- From: Ian Hickson <ian@hixie.ch>
- Date: Fri, 14 Dec 2007 22:10:10 +0000 (UTC)
I'd like to thank everyone for their continued polite participation in
this discussion. There has been a lot of argument about the relative
merits of coercion and attempts at assigining blame; however, I would once
again like to encourage anyone taking part in this discussion to consider
how to move forward, given the opinions of the various parties involved. I
think it is pretty clear at this point that we can't really change
everyone's opinions; we have to work, instead, on how we can address the
desires expressed, regardless of whether we agree with them or not.
Before posting anything on this matter (or any other matter), please
consider:
1. Does this help move the issue forward? If no, please omit that part of
your e-mail.
2. Has anyone already said what you are saying? If yes, please omit that
part of your e-mail.
3. Are you saying something that will just be denied, without leading us
to resolve the issue? If yes, please omit that part of your e-mail.
With that in mind, I will reply to some of the recent e-mails:
On Fri, 14 Dec 2007, Sanghyeon Seo wrote:
>
> I agree that most people here are not suggesting requiring H.264, but
> Nokia cleearly did, and I believe that's why many people are expressing
> concerns. I don't know how to interpret this quote in any other way.
>
> "Considering our requirements, we believe the widespread use of
> technically competitive, but not necessarily "free" open standards, such
> as H.264 for video and AAC for audio, would serve the community best."
I would like to assure everyone that H.264 can't be on the table unless
the licensing requirements on H.264 change.
On Fri, 14 Dec 2007, Shannon wrote:
> >
> > As far as I can tell, there are no satisfactory codecs today. If we
> > are to make progress, we need to change the landscape. There are
> > various ways to do this, for example: [...]
>
> Your suggestions are impractical and you are smart enough to know that.
They are not easy ways forward, I agree.
How would _you_ recommend addressing Apple's requirements while still
addressing the requirements of the rest of the community?
> You claim neutrality but YOU removed the Ogg recommendation and you
> haven't answered the IMPORTANT questions. I'll re-state:
>
> 1.) Does not implementing a SHOULD recommendation make a browser
> non-complaint (as far as validation goes)?
No, but compliance isn't a goal. The goal is interoperability. Compliance
is merely a way to achieve that goal.
> 2.) What companies (if any) would abandon HTML5 based on a SHOULD
> recommendation?
I don't believe anyone said they would abandon HTML5 based on this.
> 3.) What is Google/Youtubes' official position (as the largest internet
> video provider)? I assume they are reading this list and I'm guessing
> you still work for them.
Google is a firm believer in open standards, and strongly believes that
HTML5 should be freely implementable, without license fees, by all parties
-- whether by open source projects, corporations, or any other group. We
don't have a proposal for how to solve the codec solution. We would like
to see a codec of H.264's quality, and indeed we are growing our use of
H.264, but we agree that H.264 isn't a solution for <video> on the Web.
> 4.) What prevents a third party plugin open-source from providing Ogg
> support on Safari and Nokia browsers?
Nothing (at least in principle, although the iPhone browser cannot be
augmented by user-selected applications at this time), but most users
don't install third-party products and so this is probably academic.
> 5.) Why are we waiting for ALL parties to agree when we all know they
> won't? Why can't the majority have their way in the absence of 100%
> agreement?
Depending on how you count "the majority", this might not go towards Ogg.
Would you be happy if we used something that wasn't royalty free? If not,
why should we pick something that makes Apple unhappy? (Why are you more
important than any of the other players?)
> 6.) How much compelling content is required before the draft is
> reverted.
I don't know.
> Does Wikipeadia count as compelling?
Apparently not.
On Fri, 14 Dec 2007, Shannon wrote, in response to Dave Singer:
> >
> > Ian, as editor, was asked to do this. It was a reasonable request to
> > reflect work in progress. He did not take unilateral action.
>
> Ok, not unilateral. How about 'behind closed doors?'. Why no open
> discussion BEFORE the change?
There's been plenty of discussion over the past year about this. It was
also discussed at the (open) W3C meeting in Boston, and it's being
discussed right now here.
(Actually I took unilateral action when I put in the requirement for Ogg.
It didn't represent consensus, which is why I removed it.)
> That just about says it all, doesn't it? Is this a public standard or
> not? What is this list?
The WHATWG doesn't do spec development by vote. That basically is design
by committee, and it doesn't work. Instead, I take into account all the
input, including yours and Apple's, and the input on blogs, and Slashdot,
and input sent to me in private e-mails, and I try to find solutions that
address the needs of the most people.
> You claim 100% agreement is necessary to revoke the change. If so why
> wasn't it necessary BEFORE the change?
Consensus _was_ required before, that's why I removed the Ogg requirement.
We didn't have consensus on it. (Or rather, we didn't have the potential
for interoperability.)
> The way I see it we are expected to wait for an impossible event. No new
> protocol can possibly pass the expired patents test for at least 10
> years. Are you planning to wait that long to ratify HTML5?
On a completely pragmatic note, yes, HTML5 won't be done until 2022 or so.
But that's not relevant here, I hope we will have <video> resolved long
before then. For more information about WHATWG's timetable, see our FAQ:
http://wiki.whatwg.org/wiki/FAQ
On Fri, 14 Dec 2007, Shannon wrote:
>
> > Again, a false presumption. This was discussed in the context of the
> > HTML WG at the W3C. Those doors are not closed.
>
> Really? Does that mean I can claim a seat on the board? Where is this
> discussion about a public standard made public if not here? Please
> provide a link to these open discussions and I'll concede your point
> (and join - it is public, and free right?)
You have already joined the WHATWG, and are taking part in the WHATWG's
discussions.
Instructions on joining the W3C effort are here:
http://blog.whatwg.org/w3c-restarts-html-effort
(It is also free.)
> Nokia requested, Nokia got
Actually Nokia requested much more than the change that I committed.
All I did was change the spec to say it was an open issue and that not
everybody agreed, which is just a truism in this case. I didn't make any
decisions that Nokia wanted.
> I believe the OGG recommendation IS the way forward and I believe I
> speak for others as well as myself.
Indeed true. Others believe otherwise. Your opinion is no less, or more,
valid than theirs. I have to take _everyone's_ opinions into account.
I agree that without a codec we will let non-royalty-free codecs and
one-vendor solutions win. However, picking a codec that only two vendors
agree to implement (whether that's Ogg or H.264) isn't going to stop that
either.
> I (and many others) make a reasonable request and get stonewalled.
Your request (to have Ogg in the spec) has received _more_ consideration
than Nokia's request (to have H.264 in the spec). I even actually put it
in the spec to see if we could get consensus. We couldn't.
> Then you haven't been reading Ian's previous posts. I am certain the
> subtext of his previous remarks was that HTML5 will stall if we didn't
> remove the OGG recommendation.
I assure you that the rest of HTML5 will continue irrespective of this
issue.
> I'm certain he mentioned 'major companies' being the reason for the
> change. Surely saying you won't adopt a standard because you disagree
> with an optional part is more disruptive than my questions?
Apple has only said that they won't implement Ogg, not that they won't
adopt the whole of HTML5.
On Sat, 15 Dec 2007, Shannon wrote:
> >
> > Please look back on the mailing list archives. There's been plenty of
> > discussion about this before, and it's always ended up in the same
> > loop: A group of people wanting nothing but Ogg/Theora/Vorbis, and
> > another wanting one standard that all major implementers will support.
>
> I did, and which of these approaches was finally accepted by the
> majority (and the editor)?
None. It's still an open issue.
> THESE THINGS ARE IMPOSSIBLE! THEY ARE NOT OPTIONS!
As it says in my .signature -- things that are impossible just take
longer.
There are a number of companies who have very strong financial and
fiduciary reasons to want a solution to be found here. Don't underestimate
the power of money to solve these problems. You will, however, have to
give it time if you are not able to actually solve the problem yourself.
We don't win anything more by re-adding Ogg to the spec, since all the
browsers who are willing to implement Ogg already support it.
We don't lose anything by not making a decision today. It is better to
wait and find a solution that works for all players than to rush into a
decision that we _know_ cannot achieve interoperability.
> A 'should' recommendation for Ogg was chosen because it was the most
> popular, reasonable and realistic option. It was accepted (even
> temporarily), the issue was put to sleep.
No, not at all. The "should" recommendation for Ogg was put in, by me, as
an experiment. It failed. It wasn't reasonable, nor was it realistic. It
wasn't ever accepted by even a majority of the vendors. The issue was
never put to sleep, we've been discussing it on-list and off-list at W3C
meetings, at lunch, on IRC for months.
> (Actually I DO understand, I was simply posing THE question that needs
> to be asked - ie, whose in charge here?)
I'm the editor of the spec, if that's what you mean.
On Fri, 14 Dec 2007, Jeff McAdams wrote:
>
> And that's still disingenious because there was valid (if not completely
> agreed upon) text there that was at least in a form that could have be
> released in a final spec.
If it's not agreed upon, it's not valid.
And it couldn't have been released in a final spec without making that
part of the spec academic.
> Changing the text to something that clearly has to be changed again
> serves no purpose. If the rules of the w3c and/or whatwg require that,
> then the rules are dumb and should be changed.
The principles of the WHATWG require pragmatism and honesty. Requiring Ogg
as if people can rely on that as a common codec is neither pragmatic nor
honest. That's why I changed the spec.
> More practically, if you want to reduce the heat, revert the text and
> have the discussion with the old text in place.
Lying just to remove the heat is not really ethically acceptable to me.
I'd rather have the heat.
Note that removing the Ogg requirement has had absolutely no effect on the
intentions of implementors. Opera and Mozilla still do Ogg, Apple still
doesn't.
> In short, Nokia asked for the change, and it was granted, and end-users
> are saying not just no, but hell no, unless an equally open and freely
> implementable codec is offered.
We're working to find one.
> If you want to reduce the heat and end the firestorm, revert the text
> and have the discussion with the old text in place.
I personally have no problem with the firestorm, I just would like to
encourage everyone to look for solutions that work for everyone, not just
themselves.
On Fri, 14 Dec 2007, Joseph Daniel Zukiger wrote:
>
> A publicly acknowledged standards body should build for the ages, not
> for the current whims of the fiscally fat prima-donnas we call publicly
> traded corporations.
Actually the WHATWG is concerned very much with the current whims of all
the browser vendors and Web authors. If you want a more long-term
solution, I recommend looking towards XHTML2 and RDF at the W3C.
=========================================================================
Now, on a less technical note:
Various people wrote:
>
> I am NOT the villain here. MY interest in this matter is altruistic (as
> a web developer and user). I do not work for a company with existing
> commitments to a particular format. If I keep the debate going then it's
> because the answers given are unsatisfactory.
I entirely agree that the answers are unsatisfactory. That is why it's an
open issue. However, repeatedly saying that the answers are unsatisfactory
doesn't help us go forward.
> Then what ARE you waiting on (what PRACTICAL thing I mean)?
> Understanding what? My understanding is perfect. The MPEG-LA is upset
> with the Ogg proposal. Also, when will you stop beating your wife?
> (since you brought it up). Ian has claimed compelling content could end
> this impasse. I do not believe it (any more than I believe you beat your
> wife).
Please, let's remain rational. Rhetoric like the above does not help us
make forward progress.
> And who is blocking the light? Who created the strawmen? Was it Nokia
> claiming that Ogg was 'proprietary' (conveniently ignoring the fact it
> is public-domain).
Again, please let us avoid trying to assign blame, as it is not
productive.
> OGG Theora is not the most compressed video format, nor is it the least.
> It is however in the public domain and at least equivalent to MP3 and
> XVID, which are both popular streaming formats. While submarine patents
> may one day undermine this there is no current evidence that OGG
> contains patented technology and there is plenty of legal opinion that
> it does not. Either way it is not possible to remove this risk
> altogether by maintaining the status quo or recommending (or demanding)
> any other format.
I believe these facts have all been said before. Nobody is debating their
truth. Repeating them does not help us move forward.
> There is no evidence that recommending optional OGG support will affect
> companies adoption (or not) of the rest of the HTML5 standard. We only
> have Ian's word for that and I don't believe it anyway.
I have never said that it would. Please do not mischaracterise my
statements.
> > MPEG-LA has said *absolutely nothing*.
>
> No they wouldn't. Fortunately I can read between the lines. Nokia is
> their frontman.
I would like to encourage everyone to consider not making statements like
this, as they do not tend to be productive.
> The chosen wording was acceptable to most but it supported a format that
> wasn't obviously patented by incumbents so the incumbents reversed that
> decision off-list. Save your ire for those who deserve it, I want an
> open standard just like you. Can you say the same about Nokia, Microsoft
> or (gasp,shock,horror) Apple? Can you promise me that those who removed
> the recommendation are REALLY looking for a solution when they may gain
> from a lack of one?
> Has Steve become so much afraid of the GPL? If so, why? What does the
> board of directors want that open source prevents, other than that trip
> down memory lane to the fantasy land of patronage?
> There are several brake levers on this train, and then there is the
> option to start pulling the track. I'm personally in favor of pulling
> the track, but that's just me. I have no fondness for overloaded angle
> brackets.
>
> Who started it? Their hand is closest to the brakes. The other option is
> not the brakes that this working group wants to invoke.
Do you really believe that paragraphs like the above will help make us
forward progress?
Please, when writing e-mails, do consider:
1. Does this help move the issue forward? If no, please omit that part of
your e-mail.
2. Has anyone already said what you are saying? If yes, please omit that
part of your e-mail.
3. Are you saying something that will just be denied, without leading us
to resolve the issue? If yes, please omit that part of your e-mail.
--
Ian Hickson U+1047E )\._.,--....,'``. fL
http://ln.hixie.ch/ U+263A /, _.. \ _\ ;`._ ,.
Things that are impossible just take longer. `._.-(,_..'--(,_..'`-.;.'
Received on Friday, 14 December 2007 14:10:10 UTC