- From: Thomas B. Passin <tpassin@comcast.net>
- Date: Fri, 17 Sep 2004 09:28:51 -0400
- To: www-rdf-interest@w3.org
Eric Jain wrote: > > Howard Katz wrote: > >> So in short (if I may): in XML the typing is built-in; the >> relationships are >> implied. In RDF, the typing might or might be present, and the >> relationships >> are (generally) explicit. > > > Here's a different point of view: XML Schema seems to suggest that XML > element names are property names, not type names. The same element name > can be used with different types in different places, and you can even > do the following, assuming the "publication" element has been defined to > be of type Publication (which has several subclasses): > > <publication xsi:type="Article"> > ... > </publication> > <publication xsi:type="Book"> > ... > </publication> If you get right down to it, xml per se has no notion of "type" by that name. If you look at sgml nomenclature, we have "generic identifiers" as the closest thing. XML schema is built around "named types". An element of a given name can be part of different named types, basically because a named type can designate a whole structure. However, for rdf purposes one would decompose that whole structure into its components, which would include resource and property types (or just resource types, considering that a property too is a kind of "resource" in rdf). With OWL, one could describe that structure as classes, subclasses, properties, their domains and ranges, etc. However, since the two models are different, you might not be able to get an exact correspondence. From the point of view of the usage I was espousing, it does not matter if an element is playing the role of a property or not. The element's name indicates its type either way. -- Thomas B. Passin Explorer's Guide to the Semantic Web (Manning Books) http://www.manning.com/catalog/view.php?book=passin
Received on Friday, 17 September 2004 13:26:29 UTC