Re: Pointer to the ontology for generic resources. HTTP Link header.

Tim Berners-Lee wrote:
> Alan,
> Here is the ontology for  generic resources:

I still don't quite understand the full utility of these ontologies.

One thing is for sure, it's organizing information resources (long
discussed in this list) in subclasses but I fail to see how it actually
defines an information resource.

The most used predicates were comment and label (which are important but
means nothing to a computer) and subClassOf (which determine the
hierarchy but doesn't define anything).

To say that a Cat is an Animal doesn't define cats nor animals, you need
to say that they are alive, explain what alive is an then define their
inner structure and why and more especially, what is the relation
between internal structures in terms of input/output, domain
relationship and the context they operate.

In this ont.owl, the only other predicate used to actually say something
additional to the hierarchy is inverseOf and even this predicate is not
used evenly: timeSpecific is inverseOF timeGeneric (and vice-versa) but
Content-TypeGeneric is not inverseOf content-TypeSpecific (although the
opposite is). But that's not the point (it could be a typo)...

The real point is, why should we create an ontology without any
additional information just for the sake of structuring concepts? Is it
the whole point of ontologies and I'm missing the point completely?


Received on Tuesday, 6 November 2007 11:57:57 UTC