Re: edited TTWG draft charter, VTT?

> On Mar 26, 2018, at 12:32 , David Ronca <dronca@netflix.com> wrote:
> 
> > I think that a label from the W3C (“Rec.”) may be not that important to them.
> 
> Then the issue is solved, is it not?

That’s why I am asking the question, at the charter transition time.

We’re also bogged down in the other members of the committee nit-picking and deliberately delaying; maybe to make the other work look better. We wasted an entire month recently, for example. The timed text working group is a hostile environment for anyone who is interested in captioning in general, and not XML-dialects in specific.


> 
> On Mon, Mar 26, 2018 at 12:31 PM, David Singer <singer@apple.com> wrote:
> 
> 
> > On Mar 26, 2018, at 12:20 , David Ronca <dronca@netflix.com> wrote:
> >
> > > I don’t think it very constructive to get into a comparison argument of the two specifications.
> >
> > The only thing I am comparing is the level of support.  TTML2/IMSC2 has broad industry support with many organizations willing to invest time and money to move the work forward.  WebVTT does not.
> 
> OK, you do want an unproductive discussion, so be it.
> 
> VTT is supported in all major browsers; TTML is not, and probably never will be.
> 
> With the exception of Regions, the support for VTT is remarkably complete and consistent. TTML dialects slide through the ’N implementatuins’ requirement because there are M>N candidate implementations, and for any given feature, you can cherry-pick which implementations support it correctly; so for any row in the table (row==feature) there are indeed 3 green check-marks — but there’s no solid consistent set of columns which support the basic feature set interoperably. The amount of red, last time I checked, was extraordinary.
> 
> Yes, a lot of committees endorse TTML dialects — W3C, EBU, SMPTE, and so on — but rather fewer engineers. The opposite is true for VTT.
> 
> I have repeatedly asked where TTML is achieving ‘blind interoperability’ — where TTML-dialect files are being written without knowing or caring which client will play them. This is tthe major function of a public specification, and so far, I have not heard of a good answer. (Both the BBC and Netflix’s use of TTML (iPlayer and the Netflix player) are effectively closed ecosystems, where the player is provided by the same company as is authoring the files.)
> 
> > You should not expect those who are not committed to WebVTT to carry the water.
> 
> I’m not.
> 
> > Find one company willing to invest in the project at a level that matters, and your problem is solved.
> 
> They’ve implemented and moved on, and I think that a label from the W3C (“Rec.”) may be not that important to them.
> 
> >
> > On Mon, Mar 26, 2018 at 12:09 PM, David Singer <singer@mac.com> wrote:
> > I don’t think it very constructive to get into a comparison argument of the two specifications. Honestly, they both have problems.
> >
> > The investment in the many implementations of VTT, including in browsers, happened years ago, it’s true. There are open-source, browser-based, standalone, polyfill and translation implementations. In fact, one problem is that people don’t see a need to be involved in formal processes after the time when it was all implemented; they’ve implemented and moved on.
> >
> > Can we drop this “your spec. is X” line of discussion? I don’t see it yielding any useful insights.
> >
> > > On Mar 26, 2018, at 12:04 , David Ronca <dronca@netflix.com> wrote:
> > >
> > > > thats what Im talking about: the time spent in this group on moving WebVTT to CR has been negligible.
> > >
> > > We are all doing the W3C work as representatives of our companies/organizations, and you should not be surprised that the collective priorities of the participanting individuals and organizations is where the time gets spent.  What should concern you is that there are no companies/organizations or even individuals that believe that WebVTT is important enough to invest time or money in the effort, nor to even join the weekly calls.  How can WebVTT move forward with no one willing to do the heavy lifting, and why should it?
> > >
> > > >  It has more implementations and complete features than the very first time TTML went to CR
> > >
> > > What happened 10 years ago is not relevant.  TTML2 and IMSC2 will have multiple independent implementations of every feature.  Included in this is the open source TTT rendering project.  This is years of collective effort and many millions of dollars invested by many organizations to see this through.  Again going to my previous comments.  The WebVTT problem is that there are no companies/organizations willing to make a similar investment in WebVTT to complete the project.  It is not unreasonable, therefore, to conclude that the collective industry has decided that WebVTT is no longer relevant, and moved on.
> > >
> > > On Mon, Mar 26, 2018 at 11:32 AM, Silvia Pfeiffer <silviapfeiffer1@gmail.com> wrote:
> > > Hi Pierre,
> > >
> > > thats what Im talking about: the time spent in this group on moving
> > > WebVTT to CR has been negligible. All the effort has been done by
> > > others outside the group. This group was tasked with a transition it
> > > basically has no interest in.
> > >
> > > And about exit criteria: this is the first CR of WebVTT. It has more
> > > implementations and complete features than the very first time TTML
> > > went to CR - there's no way you can compare the status of the two
> > > specifications at first CR and not see this difference.
> > >
> > > I am not angry about this - it is what it is. I'd just like the
> > > decision to be made.
> > >
> > > Kind Regards,
> > > Silvia.
> > >
> > >
> > > On Tue, Mar 27, 2018 at 5:26 AM, Pierre-Anthony Lemieux
> > > <pal@sandflow.com> wrote:
> > > > Hi Silvia,
> > > >
> > > >> No other specification in the history of W3C has had to jump through this many hoops to get to CR.
> > > >
> > > > As far as I know, WebVTT has been held to the same criteria for CR
> > > > transition as IMSC (and TTML2).
> > > >
> > > > For better or for worse, the group has collectively spent multiple
> > > > hundreds of man-hours closing
> > > > issues leading up to the TTML2 CR.
> > > >
> > > > Best,
> > > >
> > > > -- Pierre
> > > >
> > > > On Mon, Mar 26, 2018 at 11:10 AM, Silvia Pfeiffer
> > > > <silviapfeiffer1@gmail.com> wrote:
> > > >> If there is a decision this group should take, it's to move webvtt to CR. I
> > > >> challenge you to get this done rather than throw another process roadblock
> > > >> at webvtt.
> > > >>
> > > >> It's ready and has been for a long time. No other specification in the
> > > >> history of W3C has had to jump through this many hoops to get to CR.
> > > >>
> > > >> All other groups of the W3C have deemed it ready for months of not years.
> > > >>
> > > >> Do everyone a favour and just decide to move it.
> > > >>
> > > >> Kind regards,
> > > >> Silvia.
> > > >>
> > > >>
> > > >> On Tue., 27 Mar. 2018, 3:16 am Nigel Megitt, <nigel.megitt@bbc.co.uk> wrote:
> > > >>>
> > > >>> Not so much an opinion as a restatement - at this point we seem to be in
> > > >>> the realm of process. We made the following resolution at TPAC 2016 in
> > > >>> Lisbon [1] as proposed by David Singer:
> > > >>>
> > > >>> > RESOLUTION: If we do not move WebVTT to CR in this Charter period then
> > > >>> >we will not include it in any new Charter.
> > > >>>
> > > >>>
> > > >>> [1] https://www.w3.org/2016/09/19-tt-minutes.html#resolution01
> > > >>>
> > > >>> If there's another way to meet the needs of the folk who want to progress
> > > >>> VTT, then one way to satisfy that resolution is to close this line of work
> > > >>> in TTWG and remove it from the draft new Charter; I think the Process
> > > >>> requires publication as a WG Note in this event. I assume that this does
> > > >>> not preclude taking it up again in another WG or other forum.
> > > >>>
> > > >>> I can't actually see any other ways to satisfy the resolution - are there
> > > >>> any?
> > > >>>
> > > >>> Otherwise we would need to make a new different decision very quickly (and
> > > >>> our Decision Policy review period for any decision made now expires after
> > > >>> the end of the current Charter, so time is not in favour); before doing
> > > >>> that I would ideally like to see some evidence that something has changed
> > > >>> to warrant us revisiting it.
> > > >>>
> > > >>> For tracking purposes, I raised
> > > >>> https://github.com/w3c/charter-timed-text/issues/21 on Thursday 22nd
> > > >>> March.
> > > >>>
> > > >>> Nigel
> > > >>>
> > > >>>
> > > >>>
> > > >>> On 23/03/2018, 17:21, "David Singer" <singer@mac.com> wrote:
> > > >>>
> > > >>> >I¹m in debate with Silvia over the progression of VTT.  Do others have
> > > >>> >opinions?
> > > >>> >
> > > >>> >With WhatWG on a firmer footing, much of the original motivation to do a
> > > >>> >Rec. has evaporated, and at this point looks like a lot of busy work to
> > > >>> >little reward. Are there people who see it differently?
> > > >>> >
> > > >>> >> On Mar 23, 2018, at 3:16 , Thierry MICHEL <tmichel@w3.org> wrote:
> > > >>> >>
> > > >>> >> Hi,
> > > >>> >>
> > > >>> >> I have edited the TTWG draft charter with the changes agreed during our
> > > >>> >>TTWG telecon.
> > > >>> >> Please review the draft charter
> > > >>> >>
> > > >>> >> https://w3c.github.io/charter-timed-text/Draft-2018-TTWG-Charter.html
> > > >>> >>
> > > >>> >> Deadline is Monday midnight US coast. I will send the charter on
> > > >>> >>tuesday to W3M for approval, and will request a charter extension.
> > > >>> >>
> > > >>> >> Thierry
> > > >>> >>
> > > >>> >
> > > >>> >David Singer
> > > >>> >
> > > >>> >singer@mac.com
> > > >>> >
> > > >>>
> > > >>>
> > > >>
> > >
> > >
> >
> > David Singer
> >
> > singer@mac.com
> >
> >
> >
> 
> David Singer
> Manager, Software Standards, Apple Inc.
> 
> 

David Singer
Manager, Software Standards, Apple Inc.

Received on Monday, 26 March 2018 19:35:42 UTC