> The problem (as I see it) is not one of including pieces of existing=
> documents, nor of structural validation. The main reason for
> namespaces is semantic inheritance. I want to write a scientific
> research paper quickly. HTML has the overall document structure and=
> components that I need; MathML has equations; CML has chemical
> formul=E6. I should be able to say that I'm using those things,
> associate stylesheets, and have my browser know that <html:a> should=
> be styled with the "a" rule from the HTML stylesheet.
It seems to me simpler to create a compound document rather than to
try to force everything into a single XML document -- you can
reference another XML document the same way that you can include a
graphic or audio sequence. Managing a lot of small objects directly
on the file system can be tricky, but it's trivial with proper tool
support (think of OLE under Windows, despite its warts)
> It should be *possible* to create a DTD to which such a document
> complies, but I am not as interested in automatic validation of a
> namespace document. The interrelational issues are, I think, too
> complex to solve; in the example above, I would need to change the
> text-containing HTML elements' content models to include chemical an=
d
> mathematical markup, and maybe allow HTML markup in MathML theorems.=
> Pushing selected information into the content models is too ugly.
Not at all -- you just need a single element type to hold references
to other XML documents. You could even (though this is disgusting)
use
<img src=3D"equation1.xml">
All the best,
David
--=20
David Megginson ak117@freenet.carleton.ca
Microstar Software Ltd. dmeggins@microstar.com
http://home.sprynet.com/sprynet/dmeggins/