Re: Formation of RDF terms

Aaron Swartz wrote:
> This is what I was afraid of. So there is no definitive way to determine the
> namespace of an RDF "term"? Are RDF terms even considered to have specific
> namespaces -- or is this just a side-effect of the XMLNS system?

and Dan Brickley wrote:
> If we said that, regardless of syntax, RDF predicates should be named
> with URIs whereby a regular-expression split can separate it into
> namespace-name and name-within-the-namespace parts, that'd be one way.

I have a problem here !
What do you folks call a namespace, precisely ?
RDF terms are URIs, and I am not aware that URIs have any namespace ?
Nor that they need one, anyway, since they are supposed to be unique...

XML Element names need namespace, since they can not be guaranteed to be unique, hence we use QNames.
I never liked the way RDF uses QNames to produce URIs, since it induces confusion...

The problem is : when encountering an RDF property, we would like to know where to get the schema where this property is defined...
RDF M&S says (2.2.1)

   The namespace name URI in the namespace declaration is a globally unique
   identifier for the particular schema this metadata author is using to
   define the use of the Creator property.

Which is ugly because it makes the knowledge of the schema URI dependant on the syntax,
and implies that any property URI starts like the defining schema URI...
With all due respect, Dan, if I requested an errata in RDF M&S, that would be this one :)


On the other hand, RDFS proposes the property rdfs:definedBy, which is, IMHO, the sane solution, since it belongs to the model, and not the syntax...

  Pierre-Antoine

-- 
Sometimes I think the surest sign that intelligent life exists elsewhere in the
universe is that none of it has tried to contact us.
(Bill Watterson -- Calvin & Hobbes)

Received on Friday, 26 January 2001 09:41:34 UTC