How to represent different types of web pages and rdf resources?

I have been wondering about how to best represent different types of web
pages or rdf resources.
I think it is useful to represent some pages like Wikipedia pages as
subclasses of foaf:document.

This is because I would like to be able to easily query for the Wikipedia
pages for a taxon without
getting the Bugguide pages etc.

The geonames ontology does this for Wikipedia pages about a Feature.

Also, this might be useful in assigning levels of "trust" to pages on
different sites.

Similarly, I have the same issue with linkeddata resources.

I would like to distinguish between a bio2rdftaxon resource, a dbpedia
resource, and a uBioLSID resource.

I have done this in my ontology by making subclasses of foaf:document.

The problem with this strategy is that unless you also download my ontology
these relationships are not
obvious.

What is the current thinking about this approach?

Thanks

- Pete

---------------------------------------------------------------
Pete DeVries
Department of Entomology
University of Wisconsin - Madison
445 Russell Laboratories
1630 Linden Drive
Madison, WI 53706
GeoSpecies Knowledge Base <http://species.geospecies.org/>
------------------------------------------------------------