Re: Request for clarification on HTML 5 publication status (ISSUE-19)

On Wed, 28 Nov 2007, Dan Connolly wrote:
> 
> Before we get to the business of addressing objections to publishing, we 
> have to establish a critical mass of support for publishing.

It would seem we have that -- 43 people have indicated explicitly that 
they agree that we should publish a draft, in a vote with 53 ballots cast. 
That's over 80% explicitly positive. (You yourself wrote that we had 
"considerable support for publication" [1] in response to an earlier 
survey with fewer votes asking for HTML5 to be published [2].) In fact, 
given that some of the votes were abstentions, only 11% of votes were 
negative. 43 people is also more people than most working groups ever 
have, so it's quite substantial.

[1] http://www.w3.org/2002/09/wbs/40318/wd11spec/
[2] http://www.w3.org/2002/09/wbs/40318/wd7/results#xokclose

The negative votes were as follows:

>From Roy Fielding:
|
| This draft has almost nothing to do with HTML. It is a treatise on 
| browser behavior. That is a fine standard to have, but deserves a 
| different title so that the folks who just want to implement HTML can do 
| so without any of this operational/DOM nonsense.

Roy's complaints seem to be technical issues that are easily dealt with in 
future versions (this is, after all, nothing but a FPWD). (Admittedly, 
these objections are controversial and don't really line up with our newly 
published design principles, so it is possible that the objections can 
never be dealt with in a fashion that Roy agrees with, but that, again, 
doesn't seem like a good reason to delay FPWD publication.)


>From Philip TAYLOR:
|
| No, we have not yet agreed the Design Principles. Without those, the
| specification has no formal basis for existence, and premature
| publication could jeopardise the reputation of the Working Group.

Since few W3C specs have design principles written formally before they 
are published, it seems that Philip's argument doesn't really apply in the 
W3C world. It also seems that the working group's reputation has _already_ 
been jeopardised by our lack of publication, so that the issue of our 
reputation becomes an academic one at this point.

(Note that Philip also objected to the publication of the design 
principles, but the chairs overruled him there.)



>From Gregory Rosmaita:
|
| i agree with the comments of Roy Fielding that, as currently
| drafted, the HTML5 working draft is really quote a treatise on
| browser behavior unquote.

It isn't clear why this would be an objection to publish an FPWD.


| i am also concerned by other respondents' suggestion that the WHAT
| WG's recently published (26 October 2007) quote stable version
| unquote of the spec [3] be adopted wholesale by the HTML WG. while
| the WHAT WG is free to do what it wants, i -- and others -- have
| repeatedly asked that the W3C draft reflect the input of the HTML
| WG, not input to the WHAT WG -- if WHAT WG members want to comment
| on the W3C Draft, they should feel free to do so, but it is madness
| to have 2 competing drafts, especially one that is revised outside
| of W3C space and simply ported to W3C space based on revisions to
| which the HTML WG has little knowledge. the WHAT WG's draft does NOT
| reflect the consensus of the HTML WG and should not be imposed on
| the HTML WG.  therefore, i am objecting to the release of the HTML5
| Working Draft UNTIL such time as the editors and chairs acknowledge
| that there is only 1 draft of HTML5 that reflects (or is supposed to
| reflect) the work of the HTML WG so far. until the parallel
| development of the HTML5 draft is considered just another suggestion
| stream, rather than a call for consensus on a fait accompli, i will
| continue to vote against the release of the HTML5 working draft as a
| W3C draft, until there is but one draft from which EVERYONE, within
| and without the W3C, can work with the assurance that the issues
| they raise and the suggestions they make are based upon a single
| iteration of HTML5. feedback is, indeed, critical, but only if that
| feedback is shared with the HTML WG and informs the content of the
| document identified as a W3C working draft.
|
| until the issue of the competing drafts is resolved, i cannot
| support release of the HTML5 draft. i am voting "no" rather than
| "formally objecting" because i trust the chairs and the Hypertext
| Coordination Group/Team to clarify the issue of parallel tracking
| and duplication of efforts once and for all before the end of the
| year, and i am not convinced that the best way to achieve that end
| is to formally object to the release of the W3C working draft of
| HTML5 -- i am convinced, however, that if the status quo persists,
| the HTML WG will never reach consensus, and that the number of
| formal objections lodged in straw polls and in posts to the chairs,
| editors, and list will grow exponentially.

The one decision that the HTML WG _has_ made up to this point was to adopt 
the WHAT WG draft wholesale, so it seems that this objection is ill-timed. 
It's not clear what the alternative would be at this point. If the chairs 
agree with Gregory on this point, then it would be good to know soon, 
since it fundamentally affects how the working group proceeds. If the 
chairs do not agree with Gregory on this point, then it doesn't affect 
FPWD publication.

[3] http://www.whatwg.org/specs/web-apps/2007-10-26/multipage/


>From Laura Carlson:
|
| I agree with Gregory Rosmaita, Philip TAYLOR and Roy Fielding's 
| rationale.

>From Dominik Tomaszuk:
|
| I agree with Laura Carlson, Gregory Rosmaita, Philip TAYLOR and Roy 
| Fielding's rationale.

For the same reasons as given above, these objections don't seem to be
good reasons to block FPWD publication.


>From Chris Wilson (Microsoft):
|
| We have not had the discussion yet of whether everything in the current 
| HTML5 spec is within our current charter. I believe that may have Patent 
| Policy implications.

Notwithstanding patent policy implications, it isn't clear that a decision 
on whether we should remove certain parts of the specification should be a 
prerequisite to FPWD publication.

As far as patent policy implications go, it is my understanding that in 
fact the patent policy [4] does not have any bearing on the working group 
charter, with the exception that the charter can inform what general areas 
the group will work on (which can help with preemptive patent reviews) and 
that the charter must mention that the patent policy is in effect.

Indeed, for the purposes of the patent policy, the most important step is 
the publication of a First Public Working Draft, since FPWD and LC 
publications are the only two steps that actually effect new commitments 
on the behalf of members (other than the act of joining the group that has 
already published a FPWD).

So it seems incorrect to state that not having discussed the scope of the 
first draft has patent policy implications; what has patent policy 
implications is the publication itself. If a member needs more than the 
three months after FPWD publication to perform a patent review, then the 
member should resign from the group temporarily until such time as a 
review can be complete; as far as I can see that would be acceptable and 
well within the patent policy rules.

[4] http://www.w3.org/Consortium/Patent-Policy-20040205/


> That's where Chris and I made the mistake. When I put the question on 2 
> November, I assumed that members such as Nokia and IBM and Microsoft 
> were aware of the patent policy implications of publishing current HTML 
> 5 specs under our current charter, and Chris assumed that Microsoft's 
> patent review included the immediate mode graphics stuff.

Certainly if a company needs more than the standard 3 months to review the 
spec after FPWD publication, we shouldn't prevent them from leaving the 
group and taking the time to complete this review. However, I do not 
understand how this would be a reason for blocking FPWD publication -- 
indeed, as far as I can tell from my reading of the patent policy, it 
would be a reason to expedite it.


> Of course, it would be easier to publish the spec right
> away if the spec took a much more conservative position on
> issues such as videoaudio, immediate-mode-graphics, and
> offline-applications-sql.
>  http://www.w3.org/html/wg/tracker/products/2

I think those areas are possibly the most important areas of the 
specification. If we removed them, we would be making ourselves 
irrelevant, as the community would instead refer to the WHATWG HTML5 draft 
for the definitions of those features, which seems highly counter- 
productive (and is indeed exactly what Gregory, Laura, and Dominik 
mentioned would be a problem).

It would also be dangerous to remove these from a FPWD publication if we 
had the intent of reintroducing them before LC publication, since it would 
mean we would be unaware of whether a member company intended to claim 
patent exclusions until at most 60 days after LC publication, several 
years from now, and possibly months after those features have been 
implemented and shipped by competing companies. This would in fact 
undermine the competitive playing field that we are trying to foster with 
the patent policy in the first place.


> But perhaps those are worth the wait.

I do not understand why they would cause a wait.


> I trust we'll be ready before too much longer.

We have a critical mass of members willing to publish. We have good 
reasons to publish sooner rather than later. We have a well-defined path 
for gracefully handling members who will need extra time for their 
patent-policy-triggered patent review process. We have only a small number 
of objections to publication, none of which, arguably, are good reasons to 
block FPWD publication (and most of which are in a similar vain to those 
which were overruled in the recent publication of the design principles 
document). We also have a well-developed specification which is ready for 
publication as a FPWD and which is more mature than most FPWDs.

I would like to propose that we publish the current specification as a 
FPWD on the W3C site before the new year's W3C publication moratorium. If 
this is not possible I would like to see a clear list of requirements 
which would need to be met to publish the current specification as a FPWD, 
as well as a detailed and binding timetable that the chairs believe is 
realistic for the publication of the current specification as a FPWD.

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

Received on Thursday, 29 November 2007 09:46:49 UTC