Re: [whatwg] Answering the question... (timing of table headers issue)

Hi Maciej,

On Aug 24, 2007, at 2:11 PM, Maciej Stachowiak wrote:

> Your request amounts to asking that browser implementors completely  
> ignore the current HTML5 draft in the course of ongoing  
> development. Sorry, that is not a reasonable request. I decline.

It doesn't call for completely ignoring the HTMl5 draft.  The  
experimental work to add the VIDEO or AUDIO element is one thing.  
Though we haven't decided whether those will make it into the final  
recommendation those are being pursued with the understanding that  
they are experimental or perhaps non-standard features. It is a  
completely different matter to claim that a feature in the current  
recommendation (HTML 4.01, XHTML 1.0 and XHTML 1.1) should not be  
properly implemented because a few members of the HTML WG don't want  
it. The capabilities <input usemap> provides are targeted, in part,  
at disabled users. If WebKit could improve its support for it, what  
would be the harm?

> I will add also that I personally don't have a strong opinion on  
> whether <input usemap> should be supported or not, or what it  
> should do. I never gave an opinion on any mailing list or IRC  
> channel. The reason the issue came up at all is that another html  
> working group member (Robert Burns) noticed <input usemap> being  
> removed from the draft, and thought the appropriate response was to  
> file a bug asking for it to be implemented.

That is a completely unfair characterization. The discussion on this  
list revealed to me that the feature did not work as expected in any  
of the browsers. As someone who makes use of WebKit in my own  
application, I wanted to make sure WebKit had full support for it. It  
wasn't because I shared some delusional fantasy that we were every  
going to drop this feature.

> He omitted the fact that it had been removed from the draft and the  
> reasons given for its removal.

I omitted irrelevant facts like that because they shouldn't be  
brought up on a bug tracking system. Again, if you want to file bugs  
against HTML5 as experimental features (like VIDEO or AUDIO). But  
where HTML 4.01 says one thing and a few members of our group want  
HTML5 to say another thing, it is HTML 4.01 that's authoritative  
here. When we reach candidate recommendation stage, then I'd say you  
would have a little more justification in saying what you said.  
However even then I think the phrase candidate recommendation should  
be in there somewhere.  As it is, you could have said that a few  
members of the WG don't want <input usemap> in the recommendation,  
but at that point why bother saying anything on the bug tracker at all.

> Our bug tracker exists to report problems, not for people to ride  
> their hobby horses. I don't know how you even found the bug since  
> it's uncommon for people to browse our bug tracker. If I were  
> conspiracy-inclined, I would suspect you and Robert of being in a  
> secret cabal to file provocative bug reports just so you can  
> complain about the response.

Yes, Maciej I told John and I told the co-chairs about this  
situation, because think it needs to stop.

> Can we end this silliness now? You have spent more time complaining  
> about people expressing their opinions in outside forums than in  
> actually providing research or close review that would help improve  
> the spec. I have tried to give you the benefit of the doubt but it  
> is starting to seem like your only goal here is to create conflict.

John is not complaining about people expressing their opinions. John  
is complaining about members of this WG mis-representing the work of  
this WG on bug trackers and other public forums. That's a huge  
difference. Not only that but it's occurring around yet another  
accessibility issue.

Yet on the IRC logs and the blogs, we keep hearing it's all a  
misunderstanding. Well why target an accessibility feature other than  
because its an accessibility feature. If Safari and IE implement  
<input usemap> in a way that's useable, why not simply fix the minor  
bug that makes it even more usable. On the other hand, Mozilla and  
Opera break it in a worse way: it doesn't even function in a way that  
an author would want to use it (I've filed a bug with Mozilla for  
that as well). Yet for authors who target IE and Safari (you should  
be happy about those authors Maciej), they can make use of <input  
usemap>. It provides the accessibility benefits that HTML 4.01  
sought. There are other visible benefits for authors that are missing  
in WebKit and in IE. Nevertheless it works in those browsers and it  
provides accessibility benefits.

So the WebKit bugtracker should be for tracking bugs. WebKit has a  
bug in relation to its implementation of the HTML4.01, XHTML  1 and   
XHTML 1.1 <input usemap> functionality. Why wouldn't I report that  
bug? Why would anyone try to dismiss that bug? It would instill a lot  
more faith in the rest of the WG members if those proposing the  
removal of <input usemap> would first try to understand the feature  
before dismissing it. I don't think there are any grounds for not  
including this feature.

Take care,
Rob

Received on Friday, 24 August 2007 19:38:28 UTC