W3C home > Mailing lists > Public > www-html@w3.org > May 2007

Re: Separating core elements from domain-specific

From: Philip & Le Khanh <Philip-and-LeKhanh@Royal-Tunbridge-Wells.Org>
Date: Mon, 14 May 2007 22:56:38 +0100
Message-ID: <4648DB16.1040609@Royal-Tunbridge-Wells.Org>
To: W3C HTML Mailing List <www-html@w3.org>



Lachlan Hunt wrote:

> What is the real problem you are trying to solve by restricting the core 
> of HTML to such a limited vocabulary?

Language bloat : the incorporation of elements that are
not core to the concept of HTML, but which are instead
syntactic (or semantic) sugar.  Since there is an
infinity of such elements, it will never be possible
to add sufficient for everyone.  Therefore add none,
and provide instead a clean mechanism by which additional
elements from specialised vocabularies can be added
with ease.

> You should also consider that all elements are going to continue to work 
> perfectly well in browsers, without using that, and so the only few who 
> would bother would be yourself and a few others who seem to care more 
> about theoretical semantic purity, than developing a useful and 
> practical language for everybody.

No, it is /not/ necessary to take that point of view.  Browsers
(as has been observed many many times) can continue to parse
tag soup as they presently do, but the introduction of HTML 5
allows the community to say "enough is enough : if I claim this
document is HTML 5, then process it as such, not as tag soup".
Yes, there will be browser-vendor resistance to this, since they
would actually have to do some formal parsing rather than
continue to use their present heuristics, but if a watertight case
can be made, then I believe they will listen.

> And since many authors simply wouldn't understand which one they have to 
> include, nor why it's necessary, authoring tools, CMSs, etc. would just 
> have to include them all by default; at which point, it becomes useless 
> boilerplate markup and we're no better off than we would be if we'd just 
> go without.

Yes, I cannot dismiss that risk.  Indeed, I fully expect the WHATWG,
if they find any merit in this proposal at all, to say that
<!DOCTYPE html> implies "HTML 5 + all WHATWG-supported dialects".

But that does not stop those that /do/ care writing
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 5//EN" "http://www.w3.org/TR/html5/strict.dtd">
and getting strict parsing in return, with no pre-loaded dialects.

> Perhaps resistance from yourself and a few others would reduce, but much 
> greater resistance would result from everyone else who would object to 
> your solution on the grounds of it being ineffective and not actually 
> solving any real problem.

I am more than happy to leave it to the community to decide
whether my idea has any merit whatsoever.

Philip Taylor
Received on Monday, 14 May 2007 21:56:56 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 March 2012 18:16:10 GMT