xml-dev - RE: XML Schemas: Best Practices

[Jonathan Borden <jborden@mediaone.net>]

Roger L. Costello wrote:
>
> "Michael W. Ripley" wrote:
> >
> > IMHO, application specific semantics
> > are what keep getting us in trouble with regards to data
> > interoperability.  Hence the drive for ontologies, common data
> > environments, standardized database schemas, common data models, etc.
> >
> > What should be a Best Practice?  I'm in strong favor of universal
> > semantics.  It's very difficult to implement, but it's what we should
> > try to do.
>
> Mike presents a strong argument in favor of creating elements with the
> same semantics, regardless of the application that uses them (he calls
> this "universal semantics").  He argues, for example, that if I create
> an element <car> then it should have the same semantics in all
> applications.
...
>
> What do you think?  When you create a schema component should that
> component be expected to have the same semantics regardless of the
> application that uses it, i.e., universal semantics?  Or, should the
> component be able to "semantic-morph" to each application, i.e.,
> localized semantics?

	I think this is precisely the intended role of namespaces. In particular
the mechanisms of rdf:type, rdf:subClassOf, RDF Schemas and DAML-ONT (see
http://www.daml.org/2000/10/daml-ont.daml) assign URI references to terms
(e.g. the combination of a namespace URI and an element name fragement
identifier).

One simply cannot expect the term "car" to have a universally defined
meaning. Do you intend this to mean an automobile or the contents of the
address register? Unless you place the term in context there is no
reasonable way to disambiguate.

Jonathan Borden
The Open Healthcare Group
http://www.openhealth.org