[openbiblio-dev] Curated lists and openbiblio Web UI

Mon Jul 12 14:03:11 UTC 2010

On 10-07-12 08:46, Benjamin O'Steen wrote:
> I'm going to be doing a good bit of work for the Open Citation project
> with Peter Shotton, where curated bibliographies are going to be very
> important. It would be good not to do things in different ways!

Absolutely!

> David is understandly keen to use the ontology he has developed: 'CiTO'
> -
> http://imageweb.zoo.ox.ac.uk/pub/2009/citobase/cito-20100528-1.6/cito-content/owldoc/

It's really too bad that CiTO has yet another collection of classes
lie Book, Editorial, etc. The class hierarchy builds on the frbr
vocabulary which as discussed here and elsewhere is less than
ideal.

On the other hand the predicates generally do not define domain
and range so can be reused...

> I wondered what your plans were for this area of the bibliographica
> functionality? Curated lists, or something a little more (using some of
> the argumentative predicates in the CiTO ontology, like 'confirms',
> etc?)

I've been meaning to write up the way I see aggregations/lists
being done since it was mentioned on the list last week. Briefly
making a curated list is just making an ore:Aggregation that
includes another ore:Aggregation per work/book/whatever.
The reason for two levels is that the lower level contains the
Work and its Authors since you normally want that information
together whereas the top level is a collection effectively of Works.

Richer predicates, such as the argumentative ones from CiTO
have been contemplated since the beginning but I think this
might be orthogonal to curated lists?

Not sure what happens when there is scholarly disagreement
about whether one work confirms another... Do we need to
go down the reification road here? e.g.:

    scholar1 a foaf:Person ;
        believes [ a Belief ;
                         rdf:subject book1 ;
                         rdf:predicate cito:confirms ;
                         rdf:object book2 ] .

    scholar2 a foaf:Person ;
        believes [ a Belief ;
                        rdf:subject book1 ;
                        rdf:predicate cito:refutes ;
                        rdf:object book2 ] .

This might expose a missing predicate in cito -- scholar2
might deny that book1 confirms book2 but not go so far as
to say it refutes it. I guess we need to get into beliefs about
beliefs in that case...

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/