W3C

Timed Text Working Group Teleconference

12 May 2016

See also: IRC log

Attendees

Present
Nigel, Andreas, Glenn, Pierre, Thierry
Regrets
Chair
Nigel
Scribe
Nigel

Contents


<scribe> scribe: Nigel

This Meeting

nigel: Today we have Charter, TPAC 2016, IMSC, TTML, Profiles Registry, TTML and WebVTT Mapping Document
... Any other business?

Charter

tmichel: Charter Review ends tomorrow, Friday.
... We need 5% of the AC members to respond by new rules.

nigel: I didn't see any comms about this survey before your reminder.

tmichel: The comm team only sent it to AC reps.

group: Unaware of survey minimal response requirement until now.

nigel: Some members may be able to see results at https://www.w3.org/2002/09/wbs/33280/tt2016/results

TPAC 2016

nigel: Registration is now open for TPAC: https://www.w3.org/2016/09/TPAC/
... It's from 19th to 23rd September this year.

action-464?

<trackbot> action-464 -- Nigel Megitt to Ask ttwg who would really prefer to be able to attend web & tv ig as well as ttwg at tpac 2016 -- due 2016-05-05 -- PENDINGREVIEW

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/actions/464

nigel: I did this, and got some responses back.
... Of the responses, some want to stay in TTWG all day on the Monday, others to
... move between both. Overall there's a majority in favour of a joint meeting.
... So I propose to request a max 30 minutes joint meeting, also to go back to the
... organisers to discuss the options. We seem quite inflexible for moving from the Monday and Tuesday.

pal: There may not be 100% crossover of interest in both groups so we could analyse the
... actual agenda of topics before rushing to a conclusion.

nigel: I'm not volunteering to try to meet the fine grained preferences of each individual member!

pal: It's less than ideal.
... Perhaps we could arrange our agenda for the Monday to minimise overlap.

close action-464

<trackbot> Closed action-464.

IMSC

action-461?

<trackbot> action-461 -- Thierry Michel to Add an acknowledgment section to the imsc 1 errata section. -- due 2016-04-28 -- OPEN

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/actions/461

glenn: I think the only action was to change my attribution to plain Skynav without the "inc"

pal: We did that last time, there are no pull requests for this.

https://github.com/w3c/imsc/blob/master/spec/ttml-imsc1-errata.html

tmichel: We're only on a draft on github. When the group has reached agreement for the
... errata then I can move it to the place on /TR that is linked from the Rec.

https://rawgit.com/w3c/imsc/master/spec/ttml-imsc1-errata.html

nigel: That draft looks fine to me.

action-466?

<trackbot> action-466 -- Nigel Megitt to Review notes to see who if anyone should be added to imsc acknowledgements for special thanks. -- due 2016-05-05 -- OPEN

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/actions/466

nigel: From my review I didn't identify any other contributors.

close action-466

<trackbot> Closed action-466.

PROPOSAL: Publish the IMSC Errata document at https://github.com/w3c/imsc/blob/master/spec/ttml-imsc1-errata.html

glenn: Seconded.

tmichel: I will change the date to today's date.

nigel: Should there be a ToC on this document? It looks strange with a blank panel to the left.

glenn: Is this done with respec? If it's plain HTML then it needs the correctly formatted nav section.

tmichel: It's just HTML

glenn: It needs to be updated to meet the new styling.

tmichel: We don't have to do that here though.

glenn: The default stylesheet puts up a blank ToC pane on the left.

nigel: I'm raising an issue for this now.

https://github.com/w3c/imsc/issues/181

RESOLUTION: Publish the IMSC Errata document at https://github.com/w3c/imsc/blob/master/spec/ttml-imsc1-errata.html on /TR when it has been fixed for styling

TTML

action-458?

<trackbot> action-458 -- Glenn Adams to Create issue re: line height calculation for inline -- due 2016-04-07 -- OPEN

<trackbot> http://www.w3.org/AudioVideo/TT/tracker/actions/458

glenn: I think that's been done...
... actually leave open for now, there isn't an issue yet.

nigel: What steps remain to publish the TTML2 WD. We're just waiting on the comms team
... for a date aren't we?

pal: Correct.

glenn: As soon as I have a date I'll create a package and send it over to Thierry for publishing.

tmichel: Might be worth a request from the Chair to the comm team asking for notes

nigel: Ok I will do that after this meeting.

tmichel: Please CC me and plh

nigel: Will do.
... The next topic relating to TTML is use of Pull Requests.
... First, thanks to Glenn for reviewing the pull request mechanism and offering to use it. Given the ability to auto-publish WDs based on the group decision in Sapporo
... and having reviewed all merged PRs, does that change anything?

glenn: I think the onus is on me to provide at least a week notice prior to publishing,
... rather than me just pushing a button without an opportunity to query the group.
... Given the echidna process requires me to upload a tar to a server to do automated processing,
... including validating with the new validator and new pubrules system, which I've
... started doing and found a bug in pubrules, which I think has already been fixed, I need
... a token from the team to assign it to the document. I'm not sure how often the token
... needs to be updated. Probably it doesn't need to be updated.
... My thought is to announce to the group that an update to the WD is needed, and
... to request comments within a fixed period, e.g. a week for any objections.

nigel: That's fine - the end of the e.g. week period should be no earlier than the end of the review period for any changes.
... The issue is that we should have consensus for the changes according to the
... Decision Policy, which allows for 2 weeks.
... We don't really need extra consensus for publishing a new WD beyond consensus
... for the changes.

glenn: On that topic, in the last meeting I said I would look at the PR process, which I did
... and decided that most of it was okay for me to adhere to, but that a lazy consensus
... process would be a good compromise.

pal: I think IMSC and TTML should use the same process.

glenn: Are you proposing to change the IMSC process?

pal: We had a really simple process for IMSC and I don't see why we shouldn't use that
... for TTML. It's more complicated to have two processes. The IMSC process is a 2 week
... review for substantive PRs unless overridden by a consensus agreement in meetings.
... I'm personally not happy with the proposed TTML process, because its possible that
... PRs will be merged and then it's hard to back out.

glenn: The PR process will be used - the only difference is the time for a review.

pal: If you close the PR then it gets hidden - you have to go out of your way to figure
... out what needs reviewing. It's needlessly complicated.

nigel: There's a serious point here - it can be hard for folk to see closed PRs. That's
... why I suggested using labels to mark PRs that need review.

glenn: With lazy consensus we only have to mark non-consensus, not consensus.
... Also email can be used.

pal: There's a bigger issue. Say you create 10 PRs, and then merge them a day later.
... Then someone objects to the first one merged. It's really hard to back out of that
... specific PR. I see problems with transparency with that approach.
... I object to having two different processes. If we want to adopt the same one for IMSC
... then we should talk about that.

glenn: I did include a section on post-merge issues - technical, editorial and principled objections.
... The action I proposed was not to back anything out, which I wish to avoid. Instead I
... suggested that new issues be filed addressing the perceived issue. So there won't be
... any back-outs, but there may be follow-on issues that undo former changes.

pal: I have a real issue with the concept of lazy consensus.

glenn: That's what the group has been using since 2003.

pal: Not with IMSC1.

glenn: True, but probably because folks didn't consciously realise.

atai: I very much appreciated the transparent approach taken for IMSC 1. In general
... I support any approach that encourages transparency and for people to review.
... We are discussing a lot of formal processes in our meetings.
... [leaves]

nigel: Apologies we did not have enough time to discuss the mapping document.
... We don't have consensus now, with concerns about transparency and lazy consensus.

glenn: If this compromise proposal is not acceptable then I will revert to the status quo.

nigel: The status quo is actually what we agreed in Sapporo.

glenn: That was not what I understood.

pal: I think lazy consensus will cause more trouble than the process we agreed in Sapporo.
... Also discussing the process is not an efficient use of time.

nigel: Not having an agreed process would be worse.
... My proposal is to go along with Glenn's proposal and deal with real world issues
... if they arise.

pal: Having two processes is a blocker for me. Especially because there are dependencies
... between the two documents.

glenn: We need time to pass using the proposed process, so I suggest we temporarily
... allow the new process to go forward to gather more information and then revisit it.
... Otherwise we will block progress as Andreas and Pierre have pointed out.

nigel: This is a tricky situation - we have mutually exclusive blocking issues from each
... Editor (of TTML and IMSC) and an apparent group decision in Sapporo.
... We're out of time now, and this needs more thought, so I'll adjourn for today.
... Thanks all. [Adjourns meeting]

Summary of Action Items

Summary of Resolutions

  1. Publish the IMSC Errata document at https://github.com/w3c/imsc/blob/master/spec/ttml-imsc1-errata.html on /TR when it has been fixed for styling
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.144 (CVS log)
$Date: 2016/05/12 15:19:52 $