Re: An HTML language specification vs. a browser specification

Hi Jonas,
On Nov 16, 2008, at 8:06 PM, Jonas Sicking wrote:

>
> Robert J Burns wrote:
>> HI Boris,
>> On Nov 16, 2008, at 1:39 PM, Boris Zbarsky wrote:
>>>
>>>> The key thing is that we would be able to make the draft(s) serve  
>>>> the needs of users and authors and not only a few browser  
>>>> implementors.
>>>
>>> Good luck getting consensus on any of that, then.  Or better yet,  
>>> implementations...
>>>
>>> Note that the draft(s) do need to serve all three constituencies,  
>>> but they also need to be implementable.  The other option, of  
>>> course, is to retrace the steps of XHTML2.
>> Again, you're creating false dichotomies. I never once suggested  
>> that the parsing should be done in a way that in not implementable  
>> or doesn't serve the implementor constituency What I'm saying is  
>> that parsing should:
>> 1) be compatible with existing content
>> 2) improved to allow for forward migration
>> 3) abstracted from the HTML vocabulary (markup vocabulary and DOM)  
>> and the browser behavior (what browsers do with the objects created  
>> according to the specified vocabulary once parsed).
>
> Excellent! We all agree. If you look at the HTML5 spec that is done.  
> There are separate chapters in the HTML5 spec for the vocabulary and  
> the parsing.

There are separate chapters, yes. However those are the weakest parts  
of the draft right now. Separating them out and finding more  
specialized editors for those parts would go a long way toward  
improving the draft and getting it out the door quicker.

>
>
> So if all that the TAG is requesting then I think we are done.
>
>> The presumption you're conveying is that these first two cannot be  
>> done simultaneously, and that's simply not true
>
> I'm not sure who is claiming that. In any case they are wrong, the  
> HTML5 spec is a clear proof of that.

The point is that the HTML5 draft, as it currently stands, does not go  
far enough to achieve these goals.

Take care,
Rob