Comments on: Integrating ontologies is a mess http://blog.evolvingbits.org/2008/02/04/integrating-ontologies-is-a-mess/ Ideas, bits, and bytes, mostly in the area of phyloinformatics Wed, 24 Jun 2009 11:42:08 +0000 hourly 1 http://wordpress.com/ By: jane http://blog.evolvingbits.org/2008/02/04/integrating-ontologies-is-a-mess/#comment-8 Sun, 10 Feb 2008 15:00:11 +0000 http://evolvingbits.wordpress.com/?p=9#comment-8 Great to see commentary on this issue. I believe we are in exciting times. Although we may call the current situation “a mess” — we actually have an information infrastructure, and a host of new enabling technologies, that permit integration, mapping, and linking different vocabularies, along the ontology/vocabulary continuum, in a way that was never before possible.

By our sharing different developments, and hearing from different communities…I think we may be able to improve the situation dramatically, and your [Hilmar] generating this discussion is a positive step in this direction.

It seems to me that the development of registries is a big component of our cleaning up (or improving) the mess…

Thanks for the prompting the discussion!! jane

]]> By: drycafe http://blog.evolvingbits.org/2008/02/04/integrating-ontologies-is-a-mess/#comment-6 Wed, 06 Feb 2008 13:28:37 +0000 http://evolvingbits.wordpress.com/?p=9#comment-6 Hi Mark -

that’s great (have I missed where this is documented at BioMoby.org or should it be more prominent?). However, am I missing some detail, or is this equivalent to posting the ontologies for download in RDF/OWL? I.e., it’s not an API for querying and accessing individual terms, and programmatically navigating their neighborhood, right?

Of course I could suck them up using a OWL/RDF-capable library, but isn’t that a bit akin to saying we don’t need an ontology web-service API, just a SPARQL endpoint?

]]> By: Mark Wilkinson http://blog.evolvingbits.org/2008/02/04/integrating-ontologies-is-a-mess/#comment-4 Tue, 05 Feb 2008 17:35:21 +0000 http://evolvingbits.wordpress.com/?p=9#comment-4 Hi Hilmar,

The Moby ontologies are published as OWL-RDF documents, as is the entire Biomoby registry:

http://biomoby.org/RESOURCES/MOBY-S/Objects
http://biomoby.org/RESOURCES/MOBY-S/Namespaces
http://biomoby.org/RESOURCES/MOBY-S/Services
http://biomoby.org/RESOURCES/MOBY-S/ServiceInstances

these addresses are available through an API call to ensure that they are always discoverable if they change.

The ontologies are *also* available for manipulation through code (in particular for the purpose of adding/removing nodes) but for simple browsing of the ontology almost everyone uses the dynamically-generated RDF files at the addresses above.

Cheers!

Mark

]]>