Re: ready to publish "HTML5 differences from HTML4"?

On Sun, 2007-06-24 at 14:02 -0500, Robert Burns wrote:
> On Jun 23, 2007, at 1:31 PM, Dan Connolly wrote:
> 
> >
> > The differences document is nice and short; it has gotten
> > a generally positive reaction.
> 
> I think the positive reaction to the document is due to the fact that  
> we needed such a document to move the discussion along within the WG.  
> I don't think we're even close to a public release of that or any  
> document (maybe we'll be ready next month or August).

We're already behind schedule for publishing something, strictly
speaking. We were chartered 7 March; 3 months later is 7 June.
Due to the fact that we weren't exactly firing on all cylinders
on 7 March, a slip 'till the end of June or early July is
somewhat straightforward; beyond that... I can ask for an exception
to the heartbeat rule, but I would need to make a very good case.

The heartbeat rule is
 http://www.w3.org/2005/10/Process-20051014/groups.html#three-month-rule

It was cited from the "What should the HTML WG publish first?" survey,
right at the top...

"If you're not familiar with the process of Working Draft publication,
see the list of W3C working drafts, section 7.4.1 First Public Working
Draft of the Process document, and the heartbeat requirement."
 -- http://www.w3.org/2002/09/wbs/40318/wd7/results


> > I suggest this WG should publish it in its present form.
> > http://dev.w3.org/cvsweb/~checkout~/html5/html4-differences/ 
> > Overview.html
> > $Revision: 1.22 $ of $Date: 2007/06/23 15:22:05 $
> 
> I think to release that document now would be a big debacle. It would  
> create similar misunderstandings to those faced by XHTML2 when  
> everyone thought they couldn't have <img> elements in their documents  
> anymore (like HTML was going to become text only). Such  
> misconceptions will dominate the discourse about HTML5 and the public  
> will not recognize all of the positive contributions HTML5 makes.

I see the risk; being quiet is not likely to help. Better to
let the discussion start, I suggest.

> Before we can publish this or the spec, we should wait for the  
> detailed review to come in over the next few weeks.

I agree that publishing the spec should wait until the scheduled
review completes, but only presuming this WG publishes something
else to satisfy the heartbeat requirement first.

>  On the HTML4 /  
> HTML5 differences document, we need to:
> 
> 1)  settle the issues already raised on the assistive technologies  
> features that are missing in HTML5.

I think it's perhaps worthwhile to note issues that have gotten
a lot of discussion. I intended to suggest text along those lines
as soon as I saw Anne's draft. But that was 2 weeks ago, and
I haven't managed to do it. And I'm not sure it will be easy
to come to consensus on a description of the status of discussion.
So I am content to just report, factually, which features are
in the current editor's draft.

The "status of this document" section will of course note that
there are various open issues and that the WG hasn't made
decisions on any of them.

I hope you can accept publishing a document while these issues
are still open. I don't see how we can meet the heartbeat
requirement otherwise.

>  From the discussions that have  
> occurred so far, I would say the overwhelming consensus is that these  
> features should NOT be dropped in HTML5.

"overwhelming consensus" is a contradiction in terms, in W3C process.
In W3C process, if one person objects, there is not consensus. There
is no level of support sufficient to be called "consensus" as long
as objections remain.

It might be correct to say "the overwhelming majority of opinion"
or some such, but that would be based on a lack of information;
most WG members have not given their opinion. Perhaps
"the overwhelming argument that has been presented" is accurate.

Evidently, the editors are not convinced. Or perhaps they haven't
gotten around to reading all the relevant arguments. I hope
the editors respond to this issue in due course in a way that
results in consensus. Otherwise, we'll escalate to the whole
WG and make a formal decision. I'm not sure what priority
to give to that task, but I hope I can put it at a lower
priority than a decision to make our first publication.
Otherwise, we run into the heartbeat requirement and the
risk that this WG gets closed altogether.


>  If we decide to publish the  
> differences document without settling these AT issues, we need to (at  
> the very least) explain what alternative mechanisms authors will use  
> in HTML5 for the features from HTML4 that HTML5 no longer supports
> 
> 2) provide detailed explanations for the semantic changes in elements  
> such as <ht>, <i>, <b>, <strong>, and <small>.

Feel free to suggest any specific changes you want.

<rant>
Saying "we need X" as though we're a bunch
of customers who can demand service annoys me. We are the working
group. If we want work done, the most constructive thing is
to do it.
</rant>

>  In my view, the change  
> in semantics means we do not maintain the html namespace. After all,  
> a namespace is a promise that the names used in the space will always  
> mean the same thing. New names may be added. Other names may be  
> deprecated. But the names, once introduced, will always have the same  
> meaning.

I'm not so sure. The TAG looked into that and found otherwise:

"As a general rule, resources on the web can and do change. In the
absence of an explicit statement, one cannot infer that a namespace is
immutable."
 -- http://www.w3.org/2001/tag/doc/namespaceState.html

> 3)  Further explanation off the differences in the <menu> element.

> 4) for all of the dropped attributes, we need to immediately explain  
> what the alternatives authors will use (e.g., accesskey, td@scope).
> 
> 5) the 'style' attribute issue just raised will cause great concern  
> in the population we release this to. Heavy-handed moves like that  
> will create a bad name for HTML5 before we can even get some  
> consensus amongst ourselves.
> 
> In summary, I don't think we should rush the public release on this  
> or any of the deliverables.

You make a reasonably good point. I'll continue to monitor the
level of consensus around this publication, but keep in mind
that as chair I'm obliged to follow the heartbeat rule
and push for publishing something soon.

>  It will only hurt the adoption of HTML5.  
> I would imagine we might be ready for an initial release by the end  
> of July or early August. However, we need to address the many issues  
> raised in the WG: particularly those that relate to backwards  
> compatibility. I don't think it will surprise anyone to see  
> presentational attributes and elements dropped in HTML5.. It will  
> surprise many people to see semantic attributes and elements dropped.  
> This does not at all meet the goals of backwards compatibility. We  
> need to either step up to that goal or carefully explain why semantic  
> facilities are being dropped (ie.e., backwards compatibility is not  
> being maintained in these areas). in HTML5. Keep in mind the  
> population will not view "seldom used" as a valid explanation. These  
> features were added to HTML with the understanding that they would be  
> seldom used. They were put their for those that needed their use. In  
> other words we do not have to cater the language to the least common  
> denominator.
> 
> Take care,
> Rob
> 
> >
> > Before I put the question formally, I'd like to have a few
> > WG members read the whole thing and tell us whether they
> > concur with my suggestion.
> >
> > In the "What should we publish first?" survey, the following
> > people expressed interest in publishing a differences document...
> > Would you please take a look and let us know what you think?
> >
> > Danny Liang
> > Robert Burns
> > Ben Boyle
> > Chris Veenboer
> > Murray Maloney
> > Henrik Dvergsdal
> > Charles McCathieNevile
> >
> > Anne, feel free to incorporate smallish suggestions as they
> > arise, keeping a reasonably careful change log. If this thread
> > yields large changes that you think are worth making, we'll probably
> > need to slow down a bit.
> >
> > Karl, I'd like you to give some thought to the Status section
> > and send me/us a draft.
> >
> > -- 
> > Dan Connolly, W3C http://www.w3.org/People/Connolly/
> >
> >
> >
-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/

Received on Sunday, 24 June 2007 21:31:36 UTC