xml-dev - Re: Services-based automation (WAS RE: Realistic proposals to the W3C?)

[Jonathan Borden <jborden@mediaone.net>]

Bullard, Claude L (Len) wrote:

> Well, DUH! If a semantic web means is a namespace URI points
> to a schema somewhere, I am completely underwhelmed.

RDF schemas are intended to complement not replace XML schemas. An RDF
schema defines a semantic hierarchy, or network, of element names. The URI
created by concatenating the namespace URI to the element name identifies an
RDF Schema Class to which the instance belongs, or may identifiy an RDF
Schema Property. This is totally different than syntactic level constraints
placed by a DTD or XML Schema on an XML document.

An RDF schema defines a fragment of a semantic network which includes
instances of the classes and properties. The semantic web is at its simplest
a network of nodes and arcs (properties) which use URIs for both node and
arc names. The namespace mechanism is simply the mechanism by which XML
element names are translated into URIs identifying nodes and arcs in the
'semantic web'.

The XML model defines a node labelled directed graph. In this model, arcs or
edges have the type "element" "attribute" "CDATA section" "comment" etc.

The RDF model defines an edge labelled directed graph, for example arcs may
be labelled "color", "type.of.cheese" "type.of.sauce" "topping". This may
not seem a radical difference but understand that software which makes
inferences regarding the properties of a pizza, as represented in an RDF
graph, may be totally unconcerned about whether the "type.of.cheese" is an
attribute or an element and whether this property is serialized before or
after the "type.of.sauce" property.

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