Update: I recommend anyone interested in this check out Sam Ruby's explanation of what happened. (He's one of the HTML5 WG co-chairs.)
[These are solely my opinions, not any kind of official statement from O'Reilly Media, Inc.]
Every now and then something comes up so wrong that it's time to put aside my pleasant retirement from the web standards process and get out the high horse again.
This time it's the spin that WHATWG folks have been pushing about "Adobe's secret hold":
the latest publication of HTML5 is now blocked by Adobe, via an objection that has still not been made public
I'm alternately laughing and crying at claims that Adobe, Larry Masinter in particular, has put a "secret hold" on HTML5. It's just like the US Senate, right? Individuals and corporations insisting that their position is so important that it has to block everything else?
No, not really. The sanest complete description of what's going on appears to be here, listing Masinter's objections to the publications that he believes are out of scope based on the HTML Working Group Charter and pointing out that claims that these objections block HTML5 itself don't make sense, a point reiterated here.
Who am I to say such things?
I'm very confused to be defending Adobe and the W3C here. That's not my normal role.
One of the happiest days of my life came when I decided I had had enough of tilting at the World Wide Web Consortium (W3C)'s windmills, signed out of xml-dev, and focused on other projects. I'd worked on a site keeping track of W3C work, enjoyed promoting news that caused turmoil at the W3C, questioned the model of expert-driven rather than community-driven standards, wrote a guide to help outsiders understand the mysterious and broken workings of the W3C, and spent countless hours at odd places like the www-tag list attempting to convince the W3C of its many technical and political sins. I mostly failed, I think.
I've also been a harsh critic of efforts to expand the importance of Flash, and my last interactions with Adobe's Larry Masinter involved jousting over MIME types and Borges. Sam Ruby, a current chair of the W3C HTML WG, complained that I was spitting out the seeds he was trying to plant.
Despite that, I simply can't find sympathy - any sympathy - for Hickson's position. In his position as editor, he hasn't needed any "secret holds" - he just controls the specification. This is a process, after all, that's inspired more than a few CSSquirrel episodes, and unleashed the remarkably foul (NSFW) Mr. Last Week, who'd probably be more effective if he just published IRC transcripts without the dada commentary.
The most difficult aspect of W3C process, beyond the simple fact that there is a process, is that it includes stakeholders in the Web who aren't necessarily the people who want to follow a given leader's direction. They're not exclusively implementers. They have their own opinions, motives, interests, and needs. They're supposed to ask difficult questions about specifications under development, ask hard questions about things like whether a project is still operating within its scope or not, and expect actual change to emerge in documents as a result of discussion. Friction is a key part of the process.
The W3C process hasn't worked in HTML5. Partly it's because HTML5 is managed by two separate organizations, the W3C and the WHATWG. As the WHATWG site puts it, "The W3C HTML working group and the WHATWG are working on the same specification, with the same editor." That editor, Ian Hickson, is not only editing a document, but "is acting as a spokesman for the [WHATWG] group." In practice, that has largely meant that Hickson does what he wants to do. As Jeni Tennison once put it:
So with HTML5, we have a situation where the Designer has decided not to design by consensus, and has very different priorities from the members of the Semantic Web community. The benefit is a certain level of consistency of approach with HTML5 (as much as that's possible with the legacy constraint). The cost is that the Semantic Web community, indeed any community with different priorities from the Designer, can have very little input into that design, and this causes a whole load of conflict.
Consensus, for better or worse, has been the heart of the W3C approach for a long time now, and the task of the editor on a W3C project has long been to reflect that consensus. I've complained in the past about overly determined W3C projects led by their editors, but this appears to be a whole new level of this issue. As Larry Masinter puts it in one of the "secret" messages:
My personal observation is that the current HTML5 process combines the worst elements of the IETF process and the W3C process. From the IETF, there is the chaos of an open mailing list, wide ranging comments and free participation, but without the "adult supervision" that the IETF supplies in the form of the Internet Engineering Steering Group (IESG) and the area directors. From the W3C, there is the overhead and cost of W3C activity management and staff, and a process that assumes that voting members actually have a say in the final specification, but without any actual responsibility or response to normal W3C safeguards and cross-checks.
So what now?
Ideally, I'd like to see the W3C take its consensus-based process seriously, and the WHATWG agree to abide by that.
Realistically, I just can't see either part of that happening. The W3C is too willing to bend; the WHATWG too unwilling. That seems to leave two options:
The W3C continues its wobbly HTML5 process, lending its imprimatur to a specification over which it has little real control, or
The WHATWG decides that the W3C's imprimatur isn't worth the effort of the process, drops the cooperation with the W3C, and fractures the HTML world severely.
I suspect that in the end, the WHATWG will take choice #2, likely because W3C members (like Adobe) rightly insist that the W3C behave like the organization it's claimed to be all these years. After all, no less than the Chair of the W3C HTML Working Group has called in the past for "A Bias For Action," writing:
The way that these disputes tend to get resolved in Apache is that somebody steps up to the plate and builds the darn thing as best as they can and solicits input in the form of "patches" (tangible suggestions in the form of working code). The original code provided is generally not important - it tends to get refactored away anyway into oblivion. What is important is that it focuses discussion into the form of constructive and tangible input.
That's not so far from the WHATWG approach, after all. While I do think a separation would be a disaster for the W3C's credibility as well as the WHATWG's ability to influence what Microsoft and other companies outside the WHATWG core do in its browsers, I can't help but forecast that result. I'd like to be wrong, but fear I'm not.
HTML5 will be damaged, its credibility weakened, but will still be important, one way or another.