On Thu, 18 Oct 2001, Thomas Baker wrote:
> On Thu, 18 Oct 2001, Dan Brickley wrote:
> > > All of this could be done using hard-coded URLs :-( or via appropriate
> > > links to the registry :-). RDDL looks, to me, like a good candidate
> > > language for providing these linkages. Otherwise, how is someone going to
> > > get from the namespace URI to the registry?
> >
> > How about brief human-readble 'welcome and table of contents' in XHTML
> > with the RDF Schema embedded in-line, alongside RDDL serving as a
> > machine-friendly table of contents pointing off to all the other goodies?
>
> Could those goodies include translations? Rachel alluded to problems in
> implementing this.
Sure. I don't know if RDDL provides detail for that, but in principle, yes.
FWIW here's a technique I used in a "fun and prototyping" RDF schema, the
"Friend of a Friend" vocabulary whose namespace URI is http://xmlns.com/foaf/0.1/
(do a view-source on the XHTML document at the namespace URI to see this)
(i) we have human-friendly HTML introduction to the namespace, with
hypertext pointers to relevant related materials
(ii) we have an inline RDF Schema, making the namespace description accessible
directly by software tools, eg. the Schema-aware MacOS X editor
screenshot'd at
http://rdfweb.org/people/damian/2001/10/RDFAuthor/RDFAuthor.jpeg
(iii) the RDF Schema includes assertions made in an RDF Schema extension
language, in this case, DAML+OIL March 2001 version:
<rdf:Property rdf:about="http://xmlns.com/foaf/0.1/mbox"
rdfs:label="Personal Mailbox"
rdfs:comment="A web-identifiable Internet mailbox associated with exactly one owner.
This property is a 'unique property' in the DAML+OIL sense, in that
there is at most one individual that has any particular personal
mailbox.">
<rdfs:domain rdf:resource="http://xmlns.com/foaf/0.1/Person" />
<rdfs:range rdf:resource="http://www.w3.org/2000/01/rdf-schema#Resource" />
<rdf:type rdf:resource="http://www.daml.org/2001/03/daml+oil#UnambiguousProperty"/>
<rdfs:isDefinedBy rdf:resource="http://xmlns.com/foaf/0.1/" />
</rdf:Property>
...note the rdf:type pointing to DAML+OIL's notion of an "unambiguous
property"; this shows how RDF Schema can be a carrier for richer
semantic annotations expressed using (current and future) RDFS
extension mechanisms. This addresses the concern Carl raised, regarding
the need to be able to include richer constraint machinery at a later
date.
[[
FOAF is expressed as an RDF Schema, annotated with DAML to express the
fact that a foaf:mbox uniquely picks out an individual.
]]
(iv) an RDF "seeAlso" pointer allows for distributed namespace annotation
The Schema's namespace URI is itself annotated, pointing off to
(possibly 3rd-party) further description of the vocabulary. In this
case, to another RDF document at w3.org where we make additional
assertions. This technique can also serve as a lightweight convention for
pointing to multilingual schemata scattered around the Web. This avoids
the risk of coupling the namespace-accessible content to any specific
software tool such as a registry; the Web is it's own registry.
<rdf:Description rdf:about="http://xmlns.com/foaf/0.1/">
<rdfs:seeAlso rdf:resource="http://www.w3.org/2001/08/rdfweb/foaf" />
</rdf:Description>
In this case, http://www.w3.org/2001/08/rdfweb/foaf describes a
sub-property relationship between foaf:name and a couple of other RDF
properties (foaf:personalName and cyc:nameOfAgent).
Just some implementor feedback...
Dan
--
mailto:[log in to unmask]
http://www.w3.org/People/DanBri/
|