Re: Is there a vocabulary to describe a Z39.50 interface in RDF?


I'm not sure putting all in one same place is optimal. advertises itself as a means to exchange data, it would be confusing if some packages are not data but ontologies.

That being said, there is a clear need for something, which has been identified for quite a while.

The first category of tools is "Semantic Web search engines" like Sindice ( and Falcons ( If you just publish some RDF, it should end up there, and be findable by others. But for some reason they are quite under-used, outside the research community. I'm sure they'd be willing to try and adapt their tool, if feedback was sent to them, though.

Then there are registries, like I'm pretty sure tool like that can be loaded with any ontology, even when they are not built using the registry itself. But maybe the tool is to complex for that kind of usage. Maybe Jon could tell us more about it.

Finally there are a couple of surveys like or even our own "metadata element sets" survey at (comments still welcome, by the way!).
But these are certainly too centralized.
In fact, when we started with the LLD XG work, we were happy to use a good community-based solution for listing the relevant datasets in our community, and not do it manually by ourselves.

Now, perhaps someone could set up a CKAN instance for ontologies / metadata elements sets. This is technically possible, but it requires someone to set it up and host it.

@LLD XG folks: in fact this "finding an ontology" issue is quite important in the Linked Data field in general. We should certainly put as a recommendation, that something should be done at the community level. Or that our community tries to participate larger-scale community efforts, if such are set up.



> That's a good idea. CKAN has mainly been a gathering place for "data sets" (broadly defined), but it would be a logical place to also gather ontologies. Among the "data sets" are already some vocabularies, like LCSH, which fall into the middle ground between vocabularies and data that we've struggled with in the LLD draft document.
> I wonder if it would be better to have a separate section for ontologies, or to mix them together but use an icon or keyword to make it easy to pick them out?
> kc
> Quoting "" <>:
>> The CKAN LLD Group <> could at least fill a certain
>> aspect of that need for the time being.
>> Tod Robbins
>> MLIS 2012
>> Information School
>> University of Washington

Received on Wednesday, 13 July 2011 06:47:19 UTC