- From: Philip Jägenstedt <philipj@opera.com>
- Date: Fri, 22 Oct 2010 10:38:19 +0200
- To: public-media-fragment@w3.org
On Thu, 21 Oct 2010 12:10:55 +0200, Silvia Pfeiffer <silviapfeiffer1@gmail.com> wrote: > On Thu, Oct 21, 2010 at 8:30 PM, Chris Double <cdouble@mozilla.com> > wrote: >> On Thu, Oct 21, 2010 at 10:25 PM, Philip Jägenstedt <philipj@opera.com> >> wrote: >>> >>> Having read that thread, it seems to me that SMPTE should be treated as >>> labels, not as something to be converted into a timecode. In other >>> words, if >>> the resource doesn't contain these SMPTE labels, then one can't use the >>> format. I'd be happy with the spec saying as much and simply not >>> supporting >>> the syntax, as neither Ogg nor WebM can embed SMPTE timecodes. >> >> I'm inclined to agree. > > Interesting approach. That's certainly a valid way to approach it. > > Or we can be pragmatic and say that if you yourself know that your > video has a certain framerate then you can pick the correct SMPTE > timecode and you can address frame-accurately with that timecode. Any > other SMPTE timecode will not give you more accuracy than a normal npt > time. > > I'm myself critical about a need for such frame-accurate URI > addressing, but that email thread proves there are people that think > it's required. > > I do believe we can safely ignore it for now in implementations and > wait to see the need. In all the years of doing Annodex, nobody every > needed it. I'd perhaps go even further and say that if there's no implementor interest nor compelling use case, then it shouldn't be in the spec at all. -- Philip Jägenstedt Core Developer Opera Software
Received on Friday, 22 October 2010 08:38:57 UTC