Re: Revision timing object spec - section 10 on timed data

On 2015-08-13 10:24, Ingar Mæhlum Arntzen wrote:
>
> Hi all.
>
> I've revised section 10 concerning how the timing object may be used to drive timed presentation of timed data.
>
> - Added an introduction

Pretty neat!


> - removed changes to the timing object definition in the interest of keeping a clear separation between timing object and the timing text track

Works for me. That said, I would not be surprised if we end up needing a way to talk about the "list of timing text tracks" that are associated with a given timing object in the end (the "time marches on" procedure uses the list of text tracks for instance), which would require re-introducing that list one way or the other. That's not needed right now, though.


> - defined two approaches
> 1) library approach : generic JavaScript sequencer
> 2) integration with existing standards: timed text tracks

I agree with pursing the two approaches in the group but the spec is already the second bullet there, so I don't think it makes sense to present the approach within the spec. I updated the prose accordingly, linking to the JavaScript sequencer implementation in a note.


> - added some issues for discussion
>
> I'm still a little worried that the introduction of timed text tracks might not be extremely valuable (see issues) and I would like to have your feedback on this issue - or questions if the issue is not clear.

What strikes me as good is that it presents a possible solution based on concepts that people fluent in media matters should already be familiar with. Given enough support for the feature, this can be entirely discarded in favor of a "cleaner" solution as described in the issues!

Francois.

Received on Wednesday, 19 August 2015 11:13:12 UTC