Re: Proposal from HbbTV

Dear Bob,

On 2014-09-29, at 17:53 , Bob Lund <B.Lund@CableLabs.com> wrote:

> [...]
>> Wouldn't exposing captions as a video track + burned in captions create
>> weird cases when you have multiple video tracks (multiple camera angles
>> for example)? If the captions apply to all of the videos, then you'll
>> have an explosion of video tracks instead of a simple list of video
>> tracks and a simple list of captions to go with it.
> 
> The current HTML spec doesnıt support a cue-less text track very well so
> treating it as a video track with burned-in text is an alternative. This
> could result in a large number of these tracks for media resources with N
> video tracks and M cue-less text tracks (N x M).

Now multiply that by a couple of more languages (I'm in the European market), and by a list of disabilities (hard of hearing, visually impaired, ...). That will make it four factors.

> Another alternative is to enhance text track semantics to better
> accommodate cue-less text tracks. One could define a new text track mode,
> e.g. ³UARendered" defined as ³Indicates that the text track is active, the
> user agent is actively displaying the data in the track but no cues are
> active and no events are fired.²
> 
> The only legal modes for such a track would be ³disabled² or ³UARendered².
> [...]

That would IMHO be the best solution.


Many thanks and cheers,

  --alexander

Received on Monday, 29 September 2014 16:00:39 UTC