- From: Silvia Pfeiffer <silviapfeiffer1@gmail.com>
- Date: Wed, 10 Aug 2011 20:22:01 +1000
- To: Scott Wilson <scott.bradley.wilson@gmail.com>
- Cc: Cyril Concolato <cyril.concolato@telecom-paristech.fr>, public-web-and-tv@w3.org
On Wed, Aug 10, 2011 at 8:08 PM, Scott Wilson <scott.bradley.wilson@gmail.com> wrote: > > On 10 Aug 2011, at 10:46, Silvia Pfeiffer wrote: > >> On Wed, Aug 10, 2011 at 5:22 PM, Cyril Concolato >> <cyril.concolato@telecom-paristech.fr> wrote: >>> Hi Sylvia, >>> >>> Le 10/08/2011 02:38, Silvia Pfeiffer a écrit : >>>> >>>> I wonder if instead it might be worth analysing if we can come up with >>>> a<track> kind that allows overlaying hyperlinkable regions onto the >>>> video? >>> >>> Why wouldn't it be possible to have a track element point to some animated >>> SVG file? >> >> The <track> element points to timed text, i.e. to a file that provides >> text fragments along the timeline of the video. SVG is not suitable >> for that use. Just like you cannot put a SVG into a <audio> or <video> >> element, you won't be able to put a SVG into a <track> element. >> >> However, we can e.g. create a WebVTT file with data in the cues that >> point to SVG files or whatever else we want to time-align with the >> video and link that in the <track> @src with @kind=metadata. > > > Right, so we would need a WebVTT cue node type for such a link to the SVG or whatever. Maybe a HTML <a> or <link> tag. > > However I think a major point of WebVTT is it is simple to create and to parse tracks, which is why it has such a restricted document model, so we should be parsimonious in extending the spec. Yes, we need to be careful how we extend it. Regards, Silvia.
Received on Wednesday, 10 August 2011 10:22:47 UTC