xml-dev - RE: DOMs and data models

[Kevin Williams <Kevin.Williams@ultraprise.com>]

Paul,

> I've heard a couple of DOM-developers say that now (here and in
> private). Perhaps we can call a moritorium on adding features to the DOM
> *until the underlying data models* "exposed" by the features have been
> published through a W3C REC. How about it?

I think this would be a great idea. I think that the W3C was trying to do
this to a certain extent with Infoset, but failed because the Infoset is too
general to be of any real use. Again, I think Infoset was designed to be all
things to all people, allowing non-tree processors like SAX to "conform" to
the W3C. Another recommendation that bridged the gap between the descriptive
model provided by the DOM and the APIs described by the DOM, XPath,
XPointer, and the other tree-based W3C recommendations would have immediate
benefit. Specifically, we would have a concrete model to hold the DOM,
XPath, and so on to. This would ensure the maximum possible interoperability
between these technologies. Additionally, I think this would make XML much
easier to understand by disambiguating the data model and the access
mechanisms. Adding a recommendation, rather than replacing the Infoset,
makes the most sense to me because I still think the W3C needs to describe
the information to be returned by non-tree processors.

- Kevin

Kevin Williams (kwilliams@ultraprise.com)
Co-author, _Professional XML_ (Wrox Press)
http://www.wrox.com/Consumer/Store/Details.asp?ISBN=1861003110