W3C home > Mailing lists > Public > www-rdf-interest@w3.org > November 1999

Re: Cambridge Communique

From: Arnold deVos <adv@langdale.com.au>
Date: Thu, 11 Nov 1999 10:02:14 +1100
Message-ID: <003201bf2bcf$a0c73e20$ef01a8c0@lakes.com.au>
To: <caro@Adobe.COM>, <www-rdf-interest@w3.org>

----- Original Message -----
From: Perry A. Caro <caro@Adobe.COM>
> What I'm sure most of us developers want to know is:
>
> * How do we decide which one to use (RDFSchema vs. XSchema)?
>

This was a controversial topic in the power utility area where we have used
RDF schema for power system models. See:

http://www.omg.org/cgi-bin/doc?utility/99-10-02

We propose to transfer data governed by this schema in both RDF syntax and
through CORBA interfaces. For the CORBA mapping, see:

http://www.omg.org/cgi-bin/doc?utility/99-10-01

My understanding is the an RDF schema is a specification for data (like a
database schema) while an XML schema is specification for a language syntax.
The two things are similar but distinct.

Having an RDF schema did not by itself give us an XML language for our data.
We also have to adopt RDF syntax for the data.  The two decisisions are
separate.

It was relatively easy to translate our UML schema to RDF-schema.  At that
point, adopting RDF-syntax for the data was then much quicker than inventing
our own syntax.  But in future we could adopt some other mapping from RDF
schema to XML syntax.   (Perhaps yielding a syntax more easily massaged by
XSLT.)

We can have mappings to other (non-XML) technologies too.  As I mentioned,
we have a mapping from RDF schema to CORBA data structures.

> [...snip...] RDFSchema lacks some
> obviously useful features, like concrete syntax for specifying whether a
> value is required or optional, or read-only in a named context, etc.  Is
it
> because RDFSchema is waiting until XSchema defines those concepts, and
then
> will incorporate equivalents in a future extension?
>

Yes RDF schema lacks these things. We added properties of Property
(meta-properties, I suppose) to describe multiplicity and a few other
things.  We defined these meta-properties in RDF schema (reflexively) so the
whole system is fully defined.  But it would be much better if there was a
common definition for something more than rdfs:domain and rdfs:range.

I hope this helps,
---
Arnold deVos
Langdale Consultants
adv@langdale.com.au
Received on Wednesday, 10 November 1999 18:08:09 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:51:42 GMT