tool to add skeleton (was Re: [Vorbis-dev] Re: [ogg-dev] Peer review draft for the new)

[Silvia Pfeiffer]

On 10/2/07, Ian Malone <ibmalone at gmail.com> wrote:
> On 02/10/2007, Conrad Parker <conrad at metadecks.org> wrote:
> > On 02/10/2007, Ian Malone <ibmalone at gmail.com> wrote:
> > > Martin Leese wrote:
> > > > 2.  Following on from 1, a tool which inputs an
> > > >     Ogg containier with one or more streams, and
> > > >     outputs a new Ogg container with the same
> > > >     streams plus an Ogg Skeleton stream stuffed
> > > >     in front would be very helpful.  The tool could
> > > >     also select and create the appropriate file
> > > >     extension.
> > > >
> > >
> > > This is pretty easy to write using liboggz from svn
> > > and Tahseen's skeleton creating functions (in his
> > > vorbis-tools branch).  I just haven't found an
> > > afternoon to do it yet.

Do it! :-)

> > "hogg addskel" does what we want, you can get hogg from
> > http://www.kfish.org/software/hogg
> > However, it's a lot slower than an oggz version would be (due to my
> > poor Haskell optimization skills, not due to the choice of language

Oath! ;-) Have you created binary distributables of this yet? Maybe
ppl would use it!?

> Apropos this; I was wondering about mime-types to be used in
> Skeleton.  audio/vorbis for instance is not a current recommendation
> for general use to avoid irritation to the IETF (as I understand) and
> currently, for example, audio/x-ogg is suggested.  What is the opinion
> on using the final codec mime types (audio/vorbis for instance) in
> Skeleton, as this information is static and as it doesn't overlap with
> anything anyone else is doing?

So: the mime types inside skeleton should be the same that are being
used inside encapsulation-format-free spaces, such as RTSP/RTP.
audio/vorbis is suggested for use where it just refers to a stream of
vorbis packets (see
http://wiki.xiph.org/index.php/MIME_Types_and_File_Extensions).
audio/ogg is instead recommended for any type of audio encapsulated in
ogg.
So: the ogg file (.oga extension) would have audio/ogg mime type and
inside it would live a skeleton which says audio/vorbis or audio/speex
about its audio bitstream. This enables audio applications to pick up
the file and do something with it, while possibly ignoring a .ogv
file.

If we haven't made this clear yet, that should be fixed! :-)

Cheers,
Silvia.