6.2.5 Classes of Changes
[21] https://www.w3.org/2019/Process-20190301/#correction-classes
Pierre: Section 8.2 says a document "SHOULD be authored using
characters from" the common character sets.
Cyril: There's a relationship between the reference fonts and
the common character sets, right?
Pierre: Also
… I think we should deal with these ARIB comments in the next
version of IMSC otherwise
… we may make mistakes.
Nigel: Back to Cyril's point, there does seem to be a
substantive relationship between
… reference fonts and the common character sets and the §9.3
text on rendering rules.
… So it looks as though changing the common characters changes
the code points in the
… reference fonts and therefore the rendering rules.
… (sorry that 9.3 is assuming the introduction is added,
otherwise it's 8.3)
Nigel: My conclusion is we cannot make this change now but
should add it to vNext
… with appropriate care about reference fonts, and checking
that the code points are
… indeed all available.
… Any other points to add before I summarise?
Pierre: I think that the idea of converging ARIB-TTML and IMSC
is really a great goal,
… and we should take the time to do it in collaboration with
ARIB. I see that as a pretty
… extensive but worthwhile effort.
SUMMARY: TTWG would like to adopt this change in a future
version of IMSC.
[WR/ARIB] (meta) Compatibility with ARIB-TTML w3c/imsc#545
github: [22]https://github.com/w3c/imsc/issues/545
[22] https://github.com/w3c/imsc/issues/545
Nigel: These 5 issues mainly don't need change right now but
they do need discussion.
… I don't think we can make any changes to IMSC 1.2 for them
but we can consider
… and discuss changes for future versions, if needed.
… It's also worth noting that some of the ARIB-TTML features
look like they are coincident
… with either existing IMSC 1.2 features or TTML2 features.
That is why further
… discussion with ARIB would be very useful, to understand the
best path forward for
… all parties.
Pierre: I think the devil is in the detail and we have to study
each issue. I agree with Nigel's summary.
Nigel: Any other comments?
Cyril: To make sure I understand, you say we will defer to the
next version but when
… practically will we start talking about them?
Pierre: I suggest agenda+ these for the next meeting, to start
discussing it.
Nigel: That works for me.
Cyril: Fine for me also. I would suggest that at least for the
PUA/Gaiji discussion we invite
<atsushi> +1 for agree with Nigel's summary.
Cyril: Vladimir.
Pierre: Yes. At least on that one it is an intersection of many
things - what is possible
… with fonts, what ARIB thinks and what we think! A complex
discussion.
SUMMARY: TTWG raised this for discussion today, and would like
to discuss the sub-issues in further detail in future meetings.
IMSC 1.2 PR (continued)
Nigel: We need a version with a PR SOTD.
… Please could I ask you to prepare one on a separate branch,
Pierre, based on the pull
… request branch where we add an introduction.
Pierre: Yes I will do.
Nigel: Then I will issue the CfC based on that.
Meeting close
Nigel: Thank you everyone, we're out of time and agenda! See
you next week. [adjourns meeting]
Summary of resolutions
1. [23]Proceed with this pull request under normal 2 week TTWG
procedure
Minutes manually created (not a transcript), formatted by
[24]scribe.perl version 117 (Tue Apr 28 12:46:31 2020 UTC).
[24] https://w3c.github.io/scribe2/scribedoc.html
Received on Thursday, 21 May 2020 16:32:36 UTC