Re: BioRDF: URI Best Practices

On 7/20/06, Matthias Samwald <samwald@gmx.at> wrote:
>
> On the page
> http://esw.w3.org/topic/HCLSIG_BioRDF_Subgroup/Tasks/URI_Best_Practices/Use_Cases
> I have written that an "ontology of resolvable resources" would be practical in some cases.
>
> I have created such a small ontology, it can be downloaded / imported from
> http://neuroscientific.net/ont/resolvable.owl

fyi,
While working on an ontology which mapped the Atom Syndication Format
(RFC 4287), Reto Bachmann-Gmuer was motivated to write an auxiliary
ont [1] which attempts to capture information about resources and
their representation(s), based on the notions in WebArch [2].

This was prompted by an apparent mismatch between the WebArch
concepts, the resource in RDF and the Atom spec (maybe irony there
somewhere as there were plenty of RDF people, and at least two members
of the TAG on the Atom WG). Basically expressing the simple notion of
an identified entry on a weblog or whatever gets a bit complicated
when you bring in the possibility of multiple versions in multiple
(human) languages on top of multiple mime types.

For what it's worth I believe in the vast majority of cases, including
that of ontology terms, the assumption that http scheme URIs
(/URIrefs) will be dereferenceable seems entirely reasonable. Where
other naming schemes are mapped to URIs this may be problematic, but a
404 isn't the end of the (open) world.

Having said that, I can't see any harm in asserting information in
RDF/OWL like "there exists a text/html representation of the resource
http://example.org", although in such cases the Web server should
probably be considered the authority.

Cheers,
Danny.

[1] http://atomowl.org/ontologies/infoatom
[2] http://www.w3.org/TR/webarch/

-- 

http://dannyayers.com

Received on Friday, 21 July 2006 09:58:42 UTC