Timed Text Working Group Teleconference

<scribe> scribe: Nigel

This Meeting

nigel: We aren't quorate to discuss much today but there are a few topics we can cover for the record.
... AOB?

atai: I'd like to look at the status of the profile registry.

nigel: Okay

Action Items

action-460?

<trackbot> action-460 -- Nigel Megitt to Creat an issue to draft a ttml2 section on customisation. -- due 2016-04-14 -- PENDINGREVIEW

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/actions/460

nigel: I created https://github.com/w3c/ttml2/issues/156

close action-460

<trackbot> Closed action-460.

action-457

<trackbot> action-457 -- Pierre-Anthony Lemieux to Alert Unicode upon publication of IMSC1 -- due 2016-04-15 -- OPEN

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/actions/457

pal: Can we move the date of this to after the 21st?

nigel: will do.
... done - made it the 22nd.

action-456?

<trackbot> action-456 -- Pierre-Anthony Lemieux to Collect use cases and testimonials for potential IMSC1 press release -- due 2016-04-07 -- OPEN

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/actions/456

pal: I've done this - it's in Karen's hands. I have interest from a number of W3C members and other non-member implementors.

close action-456

<trackbot> Closed action-456.

nigel: Thanks for collecting those!

Charter

plh: I've got all the internal green lines - it's now in the hands of the comm team. The
... AC review should be under way tomorrow, unless there are questions on the draft.
... I made a few changes to the draft:

<plh> https://github.com/w3c/charter-drafts/pull/50

<plh> https://github.com/w3c/charter-drafts/pull/50/commits/15ced1d6e3a42e03fa284a8a306c228b4c3032d2

<plh> and should detail any known security or privacy implications for implementers, Web authors, and end users

plh: The main change is the one regarding security requirements. They add the words:
... ", and should detail any known security or privacy implications for implementers, Web authors, and end users"

pal: For TTML there is a requirements doc.

<plh> https://www.w3.org/TR/security-privacy-questionnaire/

plh: This is from the TAG. Writing a security or privacy section should address the questions in there.

pal: That's a huge change!

plh: What do you mean?

pal: We didn't have to do this before, now we have 17 extra requirements.

plh: They are questions. I would have thought that for TTML, which doesn't use scripting,
... a lot of them will not apply. For example we're not creating an API to track user language preferences.
... So most of them will not apply. If you are defining an API then you will have to answer those questions.
... I would not worry about this. Obviously IMSC 1 will be a Rec before this applies.
... The questionnaire was developed for people to consider at least the minimum questions.

nigel: I would welcome this. The questionnaire doesn't get a link from the draft charter,
... which could help to scope/describe what is meant, but they're sensible questions and
... useful to reference.

plh: I'm happy to link to it in the charter as an example.

pal: I'd like that to be present.

plh: The other pull request changes were not as impactful. W3 legal made me slightly
... tweak the wording on patent commitments. It's slightly more open wording now.
... There were also minor grammatical changes.
... I rejected the proposal to put security into the success criteria and put them in the scope instead
... next to accessibility.

nigel: Makes sense.
... So what's the timeline now?

plh: If it goes tomorrow then I will propose a deadline for AC review of Friday May 13.
... We're looking at mid-May for end of charter review. In the meantime you can keep
... working.

nigel: Thank you.

pal: Great news.

plh: I'll add the questionnaire link to the charter.

IMSC

plh: The bad news here is that Karen is in Montreal - although I said last week that we
... were targeting April 19 I learned today that it's been moved to April 21.
... I've told them that's the latest that we'll publish the Rec.

pal: I've provided Karen with testimonials and technical bullets, as well as a picture for the PR.

plh: Thanks for doing all that. They're still committed to doing the press release.
... The document is ready by the way. There's a little internal work to get the final approval for publication.

nigel: The only final approval is from the Director right?

plh: Yes, the WBS closed without issue.
... I created the Rec version by copying the PR and updating the headers and status.
... Pierre has given it the once over and said it's fine. I've removed the link to the
... implementation report because I don't like to put it there - it has a tendency to go stale.

pal: I like that idea. Thierry has archived the IR, which will be linked from the PR version.
... He created a new live page for implementation and test vectors, which I have to update.

nigel: Are you suggesting linking to that from the Rec?

pal: No, I don't recommend doing that.

plh: The document itself is ready to go.

nigel: Is it too soon to start thinking about the next steps for IMSC?

pal: No, as soon as we're done with this my plan is to start listing features that we want
... in IMSC 2 and also tackle the question of how to handle those features in IMSC 1 that
... are also now in TTML2 but are expressed with a different syntax.
... It would be a disservice to the industry to obligate folk to do things completely differently.
... Either we deprecate but still support the old way and introduce a new way or we
... do something else [that I don't know yet].

plh: I had a separate but related question related to the IMSC 1 publication and the
... meaning of Latest Published Version.

<plh> http://www.w3.org/TR/ttml-imsc1/

plh: When I was preparing the Rec I realised that the latest published version points to the
... above link. Be aware that we have more and more WGs dealing with versioning questions.
... For example in CSS what they do nowadays is have 2 latest published links, one for

<plh> http://www.w3.org/TR/ttml-imsc1/

<plh> http://www.w3.org/TR/ttml-imsc/

plh: the latest version of IMSC and one for the latest version of this version.
... The question is do we want a "latest IMSC" link as well as a "latest IMSC 1"?
... I don't know if there's a use case for that.

nigel: I think there is a use case. What do you think Pierre?

pal: I think that's a really good question but I don't have an answer now. There's a
... difference from a device and an authorial perspective. From an authorial perspective
... if IMSC.next is a superset of IMSC 1 then that's great. From an implementor's
... perspective it may introduce a bunch of new features. I haven't thought about it yet.

plh: Can you think about it and tell me if you think there's a need for it, and with
... apologies for bringing this up so late, it's going to be much harder to do after we've
... published the Rec. If we can decide by Tuesday then I just have to ask the Director
... if it's okay to add the link.

<plh> Latest published version: http://www.w3.org/TR/ttml-imsc/

<plh> Latest published version 1: http://www.w3.org/TR/ttml-imsc1/

plh: We'd do something like the above - one version-less, and one with a level number,
... and both would appear at the top of the specification.

pal: Should we use the term "level"?

plh: No, you don't have to call it that - CSS does it that way, but they may be the only one to do that.
... Other groups don't. In WebPerf we moved to this system with a slight difference: we
... don't wait for documents to get to CR, we just publish a link to the latest work in progress.

pal: One question that comes to mind is: say IMSC 2 brings new features that need more
... thinking authorially and implementation-wise. How do you warn or explain to people
... that there are significant implications to clicking that link? Is the reader expected to
... compare the two specs and see the differences?

plh: The only way I can think of is to have a section in the spec that lists the differences.
... In CSS they don't mind it because it's normal to publish like this.

pal: And what if we do an IMSC 1.1?

plh: In that case I would say the latest IMSC 1 link would point to 1.1 when we consider
... it to be stable enough. The Maths spec have adopted an even more complex scheme.

<plh> https://www.w3.org/TR/MathML/

nigel: I think we should do this.

plh: The MathML spec as above says "Latest MathML 3 version" and "Latest MathML Recommendation"

pal: What about an intermediary page?

plh: We can add that as a top header of the specification as well.

pal: Imagine we have IMSC 1.1 and IMSC 2, where 1.1 has fewer changes than 2. Both
... are relevant for different use cases. Some people may keep with 1.1 and others may
... wait for and then adopt 2. We're in danger of stopping people from noticing 1.1.

plh: Unless you click on the "Latest IMSC 1 version" link?
... That would always point to the latest version of IMSC 1, while the "Latest IMSC version"
... link would point to 2 when it exists or 1.1 before, for example.

nigel: I think we shouldn't restrict the "Latest IMSC" link to a Recommendation though.

<plh> Latest IMSC version: http://www.w3.org/TR/imsc/

<plh> Latest IMSC 1 version: http://www.w3.org/TR/imsc/

pal: So imagine we have IMSC 1 2nd edition, IMSC 1.1 and IMSC 2. If you click on the
... "Latest IMSC 1" what should that do - go to IMSC 1 2nd edition or IMSC 1.1?

nigel: So an IMSC 1 2nd Ed would incorporate errata?

pal: Exactly. That's the most important link to have first. That's what I think should be
... "Latest Version".
... My question is from IMSC 1 1st edition would the next level/version link point to IMSC 1
... or IMSC 2?
... I'm thinking of having an intermediate "roadmap" page with the roadmap, telling you
... all the latest published and in progress IMSC documents.

nigel: That could be a wiki page?

<plh> https://www.w3.org/TR/soap/

pal: Or on /TR as a simple roadmap document.

plh: SOAP is the instance I know of.

pal: It would have no normative content.
... I can draft that quickly with Respec. Today there would be only one entry.

nigel: I like this idea - it's a good future proof way to deal with it.

pal: Okay, I'll post it to the list.

Profiles registry

atai: I just wanted to point out that we need to bring this to a state where we can
... publish it as soon as possible - people are asking for the information.

plh: Is there anything I can do to help on this?

<plh> https://github.com/w3c/tt-profile-registry/pull/3

atai: I think there's a minimal editorial thing - to separate a big table into two smaller ones.

plh: So we're waiting to merge before resolving?

atai: The only thing outstanding is to resolve conflicts, merge, then split the table.

plh: Can we merge the PR?

nigel: Yes.

plh: Shall I resolve the conflicts and merge?

group: Yes.

plh: Then we can raise splitting the table as a separate issue.

nigel: Yes, sounds good.

atai: Then the other question is if there's a reference from IMSC to the Profile Registry?

plh: It's not published yet. Do we have agreement from the group to publish?

nigel: Not yet.

pal: We don't have to have a link.

plh: We can always link to it from the TR/imsc page once we have one.

pal: True.

atai: Possibly we should also check with Mike Dolan because I think he has some more
... information to add to the table.

plh: I'll do the merge and then you can guys can proceed.

nigel: Thanks!

plh: If I cannot resolve the merge I'll reach out to you.

atai: I think it's only additional information, just has to be merged in layout and style.
... There are no substantive conflicts.

plh: ok

nigel: Thanks - that was an interesting meeting in unexpected ways. We're out of time now though. Meet same time next week. [Adjourns meeting]