Timed Text Working Group Teleconference

<scribe> scribe: Nigel

This meeting

Nigel: I think we should mention the CLDR response from Unicode at least.
... Anything else?

group: Nothing specific

F2F Meeting

Pierre: I suggest generating a detailed agenda for example pointing at the issues that we
... need to close.

Nigel: That's true, we need to close issues!

<scribe> ACTION: Nigel Generate detailed (timed) agenda for F2F to allow people to prepare in advance [recorded in http://www.w3.org/2016/12/15-tt-minutes.html#action01]

<trackbot> Created ACTION-491 - Generate detailed (timed) agenda for f2f to allow people to prepare in advance [on Nigel Megitt - due 2016-12-22].

Pierre: I'm happy to take a look at that for IMSC.

Andreas: For IMSC 1.1 and also TTML1 we need to spend time on the line gap issue.

Nigel: Agreed.

Andreas: Also Safe Crop Area is something we have discussed but do not have anything
... more concrete to discuss for inclusion in IMSC 1.1, in London.

Nigel: +1

Pierre: We cannot finalise the agenda until Jan 8 when we have all the liaisons in hopefully.

Nigel: True.
... Also we want to be able to review an updated draft of TTML2 with a view to publishing
... as a CR, so the outcome of that review process at the F2F would be a set of small(ish) actions
... or a resolution to publish. So that we can have enough time to prepare for that we need
... the updated draft from Glenn by 5th Jan at the absolute latest.

Pierre: I think that may not even be possible even if we have a draft today, but we should at
... least be able to close a good set of issues.

Andreas: The product with the most urgent deadline will be the update to IMSC 1 so that
... should be taken into account when preparing our agenda.

Nigel: So I have an action, thank you Pierre for offering to review/author the IMSC parts of the agenda,
... and I will also ping it to Glenn.

Pierre: What I will do is finalise the IMSC parts based on the liaisons I have received by Jan 8,
... as soon as possible after Jan 8.

Nigel: Thank you for that.

IMSC drifting into general TTML

Nigel: Thank you Pierre for responding to the CLDR ticket. We got a response:

-> https://lists.w3.org/Archives/Public/public-tt/2016Dec/0018.html

Nigel: Shervin is suggesting that a closer inspection of each set is necessary.

Pierre: The folk who put the sets together did not want to miss anything so they erred on the
... side of being too broad.
... I can get back to them on that because for instance the Latin Extended A set is literally
... the Unicode code block that includes u+017F so instead of trying to finesse every
... character set for every country in central Europe etc. they just said for all Latin European
... countries we are going to recommend the use of any character within the Unicode Latin Extended A code block.
... That's inclusive at low cost. I think this is a response to Shervin, to explain this.

Nigel: Do you want to write that response?

Pierre: Yes, I can reply to Shervin.
... If their recommendation is to target each language independently, then that would be
... useful feedback for us.

Nigel: Thank you!
... Anything else on IMSC?

Pierre: Something that also applies to TTML is when features that are required for subtitles
... and captions are not supported by CSS. If features are required they should be in the web
... platform; if not they should not be in IMSC (and possibly TTML). I'm not sure how we do that.

Nigel: I agree with this, and observe that if it's hard to get implementations of everything in
... specifications then the last thing you want is to add new features that don't get implemented,
... since they could hold CRs up for a long time. On the other hand no feature gets implemented
... unless it is requested/proposed first. So the first thing we should do is trawl the CSS
... issue trackers to make sure that they are recorded; getting them implemented is something
... that our members cannot necessarily do, so we do need to get buy-in from the CSS
... implementers, otherwise we will never get those features in CSS recommendations.

Pierre: We should tell CSS, maybe TAG, AC etc that we have these requirements for the web platform.

Andreas: We also have influence as members on other groups' charters.

Nigel: Also (thank you for the reminder) we have received a request from Bert Bos to review
... CSS 2.2 CR.

-> http://www.w3.org/TR/CSS22/changes.html

Nigel: Bert is asking for implementation feedback on the CR, but also possibly for wide
... review comments. Do you want to request any time for reviewing CSS 2.2 prior to
... feeding back to CSS WG?
... (that's a question for the group not just those present on this call)

Pierre: I would support spending time in the F2F analysing TTML2 to work out which features
... need to be in CSS.

Nigel: One thing we could do is to specify in our CR exit criteria that at least one implementation
... must be done using a mix of HTML, CSS and EcmaScript/JavaScript, and attempt to work
... out which TTML2 features would need a polyfill or something similar to make that happen.
... Then we could choose to mark those features as "at risk" so that, should they not satisfy
... the exit criteria, then we can drop them when we move to Proposed Recommendation
... without incurring a process delay.

Pierre: I like that idea.

Nigel: If we were to set a rule like "don't publish TTML2 unless all features are implementable in CSS"
... then we would effectively be waiting forever, or it would seem that way, since if those
... features are only ever exercised by translation from TTML2 to HTML+CSS then I think
... most implementers of browsers would deprioritise implementation of those features
... almost immediately on the basis of a low expected volume of usage.
... Ok we're out of time for today, thanks all. [Adjourns meeting]