[whatwg] more discussion regarding codecs (Was: whatwg Digest, Vol 45, Issue 16)

[Ian Hickson]

On Tue, 11 Dec 2007, bofh wrote:
> > >
> > > The whole point of the change was to make the point that we need 
> > > something that will not screw you. Ogg isn't a solution, as it won't 
> > > be implemented by Apple and Microsoft. If we require Ogg, then what 
> > > will happen is the big players will support something else, then 
> > > that will become the de-facto standard, and you will get screwed. 
> > > What we _want_ is for everyone to support the same codec. We don't 
> > > get that by having a SHOULD-level requirement for Ogg.
> 
> And we should implement OOXML support too right, just to make Microsoft 
> happy?  I don't think that's the right way to do things.  As someone 
> who's trained as a civil engineer, we were taught there was a right way 
> and wrong way to do things.  The most correct way is to follow the spec.  
> But this has the premise that the spec is not screwed up.  If you give 
> us a screwed up spec, we're doomed.

I think you may have misread what I wrote -- there's no plan to only make 
Microsoft happy. We have to find a solution that works for _everyone_, not 
just the open source community, or just Microsoft, or just the small 
companies, or just the big companies.


> This whole issue about submarine patents is bullshit.

As I noted in an earlier e-mail, it doesn't really matter what the reason 
is for a vendor not wanting to support a particular codec; just the fact 
that they won't implement it is enough to prevent interoperability. Now, 
if we are to find an actual solution, then we have to find the reasons for 
the disagreements, and there it behooves us to assume people are working 
in good faith, since otherwise it'll be very difficult to actually figure 
out how to resolve the issues raised.


> The reason is this:  If such a patent exists, Apple/Nokia/Microsoft/et 
> all would have found it, and waved it in your face.  After all, they've 
> had months to go look for it.

Actually, for legal reasons that I don't truly understand but which I am 
assured are true, big companies cannot safely perform patent searches. It 
(ironically) increases their patent liability. (I'm not a lawyer but if I 
recall correctly, "knowingly" infringing a patent results in triple 
penalities.) So big companies in fact avoid doing patent searches, and 
would have no idea what patents exist.


> Exactly.  Which browsers are acid2 compliant again?  Why was that?

Apple's browser was the first to get Acid2 compliance (at least in 
pre-release builds). I don't understand the relevance.

(Disclaimer: I wrote the Acid2 test.)


> > Theora *is* the baseline for free and open video, full stop.  
> > Everything else reasonable is encumbered (at least to my knowledge).  
> > Assuming I'm right, discussion over.
> 
> For anyone to claim this is not so, they will have to pull out patents 
> to show why this is not so.  Otherwise, they should shut up about it.

That, sadly, is not how standards development works.


> You can never please everyone.  Do the drug companies like FDA 
> regulations?  Hell no.  Do we need stringent standards from FDA? Hell 
> yes.

Unfortunately for us, Web standards do not have the strength of law. Would 
it be so -- it would make our life much easier.

Browser vendors can (and do) ignore Web standards they disagree with. Web 
standards _only_ have power so long as the vendors agree with them. Just 
look at XHTML2 for an example of this. XHTML2 ignoring the feedback of 
browser vendors is why we are doing HTML5 in the first place -- it would 
be silly of us to then just ignore the feedback of browser vendors! :-)


> Exactly.  Tell Apple to show us the patents they're worried about. 
> They've had time to go look up those damned patents.

That's not how it works.


> > > I assure you that the change was made in good faith; I (sadly) 
> > > received no money for the change. I really wish I had.
> 
> Your role here is extremely critical.  Surely there's someone that you'd 
> trust, that can do the appropriate research for you, so that you can see 
> if Jeff et all are speaking the truth, or if Apple/Nokia is speaking the 
> truth (patent, legal, etc)?

There is no way we can ever guarantee that there are no covering patents. 
Whether a patent covers a technology or not really has more to do with 
what the courts say than with what the patents say. If Apple say they 
don't want to implement Ogg, then we have to find another solution.

(Similarly -- Opera, Mozilla, et al, don't want to implement H.264. So we 
have to find a solution other than H.264.)

-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'