[whatwg] Removal of Ogg is *preposterous*

On Dec 12, 2007 4:08 PM, Manuel Amador (Rudd-O) <rudd-o at rudd-o.com> wrote:
> El Mi? 12 Dic 2007, Robert Sayre escribi?:
> > On Dec 11, 2007 6:51 PM, David Hyatt <hyatt at apple.com> wrote:
> > > SHOULD is toothless.
> >
> > Spefications aren't laws. MUSTs are toothless as well.
> >
> > >  It carries absolutely no weight.  I don't think
> > > it's appropriate for such weak language to be in the HTML5 spec.  It
> > > should either be a MUST (which is inappropriate at this juncture for
> > > reasons that Dave Singer. Ian Hickson and myself have posted about in
> > > previous messages) or just not be mentioned at all.
> >
> > It isn't weak language. It places the blame squarely on the party who
> > fails to meet the requirement.
>
> Agreed with you, Robert.  If SHOULD carries absolutely no weight... then why
> don't we just leave the paragraph there?  Stop eluding this question.

I disagree. If somebody is implementing the spec and are looking for a
recommendation of the W3C for which video codec to use, the SHOULD has
a large effect. If there is no codec mentioned, they will go looking
for what others have implemented and start a complicated decision
process with an uncontrollable outcome.

If a SHOULD clause, i.e. a recommendation, is all we can agree on,
then it is better than not mentioning a codec at all. I was happy with
the previous formulation of the paragraph and I am happy to go through
a technical assessment of the existing codecs wrt the criteria that
Ian has written into the spec right now. I am sure we will come to the
same conclusion that we did before and Ogg Theora/Vorbis will be
written back into the spec as a recommendation, but this time around
it will be stronger because we have done an assessment of its merits
and decided that they are the only ones coming even close to
fulfilling the needs.

I have no issues with this process.

Regards,
Silvia.

Received on Wednesday, 12 December 2007 06:44:01 UTC