Re: Request for review: XOP

At 3:47 PM +0200 5/1/04, Robin Berjon wrote:

>DOM Level 2 HTML also does not deal with XML. It's the DOM all the same.
>  SMIL and SVG DOMs have non-XML features, yet they integrate well, 
>strongly, and usefully with the Core DOM. If that's pollution, it 
>would seem we are much polluted already, and people are liking it.

Those work by subclassing the existing DOM interfaces. I read this 
proposal to be requesting modifications to the core of DOM, a very 
different thing. If that's not the case, and they just want to 
subclass, then why do they need support for the DOM Core group 
instead of doing it themselves as the SVG and SMIL groups have done?

However, SMIL and SVG are XML. XOP is not. I really question whether 
trying to present XOP as XML is the right path, especially since they 
seem to find it onerous to use the existing APIs for processing XML.

HTML is not XML (though XHTML is) and this is partly responsible for 
the mess that is DOM. And people do not like it. In fact, I would say 
that web developers who have to deal with HTML DOMs are even less 
fond of the DOM than XML developers are.

>Amongst the clever things DOM 3 Core adds is DOM Features[0]. If one 
>were to develop a supplemental DOM chapter that would add support 
>for the direct retrieval of binary data, and define a feature string 
>for it, then with no modification to the Core DOM implementations 
>could nevertheless support XOP, and make it cleanly available to 
>users.

Perhaps. I suspect that would not satisfy the requestors though. In 
particular I think they would object to having to create element 
nodes for their xop:include elements.
-- 

   Elliotte Rusty Harold
   elharo@metalab.unc.edu
   Effective XML (Addison-Wesley, 2003)
   http://www.cafeconleche.org/books/effectivexml
   http://www.amazon.com/exec/obidos/ISBN%3D0321150406/ref%3Dnosim/cafeaulaitA

Received on Saturday, 1 May 2004 20:59:09 UTC