Excellent - I'm glad I asked - that makes sense. I see I am probably misreading http://labs.mondeca.com/dataset/lov/details/vocabulary_akt.html And I may have been misreading those two colours. But it is actually the case that you say AKT voaf:reliesOn dcterms,, since voaf:metadataVoc is a sub property of voaf:reliesOn. (As I now understand the tree is saying on that page.) And of course that means that dcterms voaf:usedBy AKT. I still worry that this looks rather misleading, but I can see why it happens. Best Hugh
On 23 May 2012, at 10:56, Pierre-Yves Vandenbussche wrote: > Hi Hugh, > > actually, LOV states there is a relation between AKT and DC of type > voaf:metadataVoc and not reliesOn ( > http://labs.mondeca.com/dataset/lov/details/vocabulary_akt.html ) > This relation means AKT vocabulary uses DC to describe its own metadata (like > dc:title, etc.) > > I agree that one can easily confuse colors :) > > voaf:reliesOn is a very general relation and is practically never used. > Instead we prefer some more meaningful relations (reliesOn subproperties) > > Pierre-Yves Vandenbussche. > > > On Wed, May 23, 2012 at 10:47 AM, Hugh Glaser <[email protected]> wrote: > Hi Antoine, > Many thanks. > Actually, that prompts me to ask a question about LOV. > I see that LOV says the AKT ontology voaf:reliesOn dcterms (with a very big > circle). > This puzzled me because I did not think there was any connection between AKT > and DC, so I drilled in. > I found, as far as I can tell, it is because the AKT vocabulary itself (at > http://www.aktors.org/ontology/portal ) uses dct:creator, and documents that > use the AKT vocab use dc:creator, dc:title, etc., about the documents and > content itself. > This was not my intuitional reading of voaf:reliesOn - was it intended? > And if so, is it widespread in LOV? > (I realise this is probably a hard problem in general!) > Best > Hugh > > On 22 May 2012, at 13:30, Antoine Isaac wrote: > > > Hi Hugh, > > > > > >> Well Dominic's site is definitely not isolated. > >> It is very well linked at the ontology level, not instance, however. > >> I thought his question was timely, since TimBL asked the question at the > >> panel at LOD2012 as to whether the criteria for inclusion in the LOD Cloud > >> should be changed. > > > > > > > > Yep. As far as I'm concerned, something like a sort of mix between the LOD > > cloud and the LOV one (http://labs.mondeca.com/dataset/lov/) would be > > really interesting. > > But still someone needs to volunteer (as opposed to "being requested") to > > do it :-) > > > > > >> Personally I think it is a shame that such a resource should lose a lot of > >> its visibility because it does not pass the rules. > >> And I think that putting links in simply to get into the Cloud is not > >> something that should be encouraged - links should be put in because they > >> are sensible. > >> Without visibility, others (such as you!) will be less aware of it and so > >> not build the links that would actually bring it into the cloud without > >> Dominic doing anything (as you are now thinking of doing, since Dominic > >> has made you more aware of it). > > > > > > Yep. In fact this is part of the reasons why the Library Linked Data > > incubator decided to create its own group on The Data Hub > > (http://thedatahub.org/group/lld). It helped us to make the datasets from > > our community more visible to our community, without making it a hard > > pre-requisite to adhere to other communities' requirements. > > Some nodes (or group of nodes) at > > http://www.w3.org/2005/Incubator/lld/XGR-lld-vocabdataset/#Library_Linked_Data_at_CKAN > > are indeed "isolated", in the LOD cloud sense. > > > > Antoine > > > > > >> > >> On 22 May 2012, at 08:42, Antoine Isaac wrote: > >> > >>> Hi Dominic, > >>> > >>> I guess that it was with the LOD *cloud* that you had issues. It looks a > >>> bit severe, but I think I understand the motivations: if the cloud > >>> admitted isolated nodes, it would have many of them, and that would look > >>> weird... But of course that does not make your contribution less > >>> interesting. On the contrary, the BL work has incredible potential for > >>> our domain! > >>> Btw let me know if you're interested in links with data.europeana.eu. We > >>> can maybe try something... > >>> > >>> Best, > >>> > >>> Antoine > >>> > >>> PS: I'm copying the email to the LOD-LAM list: I suppose some people will > >>> be interested to continue the discussion with you there! > >>> > >>> > >>>> +1 (best I can do). FWIW, the day buying your way in ceases to be the > >>>> certain method of acceptance will be a very good day for all. > >>>> > >>>> -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- > > -- > >>> -- > >>>> *From:* Dominic Oldman<[email protected]> > >>>> *To:* [email protected] > >>>> *Sent:* Monday, May 21, 2012 1:10 PM > >>>> *Subject:* Cultural Heritage Data > >>>> > >>>> Hugh suggested that I post this. > >>>> We are currently working with other museums aligning our catalogue data > >>>> using the CIDOC-CRM ontology. We can now run single federated queries > >>>> based on semantic alignment without the need to insert specific linking > >>>> triples. When we applied to advertise our site on the LOD cloud we were > >>>> turned down because we hadn’t inserted specific links to other data > >>>> sources. I realise that I could just stuff in a few links to Dbpedia to > >>>> get accepted - but given that we can harmonise data to a very high > >>>> degree with another open CRM RDF data source perhaps we should still be > >>>> allowed formal acceptance to the open data community. > >>>> Dominic Oldman > >>>> *Deputy Head of IS * > >>>> *IS Development Manager* > >>>> *ResearchSpace Principal Investigator* > >>>> *British Museum* > >>>> +44 (0)20 73238796 > >>>> +44 (0)7980 865309 > >>>> www.BritishMuseum.org > >>>> www.ResearchSpace.org > >>>> > >>>> > >>> > >>> > >> > > > > -- > Hugh Glaser, > Web and Internet Science > Electronics and Computer Science, > University of Southampton, > Southampton SO17 1BJ > Work: +44 23 8059 3670, Fax: +44 23 8059 3045 > Mobile: +44 75 9533 4155 , Home: +44 23 8061 5652 > http://www.ecs.soton.ac.uk/~hg/ > > > -- Hugh Glaser, Web and Internet Science Electronics and Computer Science, University of Southampton, Southampton SO17 1BJ Work: +44 23 8059 3670, Fax: +44 23 8059 3045 Mobile: +44 75 9533 4155 , Home: +44 23 8061 5652 http://www.ecs.soton.ac.uk/~hg/
