Picking up a few elements of the thread:
> (Actually I think the SES URIs refer to the element declarations
rather than the complexType definitions, but I think the intent is to
refer to the complexType?)
No - the intent was to identify the element declarations. Elements are
what appear in instance documents (as here). Types are just supporting
apparatus.
> Also, I don't see any issues with using URIs that are [XML namespace]
+
> "#" + [localname] for XSD types, as that URI is not actually used by
the
> XML schema etc. (i.e. no URI collision risked).
There is an XML Schema Component Designator syntax
http://www.w3.org/TR/xmlschema-ref/ which allows more finegrained
identifiers for all components in an XML schema. Unfortunately it seems
to have stalled at working draft :-(
> I'd also like to add that not all XSD types can be mapped to RDF
> datatypes. For example, the XSD type xmls:QName depends on the XML
> context in which it appears - making the mapping from lexical space to
> value space context-dependent, which is not allowed for RDF
> datatypes....
A key point here is that the proposed XML is not merely builtin types
from the XML Schema Spec part 2. It is complex types defined by a
reputable authority with an interest in the subject domain (actually
ISO/TC 211).
Simon
______
[log in to unmask] CSIRO Exploration & Mining
26 Dick Perry Avenue, Kensington WA 6151
PO Box 1130, Bentley WA 6102 AUSTRALIA
T: +61 (0)8 6436 8639 Cell: +61 (0) 403 302 672
Polycom PVX: 130.116.146.28
<http://www.csiro.au>
ABN: 41 687 119 230
|