Re: Normative status of author-only view of the HTML5 specification

Hi Tab,

Posting back on-list as i believe your clarification and my own is necessary to provide the context of my initial comments and further any discussion.

I'm aware of the history of the WHATWG/W3C, i didn't live it, but i know the split and subsequent collaboration. What you have highlighted which i was unaware of is the continuing (and expanding!?!) divide between the two organisations and especially your comments on 4 browser manufacturers using the WHATWG as reference. This is unsurprising though as it was these companies who set up the WHATWG and where their vested interested lie.

However, the issue i have with WHATWG is precisely that. It only represents these browser manufacturers and with the notable exception of Microsoft. Whether it is liked or not Microsoft represents the requirements of the consumers and businesses who choose to use their software. It is disingenuous to suggest that they operate only under political motives and have brought nothing of value to HTML. For the manufacturer of the most popular browser to be unrepresented undermines any potential authority of the WHATWG. This is exactly why the W3C is setup the way it is - to enable universal collaboration and agreement in the face of seemingly irreconcilable differences. 

To draw an analogy with the UN, believe me, it is easy here and no-one's going to die. Politics always has at least two sides and you can't just dismiss the other party because you think you don't need them, it's about producing a better world for everyone. Don't think that WHATWG or it's members don't have political motives, everyone does it's just called self-interest. That 4 manufacturers have reconciled their differences is fantastic, akin to NATO, but don't think that they will remain in unison forever, and i can guarantee if Microsoft and their browsers were to disappear tomorrow the WHATWG members would find their differences again and require the same consensus processes which exist in the W3C.

The next issue i have with the WHATWG is that it only represents browsers. Despite common thought, HTML is not just a browser technology and as you noted the WHATWG was formed out of specific interest in the continuation of HTML predominantly from client-side authoring perspective. As an implementor of server-side HTML (yes we exist) please understand that there are different requirements and that these appear unrepresented in the WHATWG. There are many other businesses and individuals who use HTML within different contexts and their requirements are paramount in producing a HTML solution which is to be served and consumed as well as presented. 

I would like to pay homage to the WHATWG, of which i proudly consider myself a member, and to Ian Hickson and everyone who has worked so hard in getting HTML5 to where it is today. If it were not for their sterling hard work we would collectively be years behind where we are now. However, HTML did not start with version 5 and to dismiss the organization which has given it home and nurtured it through the most demanding period of life is to discredit their enlightened and innovative stewardship which is the only reason we even have HTML at all.

Sincerely,
Cameron Jones

On 10/06/2011, at 11:30 PM, Tab Atkins Jr. wrote:

> [off-list]
> [Also, this may sound abrasive.  Please take it in the spirit of
> helpfulness it is intended.]
> 
> On Fri, Jun 10, 2011 at 2:24 PM, Cameron Heavon-Jones
> <cmhjones@gmail.com> wrote:
>> Personally, i refer and work from the official specification as it stands, w3c, because i know that it is the spec which will be implemented and, while it is still in flux, it is the one i can rely on the most to be the closest to what will eventually become standard in the years to come.
> [snip]
>> The W3C, with all it's rules and procedure, produces the specification which *can* be implemented. Without the structure, accountability and formal processes it would be impossible to achieve anything even remotely close to consensus which is the whole point of a specification.
> 
> This is a reasonable position if you're not an implementor and don't
> know any of the history behind the WHATWG/W3C split.  Let me fill in
> some details for you.
> 
> After HTML4, the W3C decided it didn't want to work on HTML anymore,
> and switched its focus to XML and XHTML.  The browser vendors in the
> W3C complained about this but were ignored, because there are a lot of
> non-browser companies in the W3C who care about XML but didn't (at the
> time) care about HTML at all, and they were able to outvote the
> browsers.
> 
> So, the browsers took their ball and went home.  They organized the
> WHATWG as an informal organization to discuss the future of HTML in,
> and Hixie began writing what would eventually become HTML5.
> 
> Several years later, the browsers are still paying attention to
> Hixie's spec, and nobody cares about the XHTML2 effort in the W3C.
> W3C management finally, grudgingly, switches gears back to shit that
> the relevant people (web authors and browser vendors) care about,
> shuts down the XHTML2 WG, and asks Hixie to come develop HTML5 in the
> HTMLWG again.
> 
> Before this happened, browsers were happy to implement the WHATWG
> spec.  After it happened, browsers were *still* happy to implement the
> WHATWG spec, because for a long time it and the HTMLWG version were
> identical.  Then, gradually, differences appeared.  The first few
> decisions were to remove some stuff from HTML for political reasons
> (like Microdata, which was removed because the W3C wanted to play nice
> with its RDF working group).  The WHATWG spec and the HTMLWG spec were
> still consistent at this point, just the latter was a subset of the
> former.  More recently, substantive differences have appeared, so the
> two specs no longer have a pure superset/subset relationship, also
> usually due to political reasons (for example, a lot of the recent
> accessibility stuff is a direct result of the fact that the W3C has an
> A11Y WG filled with several loud clueless people).
> 
> I work for a browser vendor (Google).  I personally interact with devs
> from all five of the major browser companies on a daily basis.  I can
> tell you, with absolute certainty, that Chrome, Safari, Firefox, and
> Opera are implemented almost entirely by looking at the WHATWG spec,
> and will continue to be so for the forseeable future (the only
> exceptions are that, occasionally, someone new to one of the teams
> will use the W3C version until they run into one of the differences
> and are told which spec is the "real" one).  Microsoft is a little
> weirder, because they go out of their way to appear
> "standards-friendly" by publicly following the W3C spec, but several
> of the devs I know personally on the IE team actually work from the
> WHATWG spec.
> 
> So, now you know the truth.  The WHATWG spec is the more complete,
> mature spec of the two.  It's not riddled with stupid political
> compromises.  Actual browser vendors follow it instead of the W3C
> copy.  This situation shows no signs of changing, as the W3C spec is
> now regularly getting worse, as new features added to the WHATWG spec
> aren't backported to the W3C copy or are actively blocked (because
> HTML5 is supposedly in "Last Call", which means it should freeze for
> the next ten years while we iron things out), and new decisions either
> remove useful text or insert bad text.  The WHATWG (specifically,
> Hixie) has done more for interop on the web than any other single
> organization, including the W3C.
> 
> Now, in the future, you can act with better knowledge and know what's
> right.  ^_^
> 
> ~TJ

Received on Saturday, 11 June 2011 12:01:17 UTC