[vorbis-dev] RFC draft for Vorbis over RTP

[Jack Moffitt]

> Yes, but this is only valid for the current encoders. A future encoder 
> may choose to create the codebooks while encoding, making an optimized 
> set for the current input. Like sort of integrating Segher's tool 
> (which name I've forgotten) with a current encoder. A chain with 
> two such files would then not be streamable.

This still doesn't prevent streaming.  Only makes simulated live
streaming slightly more bandwidth intensive.

> Yes, but I'm not sure if I like that limitation. Broadcasters are 
> currently using chains to allow a midstream change in the metadata 
> (comment header, title, artist) and this is a feature which should 
> be preserved when using RTP too. One option would of course be to 
> allow comment headers to appear between audio packets.  

This midstream change of metadata via codebook update is a hack, always
was a hack, and will ocntinue to be a hack.  I did it because I was
lazy.  There should be a metadata protocol there.

> So what do you think about restricting the identification header 
> and the setup header to remain static for the entire broadcast, but 
> allow the comment header to be repeated? Or perhaps use some other 
> content type to deliver meta data?

The latter.  Fix codebooks for the stream, use a side channel for
metadata like you are supposed to.

jack.
--- >8 ----
List archives:  http://www.xiph.org/archives/
Ogg project homepage: http://www.xiph.org/ogg/
To unsubscribe from this list, send a message to 'vorbis-dev-request at xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is needed.
Unsubscribe messages sent to the list will be ignored/filtered.