Re: An HTML language specification

• From: Boris Zbarsky <bzbarsky@MIT.EDU>
• Date: Thu, 20 Nov 2008 10:39:12 -0500
• To: public-html@w3.org
• Message-ID: <492584A0.6070108@mit.edu>

Mark Baker wrote:
> I just quickly ran through the SVG 1.1 spec and only saw a handful of
> instances where the definition of the vocabulary was in terms of the
> DOM.

I should note that SVG 1.1 fails to define the language at the level of 
precision we're aiming for (well, some of us are aiming for) with HTML5. 
  My experience as an implementor is that it's quite badly underdefined, 
in fact.

Maybe this could be tightened up without reference to some abstract 
containment structure, but maybe not.

A related question here.  Is the problem the use of the DOM per se, or 
the general use of a tree structure for defining containment 
relationships?  That is, is the problem the specific model, or the 
general type of model used?

-Boris

Received on Thursday, 20 November 2008 15:39:54 UTC