On 30.04.2015 11:25, Jean-Baptiste Pressac wrote:
Hello,
Does this also means that the RDF data available via for instance
http://www.wikidata.org/entity/Q235382.nt or
http://www.wikidata.org/entity/Q235382 could not be queried via SPARQL
unless you download the .nt file and use for instance Jena ARQ
<https://jena.apache.org/documentation/query/index.html> on your own
computer ?
Maybe not your computer, but somebody has to download the data in some
way in order for it to be queried. Query answering is a complicated
process that does not usually work on the fly, and somebody has to do
the math in the end. There are some approaches to explore linked data in
a query-like manner in real time, but it should be clear that this will
always take much longer than if you have downloaded the data first.
Does this also means that there is no use to publish RDF data linking to
Wikidata like for instance :
@prefix mydata: <http://mydata.fr/> .
@prefix cidoc: <http://www.cidoc-crm.org/cidoc-crm/> .
@prefix wikidata: <http://www.wikidata.org/entity/>
mydata:event/1 a cidoc:E67_Birth ;
cidoc:P98_brought_into_life mydata:person/80 ;
cidoc:P7_took_place_at wikidata:Q235382 ;
This is still a useful thing to do for several reasons. First of all,
the link connects your data and clarifies its meaning. This is useful to
consumers who find your data. Second, there are linked data crawlers
that aggregate linked data from many sources to provide you with a query
service. OpenLink is running one such service, and if they managed to
find your data, you could use their service to issue queries.
Regards,
Markus
Thanks,
Jean-Baptiste Pressac
Traitement et analyse de bases de données
Production et diffusion de corpus numériques
Centre de Recherche Bretonne et Celtique
Unité mixte de service (UMS) 3554
20 rue Duquesne
CS 93837
29238 Brest cedex 3
tel : +33 (0)2 98 01 68 95
fax : +33 (0)2 98 01 63 93
Le 29/04/2015 21:44, Markus Krötzsch a écrit :
On 29.04.2015 20:56, Luca Martinelli wrote:
Dear all,
I need to know about the possibility of making queries on a Wikibase
instance. I think it is possible to make queries on data on a
particular instance only with external tools at the moment, right?
Yes, this is correct. The SPARQL query support that we currently offer
is obtained by making an RDF export and loading it into a SPARQL
database (we use Virtuoso but you could also use BlazeGraph, for
example; both have free and open source versions and are not hard to
install overall; if your data is not so large, you could also try
Jena; there are further open source RDF databases, but these are the
most prominent right now I think).
The RDF export, too, is not currently generated by Wikibase. However,
Wikidata Toolkit, which we use to make the RDF dumps, can be used with
data from any Wikibase installation in theory. In practice, nobody has
asked for this yet and we might have to make a few adjustments to
really get it to work in a convenient way. For a start, I don't know
what kind of export options a standalone Wikibase offers you at the
moment. We can use the usual XML-based page dump if it contains valid
JSON for a change (this was not the case for Wikidata last time I
checked ...). Better yet would be the JSON exports, but I don't know
if you can generate them with vanilla Wikibase or if WMF is using some
special tools for this.
Anyway, it can't be too hard to make this work and once it is done you
would have a query service that is at the same level as the one of
Wikidata. You could even combine data from more than one Wiki in one
RDF database, e.g., to run queries over data from both.
Regards,
Markus
_______________________________________________
Wikidata-l mailing list
Wikidata-l@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/wikidata-l
_______________________________________________
Wikidata-l mailing list
Wikidata-l@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/wikidata-l
_______________________________________________
Wikidata-l mailing list
Wikidata-l@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/wikidata-l
