It would be useful to have guidance here, because my perception differs. If
we are talking about the src= document of a namespace, I thought the WG had
been extremely careful not to specify what you might find in it. It could
be a DTD, a chunk of DRF or some XML-data. There seems no reason why it
shouldn't also be a BNF, chunks of java or an XSchema. So we need the
following:
the ns= URI which tells people that the namespace is identified with the
XML-DEV XSchema effort.
the src= document which has the precise machine-readable specification
This might be a DTD, might be the equivalent XSchema, or might even be both
plus some documentation and ways of sorting out which to use.
>a WG project and we are not using WG 'due process' - just public discussion
>leading to (hopefully) a rough consensus on a specification. This is indeed
>an experimental specification, but it is a specification we hope to complete
>by the end of June, _complete with namespaces_.
>
>It would be nice if some of this proposal survives as part of an eventual
W3C
>specification, but it may well not. I'm focused on producing a workable
>specification, not contemplating its eventual demise. For a discussion
>XSchema/W3C WG issues, see
>http://www.lists.ic.ac.uk/hypermail/xml-dev/9805/0454.html .
IMO the W3C process expects (and has sometimes encouraged) experimentation
and individual members of it will keep an eye on what we are doing. But
XSchema bears the same relationship to W3C as does any voluntarily produced
language effort (cf tcl, Perl, etc. in their beginnings) - i.e. it may be a
useful thing to interoperate with at some time.
Peter Murray-Rust, Director Virtual School of Molecular Sciences, domestic
net connection
VSMS http://www.nottingham.ac.uk/vsms, Virtual Hyperglossary
http://www.venus.co.uk/vhg