Re: Discovering a query endpoint associated with a given Linked Data resource

Thanks.
Yeah, pretty close, although I doubt that the RDF returned would have anything of type dcat:Distribution in it to let me form the triple.

Sorry, looking at my email I realise I was having a brain fart on seeAlso - of course endpoints are Resources (unless I am also having a senior moment :-) ).
Also, sorry, by the way, Laurens, I think backlinks services such as LOD Laundromat are great, and an important part of the LD world.
It is just that I think it should be simpler for consumers in this case (and more efficient)

Perhaps I should have suggested:
<http://sws.geonames.org/3405870/> void:inDataset :DBpedia .
:DBpedia void:sparqlEndpoint <http://dbpedia.org/sparql> .
to be delivered with the RDF.
I think I could cope with 2 triples!

Unfortunately, void:inDataset has foaf:Document as the range.
So that would mean I would need to do a bit more stuff, and it is getting more complicated again.

And also:
http://www.w3.org/TR/void/ 6.3 says
"Providing metadata about the entire dataset in such a scenario should not be done by including VoID details in every document. Rather, a single VoID description of the entire dataset should be published, and individual documents should point to this description via backlinks.”
(My brain really hurts from trying to remember all this stuff form many years ago now.)

However, I take that to mean that I shouldn’t put *all* the VoID stuff in each document.
If I have licence stuff etc., then it should be in a common void.ttl or whatever.

I don’t think there is any harm putting some selected bits of VoID in the document - in fact, if the VoID data is in the store itself (as I assume it should be), then some of it will arrive as a natural part of the SCBD.

Is there some simple idiom we could all use to carry the info?
I dunno, something like:

:DBpedia void:uriRegexPattern “^http://sws\\.geonames\\.org/3405870/“ .
:DBpedia void:sparqlEndpoint <http://dbpedia.org/sparql> .
But that does require some processing.
Any better ideas? I suspect I am just being thick.

Would that do what I want?
(Would it also do what Nandan wants? :-) )

Cheers
> On 26 Aug 2015, at 13:44, Ghislain Atemezing <auguste.atemezing@eurecom.fr> wrote:
> 
> Hi Hugh,
> 
>> Le 26 août 2015 à 14:23, Hugh Glaser <hugh@glasers.org> a écrit :
>> 
>> Another major reason is that the publisher may not have the rights to publish .well-known and its ilk.
>> And if it comes with the RDF we can be really confident of the provenance and trust of who has recommended it.
>> Also, it is a damn sight easier to maintain, than to rebuild the vOID document every time something changes.
> 
> At the data level, DCAT [1] already defines a property http://www.w3.org/ns/dcat#accessURL which points to “a landing page, feed, SPARQL endpoint or any other type of resource that gives access to the distribution of the dataset.”
> 
> Of course, the question will remain if no one uses this property at least at dataset/data catalog level. 
> 
> Best,
> 
> Ghislain
> 
> [1] http://www.w3.org/TR/vocab-dcat/ 
> 
> 
> 
> 

-- 
Hugh Glaser
   20 Portchester Rise
   Eastleigh
   SO50 4QS
Mobile: +44 75 9533 4155, Home: +44 23 8061 5652

Received on Wednesday, 26 August 2015 13:54:26 UTC