Re: [codec] WG Review: Internet Wideband Audio Codec (codec)

[Jean-Marc Valin <[e-mail]>]

Quoting stephen botzko <[e-mail]>:

> I kind of like the joint body idea.
>
> One reason is that it brings the ITU codec characterization/testing
> strengths into the process.
>
> Though it might take a little longer to get going, it could save a lot of
> time at the end (IMHO).

I fully agree here. We should seek to collaborate with the ITU if possible. At
the same time, I think we should not assume that such collaboration will be
possible in the end, so the best would be to get started while at the same time
seeking to collaborate. Considering that the main strength of the ITU process
is in characterization/testing, there's a lot we can accomplish before reaching
that stage.

   Jean-Marc


> Stephen Botzko
> Polycom
>
> On Tue, Jan 12, 2010 at 12:03 PM, Jean-Marc Valin <
> [e-mail]> wrote:
>
> > Hi Adrian,
> >
> > During the last BoF in Hiroshima, there was a very useful presentation by
> > Yusuke Hiwasaki (SG16-Q10 Associate Rapporteur) about how the ITU-T works
> > (slides at: http://www.ietf.org/proceedings/76/slides/codec-2.pdf). From
> > what I understand, there are two main reasons why the ITU-T cannot take on
> > this work by itself:
> > 1) Membership isn't open like the IETF, but most importantly
> > 2) IPR/licensing issues cannot be discussed during the development period
> >
> > There were two proposed workarounds to these (see slide 15). First a focus
> > group was proposed to allow non-ITU members to discuss. Unfortunately, that
> > solution does not address the IPR issue, nor does it address the fact that
> > ITU focus groups cannot create standards in the first place. So the only
> > alternative that was left was to do a joint body with an IETF WG (similar
> > to the JVT between MPEG and ITU that led to H.264). That means we need an
> > IETF WG that can actually develop codecs to begin with.
> >
> > In general, I think it's really time to get the work going and, as Monty
> > put it, not get into meeting pre-meetings to discuss whether we will hold
> > future meetings. At this point, there is significant interest, there are
> > people willing to do the work and there are even four proposals on the
> > table. Right now, the only concern that has been expressed over this work
> > was about having one more codec that vendors would have to support. I don't
> > think that's a very strong argument considering the existing number of
> > codecs out there and especially the fact that what we are proposing here is
> > to take *four* non-standard codecs and make one standard codec out of them.
> > I can't see how that would be a bad thing.
> >
> > Cheers,
> >
> >        Jean-Marc
> >
> >
> >
> > Adrian Farrel wrote:
> > > Stefan,
> > >
> > >> until now other SDOs have failed to produce a widely distributed good
> > >> quality wideband and full-band codec that would be suitable for the
> > >> Internet - especially one that is easily distributable - even though the
> > >> necessary technology has been available for a long time. Further,
> > nothing
> > >> has substantially changed lately to make it likely that other SDOs are
> > >> now
> > >> suddenly willing to or capable of doing that.
> > >>
> > >> The proposal to make IETF CODEC development depend on other SDOs is thus
> > >> not a constructive one and should not be followed.
> > >
> > > Your logic may be flawed.
> > >
> > > Until now the IETF has failed to produce a widely distributed good
> > > quality wideband and full-band codec that would be suitable for the
> > > Internet - especially one that is easily distributable - even though the
> > > necessary technology has been available for a long time.
> > >
> > > But you don't suggest that as a reason not to do the work in the IETF.
> > >
> > > The proposed draft charter does not state that the IETF work should be
> > > gated
> > > on other SDOs nor that the IETF shall not develop a Codec. Rather, it
> > > states
> > > the value of sharing the requirements work developed in the IETF with
> > other
> > > SDOs, and it notes the benefits of listening to other SDOs if they point
> > to
> > > existing Codecs that meet or nearly meet the requirements.
> > >
> > > In the unlikely event that another SDO says "thanks for the requirements
> > we
> > > would like to develop a solution in our SDO" we will need to examine the
> > > feasibility of their proposal and how people can best work on a solution.
> > > There does not seem to be any benefit in developing two Codecs to meet
> > the
> > > same set of requirements.
> > >
> > > As to Xavier's point: I think he is right that the wording in the charter
> > > could be usefully re-ordered so that the consultation is mentioned before
> > > the determination to develop a new solution.
> > >
> > > Cheers,
> > > Adrian
> > >
> > > _______________________________________________
> > > codec mailing list
> > > [e-mail]
> > > https://www.ietf.org/mailman/listinfo/codec
> >
> > _______________________________________________
> > codec mailing list
> > [e-mail]
> > https://www.ietf.org/mailman/listinfo/codec
> >
>