Re: TextTrack API changes

On Thu, Apr 25, 2013 at 8:55 PM, Silvia Pfeiffer
<silviapfeiffer1@gmail.com>wrote:

>
> On Fri, Apr 26, 2013 at 1:41 PM, Glenn Adams <glenn@skynav.com> wrote:
>
>>
>> On Thu, Apr 25, 2013 at 8:33 PM, Silvia Pfeiffer <
>> silviapfeiffer1@gmail.com> wrote:
>>
>>> Glenn,
>>>
>>> I am an editor of the HTML specification and while HTML5.1 is still an
>>> Editor's draft, I can make changes that address bugs and that I think will
>>> be implemented because they make sense, even if they are not backwards
>>> compatible. The particular change under discussion is one that you have
>>> yourself pursued to make the <track> element and TextTrack less dependent
>>> on WebVTT. The change that I applied to HTML5.1 supported that move.
>>>
>>> I am happy for the discussion that we have around the change, because it
>>> was suggested to backport that change to HTML5 (which, incidentally, hasn't
>>> happened yet). Changes to HTML5 that are not editorial are indeed more
>>> complicated, which is why I seeked WG feedback. This process has not been
>>> finished yet, so I don't understand why you are attacking me for the work
>>> that is done, diligently and thoroughly.
>>>
>>
>> First, I am not attacking you.
>>
>
> Your wording is intimidating and you claim I have no right to do what I
> do, which is not true.
>

I have used the words "please" and "I would ask". That doesn't sound
intimidating. As for what you have the right to do, I would suggest that,
in the absence of WG guidance, you do not in fact have the right to make
unilateral substantive changes. Or, if you do make such changes, then you
should be prepared to revert them in the face of objections and defer the
issue to the WG.

In this particular case, you appear to be insisting that you have the right
to ignore such objections without WG guidance. Whether that is an editor's
right or not, it seems to me rather poor judgment to insist on taking such
a position.

All I have asked is that you revert the changes and defer the issue to the
WG. However, since you appear to refuse doing this, I have asked the chairs
for an agenda item on this point.


>
>
> I am asking you to revert a change for which there are objections. Yes, I
>> support moving truly VTT specific APIs out of TextTrackCue, but these two
>> members are not VTT specific, and are written in a manner to abstract the
>> differences in actual text track format.
>>
>>
>>> While you and two others are now disagreeing with a part of the change
>>> that was made, Simon from Opera has agreed with that change. I want to
>>> continue this discussion until we find a consensus position. Providing
>>> specifications for TextTrackCue for other formats than WebVTT is part of
>>> that process.
>>>
>>
>> You are suggesting that it is possible to find a consensus to make your
>> proposed change in the face of objections from members. I will suggest it
>> is not.
>>
>
> The discussion is still active and not finalized. You cannot know.
>
>
>>  You propose a backwards incompatible change.
>>
>
> The main backwards incompatible change is one you do not disagree with:
> TextTrackCue has no constructor any longer, but can only be constructed
> given a particular format.
>
>
>
>> This is not something that should be implemented in your editorial work
>> in the face of member objections.
>>
>
> True for HTML5 and I have not done so. Not true for HTML5.1 - it is an
> editor's draft only at this stage. If that was the case, the Encrypted
> Media Extension would not exist.
>
>
>
>> The correct option for you is to revert the change, then ask the matter
>> to be brought to the WG for consideration. After due consideration, if the
>> WG decides it is best to make this change, then it will be made.
>>
>
> I have approached this even more carefully: I have not made a change to
> HTML5, but instead asked for input first.
>
>
>>  However, at this time, you appear to be attempting to usurp the role of
>> the WG in making substantive changes, which is not part of your role as
>> editor.
>>
>
> The role of an editor is to make changes.
>

Only when non-controversial within the WG and at the pleasure of the WG.


>
>
>>> There are no change proposals involved at this stage - it is merely a
>>> discussion on the technical mailing list. If you want to go through the
>>> change proposal process, please follow the complete process as described in
>>> http://dev.w3.org/html5/decision-policy/decision-policy.html .
>>>
>>
>> Precisely. You are making substantive changes without a CP and without WG
>> approval. This is fine when there are no objections, but that is not the
>> case here.
>>
>
> Technical discussions on the HTML WG mailing list are encouraged. This is
> a technical discussion and it has not come to the end yet. There are
> objections for making the change and for not making the change, so the WG
> hasn't made up its mind yet.
>

No, at this point it is not a technical discussion. It is a procedural
discussion about your actions in the face of member objections.


>
>
>> It is not necessary to further discuss the technical merits of this
>> change with Opera in the face of these objections. That is something the WG
>> should undertake, not the editors.
>>
>
> I hope you are not implying that the opinion of Opera in the WG has no
> merit when other WG members oppose a change.
>

If there are objections to a change, then no change should be made until
those proposing the change obtain a consensus reading on making the change.
The latitude that you have as an editor is to make changes that are not
subject to objection within the group. I do not believe you have the right
to make changes that are subject to objection in the absence of explicit WG
instruction. If you believe this is not true, then I would be happy to
raise this point at our next F2F to determine the WG position.


>
> Regards,
> Silvia.
>