[theora-dev] Re: [xiph-rtp] Re: [Vorbis-dev] Proposal: An extension to rules all others

[Silvia Pfeiffer]

An addition of the MIME type to the first Ogg page doesn't make much
sense, since that Ogg page already belongs to one of the logical
bitstreams in the stream.

We had a long discussion about how to do this right when developing
the Annodex format
(http://annodex.net/TR/draft-pfeiffer-annodex-02.txt), which has been
built to provide this functionality and others.

In the end, we defined what is now know as "skeleton"
(http://wiki.xiph.org/index.php/Ogg_Skeleton) and as an additional
logical bitstream at the beginning of an Ogg file that contains
information o the other logical bitstreams inside the Ogg file.

This is the only way to provide this information without breaking
existing specifications.

However, it requires that Ogg parsing software be adapted to also
parse Skeleton.

Some of the software has already been adapted for it - others hasn't.

Silvia.

On 1/17/07, Tor-Einar Jarnbjo <tor-einar at jarnbjo.name> wrote:
> Aaron Colwell schrieb:
>
> > Media applications are able to deeply inspect the file if they really
> > need to
> > determine which codec it contains.
> >
> In case of Ogg, the application is unfortunately only able to determine
> which codec(s) the Ogg file contains if detailed knowledge about the
> codecs is known already by the Ogg parser. I think I already suggested
> to e.g. add the stream's MIME type to the first Ogg page to make it
> easier for players to cope with new codecs embedded in an Ogg file.
>
> Tor
>
> _______________________________________________
> Vorbis-dev mailing list
> Vorbis-dev at xiph.org
> http://lists.xiph.org/mailman/listinfo/vorbis-dev
>