theora dev - Nov 2002 - Re: Ogg IETF standard (was: Re: application/ogg statusupdate)

Hmm, I know what you mean. I've actually taken these sentences out of
Monty's specifications. The first one refers to the page size of max.
~64 kByte and the second one to the logical subdivision of packets into
segments of 255 Byte size. I'm not sure how to rephrase that properly
though. Maybe it's good as it stands: that it is recommended to have 4-8
kByte size, but expected more commonly to be 50-200 Byte? I'm not
experienced enough to make anyone of these statements. Monty, your take?

Cheers,

Silvia.

<p>Erik Stenborg wrote:
> 
> I skimmed parts of the text and found one contradiction:
> 
> "  From Ogg's perspective, packets can be of any arbitrary size,
>    although a nominal size of approximately 4-8 kByte is recommended.
> [...]
>    The encoding is optimised for speed and the expected case of the
>    majority of packets being between 50 and 200 bytes large. "
> 
> /Erik
> 
> On Fri, 15 Nov 2002 Silvia.Pfeiffer@csiro.au wrote:
> 
> > Hi all,
> >
> > sorry for cross-posting, but I believe both developer communities
should
> > be interested in this.
> >
> > In order to get Ogg accepted as application/ogg MIME type, there was a
> > requirement by the IETF to get an RFC written on the Ogg encapsulation
> > format. I have now authored a draft of that document and would like to
> > get input from the community and especially from Monty. Linus and I
> > would like to submit it asap to IETF (their next meeting is next
week).
> > Please could everybody help me to make sure I have not made any
> > mistakes!
> >
> > In addition, as I have copied many words and paragraphs from the
online
> > documents, which I presume were authored by you, Monty, I would like
to
> > suggest to include you as author in this RFC.
> >
> > Let me know what you think ...
> >
> > Cheers,
> >
> > Silvia.
> >
> >
> > Silvia Pfeiffer wrote:
> > >
> > > Hi Linus,
> > >
> > > I'm just wondering how much further you got with writing the
RFC on the
> > > Ogg stream/file format (and of course the RFC on the
application/ogg
> > > MIME type).
> > >
> > > I would like to offer my help in writing an RFC on the "Ogg
media
> > > encapsulation format" and in pursuing it with the IETF for
> > > standardisation. We could e.g. write up a document in
RFC-compliant
> > > format (see RFC2629) rather quickly by using the
"xml2rfc" tools
> > > provided by IETF.
> > >
> > > I believe this effort is necessary and an excellent way to
further the
> > > publicity of Ogg and all the related media encoders. I'd be
very happy
> > > to get involved as I'd like to support the excellent efforts
of this
> > > community.
> > >
> > > Cheers,
> > >
> > > Silvia.
> > >
> > > Linus Walleij wrote:
> > > >
> > > > On Mon, 26 Aug 2002 Silvia.Pfeiffer@csiro.au wrote:
> > > >
> > > > > are you planning to write an RFC on the Ogg file format
or on the ogg
> > > > > vorbis encoding format? Also, are you thinking about
getting
> > > > > audio/ogg-vorbis and video/ogg-vp3 registered as mime
formats?
> > > >
> > > > I have written an RFC for the application/ogg MIME type, so
to support
> > > > this I might have to write an RFC for the Ogg stream/file
format. I yet
> > > > don't know if I have to do that though, as my aim was
primarily to get the
> > > > mimetype registered.
> > > >
> > > > There is a definition for audio/vorbis that will be common
for both RTP
> > > > streamed "raw" vorbis and Ogg-packaged vorbis.
However it is not yet
> > > > finished, some guy in Finland is working with it.
> > > >
> > > > Noone has thought about video/theora (as I think it should
be named) and
> > > > with the extreme complexity that has surrounded the RTP
version of Vorbis
> > > > I really don't know which way to go with that. Any
suggestions?
> > > >
> > > > Linus Walleij
> >
> >
--- >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
'theora-dev-request@xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is
needed.
Unsubscribe messages sent to the list will be ignored/filtered.

At 04:03 PM 19/11/2002 +1100, Silvia.Pfeiffer@csiro.au
wrote:
>Hmm, I know what you mean. I've actually taken these sentences out of
>Monty's specifications. The first one refers to the page size of max.
>~64 kByte and the second one to the logical subdivision of packets into
>segments of 255 Byte size. I'm not sure how to rephrase that properly
>though. Maybe it's good as it stands: that it is recommended to have 4-8
>kByte size, but expected more commonly to be 50-200 Byte? I'm not
>experienced enough to make anyone of these statements. Monty, your take?
>
Ogg is designed for an nominal page size of 4-8 kB (libogg will use
pages of just over 4 kB usually), so the 4-8 kB figure refers to
nominal _page_ size. For latency reasons, constraining this to a
reasonably low figure (rather than the 64 kB maximum, or close to that)
is reasonably important, so this should stay (after being corrected,
of course). 

The packet size figure as given (50-200 bytes) is also correct, but not
as important, since this is a design justification rather than a
recommendation for usage.

Michael

--- >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
'theora-dev-request@xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is
needed.
Unsubscribe messages sent to the list will be ignored/filtered.