Re: Kicking off to the Text Tracks Community Group

It also should be a specification that someone who is creating time text
material can read to figure out what to output. There are always two sides
to an interface, not just the consumption side.

Jim King
-----Original Message-----
From: Silvia Pfeiffer <silviapfeiffer1@gmail.com>
Date: Mon, 10 Oct 2011 20:01:41 -0700
To: Ian Hickson <ian@hixie.ch>
Cc: "public-texttracks@w3.org" <public-texttracks@w3.org>
Subject: Re: Kicking off to the Text Tracks Community Group

>On Tue, Oct 11, 2011 at 11:25 AM, Ian Hickson <ian@hixie.ch> wrote:
>> On Tue, 11 Oct 2011, Silvia Pfeiffer wrote:
>>>
>>> 4. The WebVTT specification
>>>
>>> I'm sure you're all waiting to see the WebVTT specification in a place
>>> where we can all start contributing. Right now, there are some
>>>licensing
>>> issues to be sorted through, but I hope this can be resolved very soon.
>>
>> In the meantime, people should feel free to continue sending feedback as
>> they have been.
>>
>> One thing that would be helpful for me as editor is determining exactly
>> what aspects of the HTML standard people consider to be "WebVTT" so that
>> we can extract just those. Is it just the syntax? Does it include the
>> caption data model? Are the DOM APIs part of WebVTT? Is the rendering
>> model part of WebVTT? How about the CSS extensions? Unfortunately the
>> "webvtt.html" file people may have been looking at in the past is not a
>> particularly good starting point as it is very coarse -- it's missing
>>huge
>> chunks of defining material (e.g. there's no conformance section).
>
>
>These are great questions. I suggest we determine the answers by
>applying this question: "is this an important aspect to know about
>WebVTT when I am writing a WebVTT parsing application that is not a
>Web browser (i.e. basically a media player)?"
>
>The DOM and CSS side of things should be generic features of the
><track> element, so they need to be inside the HTML spec IMO.
>
>The abstract rendering description should, however, definitely be in
>the WebVTT spec, because we want the presentation the be the same
>across all players.
>
>So, I'd basically start with the WebVTT spec at
>http://www.whatwg.org/specs/web-apps/current-work/webvtt.html and add
>the text from 
>http://www.whatwg.org/specs/web-apps/current-work/multipage/rendering.html
>#webvtt-cue-text-rendering-rules
>in a media-element independent manner by just referring to the video
>viewport.
>
>Do you think that would be possible?
>
>Cheers,
>Silvia.
>

Received on Tuesday, 11 October 2011 14:25:36 UTC