Re: Moving on (was Re: URIs quack like a duck)

> The fact that there be a definitive schema for XHTML would not
> stop you doing this.
On the contrary, it would help, You'd want to refer to it rather
than redo the work. (see we do agree sometimes:-)

> Are you at all interested in a schema for MathML which explains
> where you can put MathML in an XHTML document by cross-reference
> to the XHTML schema?

I am interested yes, but...
The way I'd see it working with schema is the way it works with DTD
that the mathml schema itself does not refer to XHTML. (No other part
of the mathml spec explictly relies on HTML) but that one of the
most important uses of this schema would be to make a 'larger' schema
which includes the XHTML one and the MathML one and then specifies
where MathML can be included in HTML. (But I'd need to re-read the
schema draft again to check exactly what facilities are offered).

But having done that, I'd have an XHTML+MathML schema, and the top
level element in my documents would be <html> so I'd want to keep
xmlns pointing at the XHTML namespace (because that's the html element
in question) but use schemaLocation to point to my MathML+Html schema.

I definitely don't want to point to the combined schema with xmlns as
then (for example) mozilla won't recognise any of the html elements as
being html, because they wouldn't be (x)html.

> Facet is your word? 
No, yours:-)

> Message-ID: <041801bfcd0b$62324820$a60a1712@col.w3.org>
> From: "Tim Berners-Lee" <timbl@w3.org>
> A schema of course represents on facet of a namspace, and an RDF
> schema another.


> Would you not consider a schema language to be a way to
> define some things about a namespace?

No. Not at all. A namespace is just a set of names.

It's got about as much structure as saying "all XML names beginning
with X" Given one or more of these sets of names, you can take a
(usually finite) subset of those names and use them as tokens in a
language that you may define the syntax of with schema, and the
semantics of with rdf or the prose text of a recommendation,
or implies semantics by implementation in a browser, or whatever
other features you want.


David

Received on Tuesday, 6 June 2000 12:56:00 UTC