Re: RDFCore WG: Datatyping documents

Regarding both candidate RDF datatyping WD's:

I am concerned that both:

1) seem to assign the "xsd" prefix to a namespace other than
As you should know, and must realize, XML namespace names are compared by
_literal string_ comparison, and not using any sort of URI canonicalization
scheme. The "xsd" prefix has traditionally been bound to the XML Schema
namespace name, and use of this prefix bound to a _similar but not
identical_ namespace URI, simply creates confusion.

Please do not bind the "xsd" prefix to the
namespace name. Furthermore, please do not refer to this namespace name as
the XML Schema namespace, in fact the two _namespaces_ are entirely
unrelated. For those who disagree, please reference a document which states
otherwise, and normatively reference this in the WD.

> A discussion of XML Schema datatyping as relevant to RDF, and a proposal
> using it within RDF:
> typing.htm

I am concerned that this document  element names into the XML Schema
namespace. It seems to me that concepts that RDFCore introduces should be
labelled by an RDF namespace. It seems to me that the XML Schema namespace
should be reserved for XML elements and URIs introduced by this WG.

How do you expect one to find documentation for the "" and other
elements? People will naturally look to the XML Schema namespace
documentation and find themselves SOL. I believe that Dan Connolly has made
this point in a pointed fashion in the past in other situations, and I

On the other hand this draft seems to do a much better job of defining
datatypes in an independent fashion to XML Schema, yet using the same
concepts, so I suspect that simply changing how the concepts are named will
be an effective solution. Whether this can still be called "XML Schema
datatypes" will remain to be seen, but nonetheless, the solution will be
compatible with XML Schema datatypes:

i..e. just don't call it "xsd:integer" rather "rdfdt:integer"

> A second proposal for using XML Schema datatyping within RDF:
> typing.htm

This proposal appears to be more dependent on XML Schema datatypes and hence
suffers more acutely from the probem that there is currently no defined way
to assign a URI to a general XML Schema datatype (which is refered to by its
QName). Same issues as above. I fear that if you simply change the names to
"rdfdt:integer" and "rdfdt:string" there is not much enough here to say what
these datatypes actually are (beyond the fact that we all know what an
'integer' is).


Received on Sunday, 27 January 2002 14:41:10 UTC