RE: Formal Objection to One vendor, One Veto

Shelley Powers [mailto:shelleyp@burningbird.net] wrote:
>> Lachlan Hunt [mailto:lachlan.hunt@lachy.id.au] wrote:
>>> implement.  When the requirement for Vorbis and Theora was first added
>>> to the spec, and Apple objected, we looked at the situation and searched
>>> long and hard for an alternative that would address their concerns.
>> Precisely.
>
>And the alternative is?

For a video codec that can be mandated in HTML5?  As I've said before, I'm not sure there is one.  The original MPEG, perhaps?

The problem I see is that video is a patent-laden area, and there's been a for-pay industry initiative around H.264 et al; though I recognize why that is untenable for the open source community, for projects like Mozilla, and why some other vendors like Opera might just not want to pay the price, I would beg indulgence from others in understanding that just because On2 Technologies donated VP3 with an appropriate controversy does not mean that it is without risk or cost, particularly to a vendor with large revenues.  It's been stated[1] that MP3 or H.264 might also have vulnerabilities to unknown patents; though that is technically true, it is highly unlikely that a holder of such a patent would not have come out of the woodwork by now, given the obvious profitability of pressing such a patent (after all, Fovent made > $100m in two years before the JPEG patent was invalidated) and fairly unlikely that the MPEG-LA would not attempt to buy the rights.  I expect it's unlikely that such a patent pool protection would be set up for Theora, as vendors want to have to pay zero money to use it.

As always, I am not a lawyer, and this is not legal advice.

>> And if we were to proceed with the requirement, as Ian correctly surmised,
>>we'd simply have two of the leading implementation choose not to implement
>>that part of the spec (at least until the patent clock were to run out,
>>which is quite a while in the future) - which of course, does not further
>>interoperability, which is the whole point.
>
>I've never seen Microsoft reluctant to ignore specifications in the
>past, so the situation you describe is not new. Oddly enough, the web
>still continues to progress. There is something to be said for using an
>HTML specification to describe a web of the future, rather than allow it
>to be used by vendors as some kind of current day validation [1].

Let's not mud-sling.  We're all here, presumably, because we want to improve interoperability, and make an Open Web.  Open, to me, means everyone can implement, and without penalties that are specific to a specific set of vendors.  Just as H.264 would fail that test, I think Theora fails it too - due to the strong concern over unknown video technique patents.  I wish that were not so, because I actually WOULD like to mandate a codec, but I don't see a fair one to mandate.  Perhaps mandate Theora OR H.264?

>>Not quite true.  Though certainly I think EOT has a lot to recommend it
>>(not least of which its deployment across 65% of the current browser
>>market, and precisely that segment that updates least often), I've
>>consistently said for a couple of years now that we'd accept another
>>solution if it were acceptable for enabling commercial font vendors.
>
>I believe George Bush had a 65% approval rating at one time.

Indeed, for about a year, from September 11, 2001 through around November of 2002.

>How is this relevant to this discussion? It isn't, except to demonstrate
>that nothing lasts forever.

I never said it did, and you can certainly take the viewpoint that Microsoft's opinions are irrelevant, because you believe their share will eventually decrease to the point of irrelevance.  I'm not betting on that, personally.  But regardless, note that I said that share is across the market that deploys least often; I've done projections on that for web fonts specifically, and a BEST case for any solution other than EOT deploying to 90+% of the browser market is like 5 years from now.  That was all.

>>> I'm not particularly surprised that [Microsoft] haven't said they will support
>>> it, as they, like many companies, tend to keep information about future
>>> products confidential.
>>
>>Correct.
>
>No offense, but foolish, short sighted, counter-productive. Web
>developers and designers consider IE the tool of the devil. Even a hint
>that Microsoft will eventually support SVG, the Canvas element, or XHTML
>would go a long ways to generating a more favorable view of the company
>and the browser. One can't count on 65% forever.

1) No offense, but that's a short-sighted viewpoint that doesn't consider that browser vendors are, in fact, in competition, and some of them (in fact, pretty much all of them other than Mozilla, in my opinion) keep at least some elements of what they are planning to do hidden.
2) ...And I am not now, and have never been, in control of the IE disclosure policy.  Railed against it for extended periods, in fact.  Indeed, personally I would be more (though not "totally") open about planning; however, that has some significant downsides too.

>> That hasn't changed either, and I do not expect Microsoft to suddenly object to an XML syntax for HTML.
>
>So, Microsoft is committed to supporting XHTML? Cool, good to know.

I would great appreciate your not twisting my words, or my participation in this conversation would be foolhardy.  I've learned from long experience to be intentional about what I say, and that's not a rephrasing of what I said.  I made no commitment to supporting anything (and am not, in fact, in a position to make any such commitment on IE's behalf) - I stated that I do not expect Microsoft to start objecting to an XML syntax for HTML.  I've tried to be forthcoming, from the first day I joined the HTML WG, about what I think is egregious for Microsoft, even when it is an unpopular stance - and some of those issues might still be open (e.g. the lack of versioning in the DOCTYPE); however, I don't personally believe that an XML syntax for HTML is one of them, and I don't think I've heard any Microsoft person say that, in public or in private.

>I don't agree with Sam's alternative, which is people make their own
>versions of HTML5. modified to fit their own agenda, and we let the
>copies somehow duke it out at some point [3]. But the HTML WG seems to
>support the concept, and I don't want to continue being a road block.

I don't support making copies of HTML5 and letting them duke it out; I do support people making explicit proposals for what they'd like to see.  I may still disagree with those proposals, of course (e.g. I disagree with Rob Sayre's proposal to require Ogg codecs).

-Chris

[1]: http://en.wikipedia.org/wiki/Ogg_controversy_(HTML_5)

Received on Tuesday, 14 July 2009 23:46:19 UTC