Re: HTML5 CR "at risk" features

On Wed, Apr 16, 2014 at 9:12 AM, Jay Munro <jaymunro@microsoft.com> wrote:
> I'm curious why the wg went from a more generic addTextTrack to a very specific VTTCue and DataCue? The W3C also specs out TTML[1] and Simple Delivery Protocol (SDP)[2] which is based on TTML. Seems to be kind of limiting.

I don't want to repeat the arguments or re-open the discussion - you
can read up on the following bugs:
https://www.w3.org/Bugs/Public/show_bug.cgi?id=21851
https://www.w3.org/Bugs/Public/show_bug.cgi?id=13359
https://www.w3.org/Bugs/Public/show_bug.cgi?id=22903


> Are there plans on making a TTMLCue or something similar.

Glenn will know, but I think so:
http://www.w3.org/WAI/PF/HTML/wiki/TTML_mapping_to_HTML_5
Alternatively, DataCue could be of use, too.

Cheers,
Silvia.

> 1. http://www.w3.org/TR/ttaf1-dfxp/
> 2. http://www.w3.org/TR/ttml10-sdp-us/
>
>
>
> -----Original Message-----
> From: Jay Munro [mailto:jaymunro@microsoft.com]
> Sent: Tuesday, April 15, 2014 10:46 AM
> To: Silvia Pfeiffer
> Cc: Sam Ruby; public-html
> Subject: RE: HTML5 CR "at risk" features
>
> Good idea.
>
> -----Original Message-----
> From: Silvia Pfeiffer [mailto:silviapfeiffer1@gmail.com]
> Sent: Monday, April 14, 2014 2:43 PM
> To: Jay Munro
> Cc: Sam Ruby; public-html
> Subject: Re: HTML5 CR "at risk" features
>
> On Tue, Apr 15, 2014 at 4:28 AM, Jay Munro <jaymunro@microsoft.com> wrote:
>> I brought this up with Robin, and we tried it in a couple of browsers using the debugger. It appears to be supported by at least two.
>>
>> I wrote a demo that I use in my documentation, and I've posted it on the sample server. It appears to work fine in IE11 and Chrome. Firefox seems not to like the TextTrack info, but I didn't debug it.
>>
>> Here's the url:
>> http://samples.msdn.microsoft.com/Workshop/samples/media/addtexttrackt
>> est.html
>>
>> Sample's pretty simple:
>>
>> <!DOCTYPE html >
>>
>> <html >
>>   <head>
>>     <title>Add Text Tracks example</title> </head> <body>
>>
>> <video id="video1" controls="controls" muted="muted">
>>      <!-- change to your own mp4 video file -->
>>   <source
>> src="http://ie.microsoft.com/testdrive/Videos/BehindIE9ModernWebStanda
>> rds/Video.mp4" />
>
> You need to add an alternative source file in WebM or Ogg format for Firefox and other browsers to work.
>
>> HTML5 Video not supported
>> </video>
>>
>>  <script>
>>     var video = document.getElementById("video1");
>>     var startTime, endTime, message;
>>     var newTextTrack = video.addTextTrack("captions", "sample");
>>     newTextTrack.mode = newTextTrack.SHOWING; // set track to display
>>     // create some cues and add them to the new track
>>     for(var i=0;i<30;i++){
>>         startTime =  i * 5 ;
>>         endTime = ( (i * 5) +  5);
>>         message =  "This is number " + i;
>>         newTextTrack.addCue(new TextTrackCue(startTime, endTime,
>> message));
>
>
> TextTrackCue has no constructor any more, so this is not spec conformant. You have to create a VTTCue or a DataCue.
>
> Chrome and IE's support of the constructor need to be deprecated.
> (This is the case for the WHATWG and the W3C specs).
>
>
> HTH.
>
> Cheers,
> Silvia.
>
>>     }
>>     video.play();
>>   </script>
>> </body>
>> </html>
>>
>>
>> -----Original Message-----
>> From: Silvia Pfeiffer [mailto:silviapfeiffer1@gmail.com]
>> Sent: Saturday, April 12, 2014 9:37 PM
>> To: Sam Ruby
>> Cc: public-html
>> Subject: Re: HTML5 CR "at risk" features
>>
>> On Sat, Apr 12, 2014 at 8:31 PM, Sam Ruby <rubys@intertwingly.net> wrote:
>>> On 04/11/2014 09:41 PM, Silvia Pfeiffer wrote:
>>>>
>>>>> HTMLMediaElement.addTextTrack - Discussed at F2F meeting
>>>>
>>>>
>>>> I tried to find out from the minutes why addTextTrack is at risk,
>>>> but it was only mentioned without reasons.
>>>
>>>
>>> From http://www.w3.org/2014/04/09-html-wg-minutes.html :
>>>
>>> ... lots of failures on HTMLMediaElement.addTextTrack ... mostly
>>> failing at the moment
>>
>> That's what I saw. I dug around a bit and assume it's based on this test:
>> http://w3c.github.io/html/test-results/less-than-2.html#test-file-62
>>
>> At first sight, I think the test might expect the wrong kinds of errors.
>> Will investigate.
>>
>> Cheers,
>> Silvia.
>>

Received on Wednesday, 16 April 2014 07:32:05 UTC