- From: Butler, Mark <Mark_Butler@hplb.hpl.hp.com>
- Date: Fri, 10 Oct 2003 14:23:19 +0100
- To: "'www-rdf-dspace@w3.org'" <www-rdf-dspace@w3.org>
- Message-ID: <E864E95CB35C1C46B72FEA0626A2E8082061B6@0-mail-br1.hpl.hp.com>
Hi team I have put together a short "hello world" mapping demonstration Java program that uses the Jena Inferencing API and RDQL. It takes the artstor sample data, converted from XML to RDF/XML using the stylesheet I wrote, and the artstor schema I wrote which includes the Dublin Core mapping from the VRA Core spec, and uses this to generate a Dublin Core representation of the sample data. Now as Andy has pointed out, the VRA to Dublin Core map is broken, and the artstor schema and hence the stylesheet need further refinement. However this program demonstrates we can leverage the Jena 2 Inference API and RDQL to do the map with only a small amount of Java code. The other point is even though we know the map is broken, examining the results of the broken map do help us understand how the map is flawed. I've incorporated the file into the CVS in simile/sandbox/mapartstordc, but in case you want a quick look, here is the Java code and an RDF/XML and an N3 version of the result of applying the map to the sample Artstor data record. regards Dr Mark H. Butler Research Scientist HP Labs Bristol mark-h_butler@hp.com Internet: http://www-uk.hpl.hp.com/people/marbut/
Attachments
- application/octet-stream attachment: Map.java
- application/octet-stream attachment: sample_dc.rdf
- application/octet-stream attachment: sample_dc.n3
Received on Friday, 10 October 2003 09:23:43 UTC