- From: Andrew Gibson <a.p.gibson@uva.nl>
- Date: Tue, 13 Jul 2010 11:11:07 +0200
- To: Alistair Miles <alimanfoo@googlemail.com>
- CC: Helmut Nagy <h.nagy@semantic-web.at>, public-esw-thes@w3.org
Hi, this is an interesting topic that I have some interest in myself, one comment inline below Alistair Miles wrote: > Hi Helmut > > On Mon, Jul 12, 2010 at 09:18:29PM +0200, Helmut Nagy wrote: > >> Hi, >> >> e.g i got an Concept "Vienna" and i want to make the following mapping: >> >> "Vienna" skos:exactMatch http://dbpedia.org/resource/Vienna >> > > Possibly a reasonable thing to do. Personally, if it's useful for you to > do this, then I would just go for it. If it turns out to have unexpected > consequences, then you can always change later. Might also be worth asking > the dbpedia folks if and how they use SKOS. > Being a bit of a pedant, I do not like to use the skos matching relations unless I can already see that the resource I am linking my concept to is of rdf:type skos:Concept, skos:inScheme to a concept scheme, has a skos:prefLabel and preferrably at least one broader/narrower relationship. Unfortunately this means I rarely use matching relations to anything other than skos resources that I have created myself ;-). In the real world Im not sure this makes any difference, and the Semantic Web will probably survive either way, but I think it gives other people more confidence in your vocabulary modelling if its clear some care has been taken. DBPedia asserts no skos relations for any resource I have seen (which is not necessarily a bad thing, and I havent checked everything), and so in the absence of anything more specific, I just link from my concepts to dbpedia with rdfs:seeAlso, which "is used to indicate a resource that might provide additional information about the subject resource", so for example, its rdfs:labels or rdfs:comments might be useful to a semantic web application, but do not expect any directly relevant skos relationships that it could do skos things with. Cheers, Andrew > >> skos:it? t-shirts? :) >> > > Sorry, Danbri and I had a discussion a while back about ideas for a new inverse > property pair in SKOS that would be used for linking from SKOS concepts to > "real things" like people, mountain ranges, etc., and back again. Danbri wanted > to call the properties skos:it and skos:as, e.g., <mySkosConceptOfVienna> > skos:it <http://dbpedia.org/resource/Vienna>. Danbri also thought this would > be a great marketing ploy, as you could print t-shirts with "skos:it!" on the > front (meaning like, if you've got a thesaurus, SKOS it!) ... unfortunately > there wasn't enough desire for something like this in the SKOS reference > at the time, and it was hard to figure out what the semantics should be, > so we left it. > > It would be interesting to see if anyone else is doing this kind of linking > into and out of SKOS concept schemes, and if so, what properties they're > using. Anyone? > > Cheers > > Alistair > > >> Cheers >> hel. >> >> >> >>> Hi Helmut, >>> >>> On Mon, Jul 12, 2010 at 04:51:38PM +0200, Helmut Nagy wrote: >>> >>>> Hi, >>>> >>>> i got one question, should skos:mappingRelations only be used to map to skos:Concepts of a different skos:conceptScheme or would it also be consistent to use skos:mappingRelations to map to linked data like DBpedia or FreeBase? >>>> >>> I'm not sure, I think it depends what type of resource you want to map >>> to. What assertions are made by DBpedia and FreeBase for the resources you >>> want to map to? >>> >>> Danbri: the return of skos:it? (we should have printed those t-shirts, >>> then it would have definitely gone in the spec... :) >>> >>> Cheers >>> >>> Alistair >>> >>> >>>> Thanks, >>>> hel. >>>> >>>> | Helmut Nagy, Consultant >>>> | Semantic Web Company GmbH >>>> | Lerchenfelder Guertel 43/5 >>>> | A - 1160 Wien, Austria >>>> >>>> COMPANY INFORMATION >>>> | http://www.semantic-web.at/ | http://www.i-semantics.at/ | http://blog.semantic-web.at/ >>>> >>>> PERSONAL INFORMATION >>>> | h.nagy@semantic-web.at >>>> >>>> >>> -- >>> Alistair Miles >>> Centre for Genomics and Global Health <http://cggh.org> >>> The Wellcome Trust Centre for Human Genetics >>> Roosevelt Drive >>> Oxford >>> OX3 7BN >>> United Kingdom >>> Web: http://purl.org/net/aliman >>> Email: alimanfoo@gmail.com >>> Tel: +44 (0)1865 287669 >>> > >
Received on Tuesday, 13 July 2010 09:11:53 UTC