[openbiblio-dev] ORE in openbiblio

[William Waites]

On 10-06-28 04:08, Peter Murray-Rust wrote:
> On browsing the openbiblio changlelog (which on first sight looks
> exciting) I noticed that it was importing ORE. Our group has a project
> on ORE (OREChem, coordinated by Carl Lagoze) so we're certainly
> interested in how it is used and develops in Openbiblio. Also any
> communal tools would be very useful.

In openbiblio I've tried to make it so each entity (understood
as a distinct thing, analogous to an object in an ORDB, such
as a MARC record or a Work or a Person etc) gets its own
named graph. In other words the named graph acts as a
container for statements about a single thing. This way when
changes are made they need only be made in one place.

Usually when people are looking at data they want to see
information that comes from more than one thing. This is
where the ORE comes in, using Aggregations. Processing
a single MARC record might result in,

  * one obp:Work
  * one obp:Manifestation
  * one obp:Item (possibly)
  * one or more foaf:Person (authors, translators, biographical
     topics, etc.)
  * one or more dc:Agent (publisher)

then several ore:Aggregations are made,

  * Work + Authors + Subjects
  * Author + Work for each Work the Author has written
  * Manifestation + Publisher
  * Publisher + Manifestation for each Manifestation that
     has been published by them

Also intended is to use ore:Aggregation for reading lists...

This hasn't yet been re-implemented after the MARC parser
refactor. Might be good to think about how to do it in a
more generic and reuseable way...

Cheers,
-w

-- 
William Waites           <william.waites at okfn.org>
Mob: +44 789 798 9965    Open Knowledge Foundation
Fax: +44 131 464 4948                Edinburgh, UK

RDF Indexing, Clustering and Inferencing in Python
		http://ordf.org/