Meeting minutes
This meeting
Nigel: Today, we have IMSC 1.2 FPWD next steps, two issues/pull requests to discuss.
… Plus TTML2 2nd Edition CR publication.
… Is there any other business?
Pierre: Can we discuss WebVMT?
Nigel: Sure, in AOB
Potential semantic conflict between ttp:profile and ttp:contentProfiles. imsc#506
github: https://github.com/w3c/imsc/issues/506
Nigel: Just checking in, since when we discussed this last the Editor was not present.
… Did the consensus last time make sense, Pierre? Any other questions.
Pierre: Seems fine with me.
… But why in IMSC 1.2 instead of TTML2? We have to remove the may, but if folks think a note here would be
… useful that's fine. I'll change the wording a little bit but keep the intent.
SUMMARY: Editor to prepare Pull Request
Glenn: I would accept this under TTML2 in a future edition.
Nigel: But I wanted it sooner so suggested put it in IMSC 1.2 now and TTML2 later.
Glenn: There's some similar existing language in TTML already.
Pierre: Okay, got it, thanks.
Permit CSS font matching and font-face mapping imsc#517
github: https://github.com/w3c/imsc/pull/517
Pierre: I have one last editorial comment.
Nigel: Also Glenn raised more substantive comments, let's do those first.
Glenn: 1. On the font-face issue, we didn't actually discuss the font-face matter at the last meeting as
… far as I recall. It doesn't make any sense at all as far as I am aware to even mention font-face because it
… has no relevance to us because there is no way to refer to font-face in our context whatsoever.
… We don't have any way to put font-face in our content, syntactically, so any use of it would have to be
… completely out of the scope of TTML or IMSC unless I'm missing something.
Nigel: We did discuss font-face and the intent here is to say that using the font-face algorithm is compatible with
… the TTML font matching semantics which are implementation defined.
Glenn: The font-face rule describes a syntax. I thought from last week we were only talking about CSS Fonts §5.1
… The first paragraph, by itself, is perfectly fine and is exactly what I thought we agreed last week, and I think it is
… sufficient. Anything else is extraneous. The implementation can decide all the details.
… It's going overboard by putting extra hints on what the implementation might do.
… You're saying "implementor, here are other things you might think about", but I don't see that as a necessary hint.
Nigel: I thought it is unclear what we mean unless we explain the mapping.
Glenn: I think implementers will find it obvious so the table is not necessary.
Pierre: Just for my own clarification, which paragraph are you referring to?
Glenn: I'm suggesting deleting the paragraph about the font-face rule, and just leaving the first new one
… that says the font matching algorithm in place.
Pierre: That would address my comment as well.
… My question was why have those two paragraphs disconnected.
… From an implementer it was weird that I could use one without the other.
… If the conclusion is the first one is sufficient then it would be good to keep just that one.
Glenn: In the note, the first sentence is already written into the text of the spec, so it isn't needed.
Nigel: You think the second sentence works without the first one?
Glenn: OK if we just remove the last sentence we could keep this.
… For locality of reference, put it after the new paragraph about the font matching algorithm.
Nigel: That works for me.
Glenn: Thank you for accommodating me.
Nigel: Not at all, feedback very welcome.
Pierre: That change would address my comments too.
Nigel: Okay, I will implement that in the pull request.
SUMMARY: @nigelmegitt to amend pull request taking into account feedback from today's discussion.
TTML2 2nd Edition CR Publication
<atsushi> :yay:
Nigel: We published on the 28th as planned. Thank you to Glenn and Atsushi for making it happen.
Nigel: There's a question, whether to point people looking at 1st Ed Rec at this new 2nd Ed CR.
Glenn: Tricky question. I can't recall what we did with TTML1
Nigel: I don't think this facility existed then.
Glenn: It may have for 3rd Ed, which was late 2018.
… I'm of 2 minds. I wouldn't want it to warn to say the Rec is obsolete.
… If we did a warning I would want it to say something like "A revision is in process, look here if you want to see it"
Nigel: Is this something where we can control the text?
Atsushi: Usually when a WG updates a Rec a pop-up will be raised to note it is an old version so please look at the new version.
<atsushi> https://www.w3.org/TR/2018/PR-ttml1-20181004/
Atsushi: This one says "This version is outdated!"
… Do you want this popup in TTML2 1st Ed Rec to point to TTML2 CR?
… I suppose if we consider the TTML1 Rec is still alive, it may be possible to keep TTML1 Rec as the latest version for TTML1.
Pierre: Yes, please don't touch TTML1.
Nigel: Agreed, this is only about TTML2
Glenn: The question is can we supply the text for the warning?
… In this case the TTML2 1st Ed is still active. It is not outdated yet.
… But it is in process of revision so we could provide a warning saying that it is in the process of revision and linking to the revision.
… We would not want to say it is outdated. Please could you check to see from the publishing people if we can provide
… the text of that? If they only have canned text then we may need to leave it the same.
Nigel: It might depend on what the canned text options are.
Glenn: We'd want to provide text.
Atsushi: That pop-up is provided by a central database so I don't think we can customise the note.
Glenn: Can you ask the publishing people, maybe their database has a way to include a custom message?
<atsushi> <script src="https://www.w3.org/scripts/TR/2016/fixup.js"></script>
Atsushi: When I checked before we usually just put one line of inclusion of script.
Nigel: That script goes to https://www.w3.org/TR/tr-outdated-spec and the response depends on the referring document.
… So Atsushi could you ask what flexibility we have in directing the server to send our own text for that, in response
… to the referral?
Atsushi: Yes, I can ask
Nigel: I think our conclusion is that we would like some text, but only if we can check it in advance.
Atsushi: It seems all of the contents are written by JSON data from the URL - let me go back to them with this request.
Nigel: Thank you
… In terms of this CR, I think the next steps are to work out which tests should go into the implementation report
… and prepare that.
Glenn: I counted 30 substantive PRs going into this version (from memory) and 15 (?) of them were marked as
… untestable, or as having test, I think 11 have tests and 4 are untestable.
… TTV is one implementation of those 11, so that leaves 15 others that were merged prior
… to our arrangement to create tests before merging, so we need to go back and create tests for the other 15.
… Then we need to find another implementation for the remaining testable ones.
… Of the 15 that are left some of them might end up being untestable. We haven't gone through them yet.
… We do need another implementation. Right now we've written in the SOTD that we won't go to PR before
… March 17 so I've already created a milestone in GitHub and marked its pending completion date as that date.
… It can be changed.
… We have one issue tentatively right now which is the privacy review response.
… There's also a request from internationalisation to change an example to use isolate instead of embed, which
… we can probably accommodate prior to Proposed Rec.
… So we need to identify a possible 2nd implementation.
… So we need to look into that. Previously BBC, Netflix and I believe Pierre, you guys, had done some implementation
… work so perhaps we can get another implementation.
… I need to look at the possibility of TTV implementing the other testable tests.
Nigel: Yes, good summary and reminder points about our next steps.
<atsushi> https://himorin.github.io/ttwg/TTWG-2019-spec-timeline.html
Atsushi: I need to update the above for TTML2.
Glenn: I used Philippe's tool for generating 17th March. It said Rec publishing date in April if all goes to plan,
… but Atsushi should maybe generate a timeline incorporating current practice and all the steps necessary.
Atsushi: Also let me propose one thing on this.
… For now, that timeline is put to w3.org as a single HTML5 but is it possible to have one directory to be redirected
… to github.io so we can host these files in there to be published at w3.org space, like our current index.html for WG
Nigel: Yes of course, it would make sense to put this in the TTWG repo.
Atsushi: Currently only the home page redirects, but we could add a subdirectory used for us to easily edit via PR.
Nigel: That's a really good idea, yes please.
<atsushi> https://www.w3.org/immersive-web/
Atsushi: I will send an email requesting a directory. I'm not sure what directory to use.
<atsushi> https://github.com/immersive-web/homepage
Nigel: I'm not sure either, maybe follow another group, or otherwise call it something like 'wip'.
… Can we follow this up off-line with Atsushi, Gary and Nigel?
Atsushi: Yes
AOB - WebVMT
Pierre: WebVMT is a fork of WebVTT.
… I'm interested in keeping all the Timed Text discussions in the same group in W3C.
… Do you have any insights, maybe Gary?
Gary: Rob Smith is the one heading this up.
… It sounds like it started off with them trying to use WebVTT for metadata.
Nigel: Specifically geographic metadata.
Gary: Yes, for tying map data to a video, like showing locations moving on a map.
… It seems like it's mostly an extension rather than a full fork, but there are definitely some changes.
… They potentially should be moved back to WebVTT, I don't know.
Nigel: Last I heard on this, it is WebVTT minus some features that Rob thought were unnecessary for the use case.
Pierre: It is confusing to have a proliferation of groups working on timed text.
… It would be good to ask them why not join TTWG?
Gary: He's also working on the datacue proposal.
Pierre: It would be good if these were all under the same roof.
Gary: That makes sense. There was a question from Silvia on the M&E mailing list asking why different from WebVTT.
Pierre: I don't want to prevent that work, and I don't want to presume of the outcome, but also looking at what is
… happening with bullet chatting, it's great to have loads of initiatives but hard to keep track of them.
Nigel: Sounds like W3C all over!
Pierre: Well indeed, it is a recurring comment.
Gary: It is also possible that forking is the best decision.
Pierre: Exactly, I don't want to make an assumption but it would be good to do it in the same group.
Gary: Yes, my only issue is that it is harder to trickle back good features into WebVTT.
Pierre: Yes, and with bullet chatting, people get excited about WebVTT because it is implemented in browsers,
… but if they extend it and do it without browser implementation then they're back where they started.
Gary: Exactly.
… They're back to implementing themselves.
Pierre: Exactly, I always thought that was a strange argument.
Nigel: Okay, that's noted, one for Gary and I to think about. I'm not sure what actions to take.
Meeting close.
Nigel: Thanks everyone. We're a little over time, so let's adjourn. [adjourns meeting]