DanC: W3C Coordination Group Note 13 April 2004
danbri: They tried to generalise Wordnet to deal with multiple languages, including bridges between.
danbri: Idea is (per my wordnet vocab) to show how we might map hypernym/noun hierarchies into RDF for purpose of mixing with other info.
danbri: Cost/benefit of producing multiple class and property URIs (IRIs) when the meaning doesn't differ is questionable. But for concepts that aren't lexicalised in (say) English, why not use Japanese characters in the XML element name?
danbri: I gave a talk at NICT yesterday that touched on these issues.
danbri: The example here uses Kota's "Uranai" vocabulary, but gives it a Japanese property name instead of English (as an experiment). Also I declare a class and make it (strong claim!) owl:sameAs foaf:Person. That is an intentional mistake, should probably use reciprocal subClassOf instead.
danbri: Feedback welcomed.
danbri: Cost/benefit of producing multiple class and property URIs (IRIs) when the meaning doesn't differ is questionable. But for concepts that aren't lexicalised in (say) English, why not use Japanese characters in the XML element name?
danbri: I gave a talk at NICT yesterday that touched on these issues.
danbri: The example here uses Kota's "Uranai" vocabulary, but gives it a Japanese property name instead of English (as an experiment). Also I declare a class and make it (strong claim!) owl:sameAs foaf:Person. That is an intentional mistake, should probably use reciprocal subClassOf instead.
danbri: Feedback welcomed.