Hi Alex As long as the Fedora object (or a datastream) forms the subject of each triple in RELS-EXT (or RELS-INT) in that object, I don't see there would be a problem doing what you describe.
FYI we have discussed enhancing Fedora's RDF capabilities, you might like to take a look at https://wiki.duraspace.org/display/FCREPO/Supporting+the+Semantic+Web+and+Li nked+Data for some ideas on the directions this might take. Regards Steve -----Original Message----- From: Alex Rodriguez Lopez [mailto:alo...@flordeutopia.pt] Sent: 10 December 2010 11:35 To: fedora-commons-users@lists.sourceforge.net Subject: [fcrepo-user] Resource Index, addRelationship and relations between resources (other than objects and components) Hi Fedora users! I understand that, according to the documentation [1] there is a resource index backed by a triple store which can be queried with SPARQL (and others) in a "semantic fashion". It is even automatically indexed and full-text indexing is easy to enable on it. Now here comes my doubt, for it is also mentioned that this is to be mainly used with Fedora Relationship Ontology [2] and: "The RELS-EXT Datastream is intended to be dedicated to solely object-to-object relationships and not used to make general descriptive assertions about objects." So what is wrong with this? Does it means that there is a better way to assert relations between objects and other resources that may not be objects in the repository? I find it really handy to be able to use addRelationship method to add triples one by one, having as subject a fedora object, but having as predicate any relation from any ontology and as object either: a) other objects in the repo (as intended, good for composite objects), b) resources or classes identified by URI (persons, organizations, etc), c) literals (plain text, automagically indexed by lucene if full-text indexing enabled). So I'm I supposed to work on new Datastreams for this, enable GSearch etc, when using addRelationship + sparql queries (risearch) + the full-text index already does the trick? What are the best-practices to semantically enrich fedora objects and later query about those relations? Do people manage these relations outside Fedora? 1000 thanks for any suggestions about this!!! Kind regards, Alex [1] https://wiki.duraspace.org/display/FCR30/Resource+Index [2] http://www.fedora.info/definitions/1/0/fedora-relsext-ontology.rdfs ---------------------------------------------------------------------------- -- _______________________________________________ Fedora-commons-users mailing list Fedora-commons-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/fedora-commons-users ------------------------------------------------------------------------------ Oracle to DB2 Conversion Guide: Learn learn about native support for PL/SQL, new data types, scalar functions, improved concurrency, built-in packages, OCI, SQL*Plus, data movement tools, best practices and more. http://p.sf.net/sfu/oracle-sfdev2dev _______________________________________________ Fedora-commons-users mailing list Fedora-commons-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/fedora-commons-users
