Re: WebM id mapping

Makes sense to me. We’ll probably want to change Chromium to use TrackNumber once that change it made.

From: Aaron Colwell <acolwell@google.com<mailto:acolwell@google.com>>
Date: Thursday, May 8, 2014 at 10:09 PM
To: "public-inbandtracks@w3.org<mailto:public-inbandtracks@w3.org>" <public-inbandtracks@w3.org<mailto:public-inbandtracks@w3.org>>
Subject: WebM id mapping
Resent-From: <public-inbandtracks@w3.org<mailto:public-inbandtracks@w3.org>>
Resent-Date: Thursday, May 8, 2014 at 10:09 PM

Hi,

I was looking at the Sourcing In-band Media Resource Tracks from Media Containers into HTML<http://rawgit.com/silviapfeiffer/HTMLSourcingInbandTracks/master/index.html> spec and wanted to suggest that the WebM mapping for id be changed to use TrackNumber instead of TrackUID. My reasoning is that TrackNumber is akin to the track_ID in ISOBMFF, pid in MPEG2-TS, and serialnum in Ogg. Like all these other formats it is the actual ID that is used to demux blocks/samples/packets in the container.  A container without a TrackUID, while not totally compliant, is still usable because the TrackNumber is what actually binds the track info to a sequence of blocks. A missing TrackNumber leads to an unusable file. I think using TrackNumber is more consistent with your choices for the other formats.

Thanks,
Aaron