{Minutes} TTWG Meeting 2020-12-03

Thanks all for attending today's TTWG meeting. Minutes can be found in HTML format at https://www.w3.org/2020/12/03-tt-minutes.html

In text format:

   [1]W3C

      [1] https://www.w3.org/

                Timed Text Working Group Teleconference

03 December 2020

   [2]Previous meeting. [3]Agenda. [4]IRC log.

      [2] https://www.w3.org/2020/11/26-tt-minutes.html
      [3] https://github.com/w3c/ttwg/issues/161
      [4] https://www.w3.org/2020/12/03-tt-irc

Attendees

   Present
          Andreas, Atsushi, Cyril, Gary, Mike, Nigel, Pierre

   Regrets
          -

   Chair
          Gary, Nigel

   Scribe
          nigel

Contents

    1. [5]This meeting
    2. [6]IMSC Tests open pull requests
    3. [7]MPEG Liaison #167
    4. [8]Patent Policy 2020
    5. [9]AOB: frequency of TTWG teleconferences
    6. [10]Meeting close

Meeting minutes

   <atsushi> (still in previous meeting, sorry)

  This meeting

   Nigel: today, we have IMSC Tests pull requests to iterate
   through and form a plan of action
   … the MPEG liaison response draft
   … confirmation of the resolution to adopt patent policy 2020
   … and an AOB on telecon frequency
   … Any other business?

   group: [no other business]

  IMSC Tests open pull requests

   Nigel: 4 open PRs, some quite old.
   … Thanks Pierre for raising this last week. We need a plan of
   action to resolve them.
   … Should we iterate through them?

   Pierre: #97 is ready to go I think. I asked Cyril to review but
   others can too.

   Cyril: Yes I will review.

   Pierre: The other ones have outstanding comments to address

   Nigel: Thanks for reminding us about these - some have been
   open for a long time.
   … I will try to get round to looking at them, and also welcome
   everyone else to as well.
   … Any points to discuss?

   Pierre: No.
   … just to note that Glenn mentioned they don't look like CR
   exit criteria tests (see #96) and I
   … pointed out that they aren't intended for that. He seems to
   have accepted that.

   Nigel: Yes, I noticed that too, I think we can continue.
   … It's the Implementation Report that lists the relevant CR
   Exit Criteria tests.

   Cyril: On the subject of testing, I was wondering if Andreas's
   examples for ttml2#1211 could be added to a test suite?

   Andreas: Yes, I'm fine with that. I think the main case why I
   contributed is we were missing some examples
   … where weak, neutral, ltr and rtl characters are mixed.
   … Of course yes I'm fine with it.

   Nigel: And Cyril, you noticed you didn't agree with one of the
   renderings?

   Cyril: Yes, but I just responded, I made a mistake, I think
   they're fine.
   … They're very interesting tests and it would be a pity not to
   have them in the test suite.

   Nigel: Who can take the action?

   Cyril: I can.

   Nigel: Great, thank you.

  MPEG Liaison #167

   Nigel: [reviews document shared via member-tt reflector link]

   group: [discusses content of draft outgoing liaison text]
   … [discussion of clipping behaviour defined by ISOBMFF as being
   related to the document processing context in TTML1/2]

   Mike: More clarity on the clipping behaviour would be helpful.

   Cyril: Root temporal extent and presentation processing context

   Nigel: Any other points to be made?

   group: [discusses the time coordinate Ti and their timeline]

   Nigel: Thanks, I wanted there to be some draft text captured -
   I will leave this up and not send it without further review.

  Patent Policy 2020

   Nigel: The decision review period for this ends now,
   effectively, and I have seen no objections to our resolution to
   adopt.

   Gary: I have not either.

   Cyril: I checked with our legal team and they're fine with it.

   Nigel: In that case we can go ahead. Not sure the next step -
   there's a WBS I think?

   Atsushi: [checks]

   Nigel: I've found the WBS poll
   … I've submitted the form.

   Gary: One thing to remember is that some people will have to
   rejoin the group after re-chartering.

   Atsushi: This will go through AC review and then recharter,
   then everyone will need to rejoin.

   Nigel: Thanks for the clarification.

   Atsushi: It will be 8th January or later.

   Gary: Hopefully there will be a heads-up when it happens.

   Atsushi: An automatic notification will go through and I will
   remind in meetings and by email.

   Gary: Thanks.

   Nigel: Thank you everyone.

  AOB: frequency of TTWG teleconferences

   Nigel: Thanks to Pierre for raising this. I'd like to say my
   thoughts are not fully formed on this yet, and I'm happy to
   discuss.
   … The main point I would make is we have a lot of deliverables
   slated, and progress has been slow on most of them.
   … Work drives the need for discussion!

   Pierre: As per my email, I think teleconferences are really
   useful for controversial or complex issues or as a forcing
   function.
   … Because of the workload today I don't see the need for it
   every week.
   … We should try to do more work asynchronously using electronic
   tools.

   Andreas: I mentioned on the mailing list, our workflow could
   evolve a bit.
   … This combination of mailing list and telephone conferences
   has been established many years ago.
   … We have moved a bit. Especially more chat tools with
   threading and real time or asynchronous communication offer a
   great
   … way to collaborate efficiently. If we add this to our tools
   or standards work it could possibly reduce the
   … frequency of telephone conferences.

   Gary: There is a W3C Slack channel

   Pierre: Also email and GitHub too

   Andreas: I think the threads on GitHub can explode - I'm not
   sure if they are the right place.
   … I know GitHub is looking to establish a different kind of
   communication.
   … The writing mode discussion was a good example.
   … Email is also possibly not ideal, because the whole list gets
   flooded with issues that are concerning only for a few people.
   … I could imagine making Slack more efficient as a tool for
   communication on certain topics.
   … The Slack channel itself won't help, we need to discuss how
   to use it.

   Nigel: I'm concerned about getting the right balance on is the
   public nature of the decisions.
   … Sometimes the record of how we reached a decision is helpful
   in retrospect, and if we move discussion off easily searched
   … or publicly accessible media then we may lose that benefit.

   Nigel: But anything that encourages more frequent/lower latency
   discussion is a good thing.

   Cyril: I would agree to the request to meet less often, in
   general. But in the past months for example, I don't think we
   had so
   … many cases where we could skip the meeting. We skipped some
   meetings and had one or two short meetings.
   … Maybe simply encouraging the chairs to skip a meeting when
   the agenda is too light?
   … Move more towards a meeting on a needs basis rather than a
   fixed cadence could help.

   Pierre: The downside, Cyril, is that you still need to reserve
   that 1 hour every week at a critical time for international
   meetings.
   … There aren't many timeslots that work globally so there's an
   opportunity cost,
   … We can all talk easily but the question is do we really need
   to?

   Cyril: I was also wondering about the impact on our decision
   policy if we move to a monthly cadence.
   … It means we could take a decision asynchronously and never
   discuss it. Maybe that's okay.

   Andreas: This is exactly what the meetings are for, to confirm
   decisions.
   … To be clear about agreement, the teleconferences are good.
   … They need to be prepared in a way that the agreement is clear
   before or any controversy has been worked out beforehand.

   <atsushi> fyi. decision policy from charter

   <atsushi> > If no objections are raised on the mailing list by
   the end of the response period, the resolution will be
   considered to have consensus as a resolution of the Working
   Group.

   Gary: Also we can always schedule an off-cadence meeting to
   discuss if something like Slack or another chat medium ends up
   not being good enough.

   Mike: What's the meeting notice requirement for WGs?

   Nigel: I'm not sure but I think it primarily applies to face to
   face meetings.

   Mike: They likely do say something about WG calls, anyway it
   should be factored in.

   Atsushi: Calls need to be announced 2 days before, and f2f 4 or
   8 weeks before, I think. (searching for the docs now)

   Nigel: Could be in our Charter, I haven't checked.

   Atsushi: The Process may enforce something.

   Nigel: The Charter does not say anything about it.

   Nigel: I'm definitely open to a change. My concerns are:
   … 1. visibility of discussions and accessibility to those who
   do not regularly participate
   … 2. potential further loss of momentum

   <atsushi> found! [11]https://www.w3.org/2020/
   Process-20200915/#GeneralMeetings

     [11] https://www.w3.org/2020/Process-20200915/#GeneralMeetings

   Nigel: 3. (ought not to be a concern, but) possibly poorer
   quality decision making if people don't think things through
   except by talking about them

   Gary: The document Atsushi found is a "should": announcement of
   meeting 1 week ahead and agenda >= 24 hours ahead.

   [12]Meeting requirements in the Process

     [12] https://www.w3.org/2020/Process-20200915/#GeneralMeetings

   Gary: It's possible that momentum could be increased by having
   availability on asynchronous media instead of having to wait
   until the next call.

   <atai> +1

   Nigel: Yes, good point.

   Gary: Maybe the quick thing here is that we should consider
   creating a TTWG chat and start using it.
   … If we are using that and don't feel the need for meetings
   then we can reduce the frequency.

   Nigel: Sounds like a good way forward to me.
   … Dip our toe in the water and see if we enjoy it.

   Cyril: So let's say experiment for 2 months with discussing in
   the chat and cancelling meetings if not needed,
   … and if that works then reduce the cadence officially.

   Gary: Seems reasonable to me, but this month might be weird -
   maybe extend through to the end of February.

   <atsushi> > Meeting Schedule Teleconferences: Usually once per
   week. (in charter)

   Pierre: My suggestion would be to remove the meeting cadence to
   2 weeks, and in between use those electronic means.

   Andreas: I agree with Pierre, try biweekly meetings and putting
   extra ones in if needed.

   <atsushi> +1 charter just says 'usually'

   Nigel: How about I raise the github issues for meetings through
   to the end of February and mark those that are slated for
   potential cancellation,
   … every 2 weeks?

   Andreas: Sounds good to me

   Nigel: And we're talking about using W3C Slack for async chat,
   right?

   Gary: That is my proposal, unless people think others are
   better.

   Cyril: W3C is the place, right.

   [13]Link to join the ttwg channel (active for 7 days)

     [13] https://join.slack.com/share/zt-jtt153mc-PpPGgklDCuDTb6igRnCxzg

   Nigel: Thanks everyone, we're out of time, I'm always happy to
   look at new ways of working.
   … The main thing is to make progress on our deliverables, how
   we do that is up to us.
   … By the way that Slack channel will not show older archived
   messages as it is on the free plan at the moment.

  Meeting close

   Nigel: Thanks everyone, we're out of time for today. [adjourns
   meeting]


    Minutes manually created (not a transcript), formatted by
    [14]scribe.perl version 124 (Wed Oct 28 18:08:33 2020 UTC).

     [14] https://w3c.github.io/scribe2/scribedoc.html