- From: Silvia Pfeiffer <silviapfeiffer1@gmail.com>
- Date: Fri, 29 Jul 2011 14:31:09 +1000
- To: tmichel@w3.org
- Cc: joakim.soderberg@ericsson.com, soohongp@gmail.com, Felix Sasaki <felix.sasaki@dfki.de>, "Evain, Jean-Pierre" <evain@ebu.ch>, "public-media-annotation@w3.org" <public-media-annotation@w3.org>
Hi Thierry, On Fri, Jul 29, 2011 at 12:20 AM, Thierry MICHEL <tmichel@w3.org> wrote: > Sylvia, > > I have updated the 3 HTML tables with your info. > I am not sure about my edits, please verify, would be easier if you do it, > really. Do you have the updated files somewhere? > It still have some issues: > > OGG > ----- > http://dev.w3.org/2008/video/mediaann/mediaont-1.0/mediaont-1.0.html#ogg-table > > > 1-fragments and namedFragments are marked as "N/A" in the "Relation" column > and are used in the RDF file marked as "Yes" in the "RDF tested" > column. This seems not right. Well, this is indeed dependent on whether you accept that the Media Fragment URI specification is applicable or not. You can get temporal fragment URIs for any media file with that spec and the named tracks provide you track fragment URIs. As for named sections, you can always map a name to a time. So, this is how I constructed the RDF file which has all these three types of fragments. > 2- not sure what to put in the following > title: second and third lines They would be a "no", since I haven't been able to add those fields. > 3- The following Descriptive Properties (Core Set) dont have a equivalence > in the "Ogg properties" column > > frameSize > duration > samplingRate > frameRate > bitRate That's because they are encoded in binary headers and can't just be pointed to and read like text. Thus I described how they are being calculated from the file. > WebM container > -------------- > http://dev.w3.org/2008/video/mediaann/mediaont-1.0/mediaont-1.0.html#webm-tablet > > > 1- not sure what to put in the following > identifier: second and third lines > title: second line > language: second and third lines Right now none of these are supported by WebM, but only by Matroska. So they would be a "no". > 2- The following Descriptive Properties (Core Set) dont have a equivalence > in the "Ogg properties" column > > averageBitRate That's because there is not an explicit field in the file for it, but it is calculated from binary headers. Hope this helps. Cheers, Silvia.
Received on Friday, 29 July 2011 04:31:56 UTC