- From: Alan Ruttenberg <alanruttenberg@gmail.com>
- Date: Wed, 9 Apr 2008 09:09:47 -0400
- To: Dan Brickley <danbri@danbri.org>
- Cc: Pat Hayes <phayes@ihmc.us>, Leo Sauermann <leo.sauermann@dfki.de>, www-tag@w3.org, SWIG <semantic-web@w3.org>
On Apr 9, 2008, at 4:16 AM, Dan Brickley wrote: > And for massive ontologies, like the various wordnet > representations, breaking them up into parts has its own merits: why > download a description of 50000 classes just because you've > encountered @yone. I concur. Moreover, even without cacheing 303s, if each document includes an owl:imports, and/or a rdfs:isDefinedBy for the subject term, and a client decides to follow those links, they will find the rest of the terms and can cache that and not need to fetch the rest. An an example of this approach, here is a prototype I am working on for the OBI ontology. http://purl.obofoundry.org/obo/OBI_0000225 The intention is that each page organized around returning useful enough information about a single term in the ontology - exactly what is a judgement call. Each has an owl:imports statement so that clients that need the correct semantics are able to retrieve them, and there are rdfs:isDefinedBy statements for the main term, as well as any terms that I have borrowed from ontologies that are not, as yet, OWL DL (barring oversights - that's the intention). Status of prototype: Only one of the html pages exists, and it is not yet generated with hyperlinks. RDF/XML is available for only classes in the ontology at the moment, not properties. There is an unresolved issue regarding mime-types: application/rdf+xml doesn't trigger use of the stylesheet, so the page is served as application/xml (shouldn't every application/rdf+xml also be a application/xml ?) -Alan
Received on Wednesday, 9 April 2008 13:17:04 UTC