Re: other proposal for the graph task-force

This brings back the question of graph literals. Or should I say, 
literal graphs?

I have no problem with graph literal per se, and would be happy that a 
datatype for graph serialisation exist, but I would not like it to be 
mandated for talking about graphs.

I would rather use a different approach, which has been proposed already 
by Pat, IIRC:

"Graph IRIs" are just tags attached to an RDF graph and don't need to be 
interpreted as graphs in an RDF interpretation. For instance,

:cutekitty { :cutekitty a animal:Dog }

is fine. However, when you want to talk about the graph tagged with the 
Graph IRI :cutekitty, you provide a new URI (which must be interpreted 
as a graph) and connect it to the graph IRI as follows, for instance:

:cutekitty { :cutekitty a animal:Dog }
:graph1 :hasGraphIRI "http://example.org/cutekitty"^^xsd:anyURI ;
:me :believes :graph1 .

This leads to other problems if you publish this kind of things on the 
web but is probably working well inside a graph store.


Le 22/06/2011 19:26, Pierre-Antoine Champin a écrit :
> After reading the Dataset proposal [1] on the wiki,
> I put a different proposal
>
>    http://www.w3.org/2011/rdf-wg/wiki/TF-Graphs/RDF-Quadless-Proposal
>
> which attempts to reconcile the "loose naming" vision with the "strict
> naming" vision by providing a common ground for both.
>
>    pa
>
>
> [1] http://www.w3.org/2011/rdf-wg/wiki/TF-Graphs/RDF-Datasets-Proposal
>


-- 
Antoine Zimmermann
Researcher at:
Laboratoire d'InfoRmatique en Image et Systèmes d'information
Database Group
7 Avenue Jean Capelle
69621 Villeurbanne Cedex
France
Tel: +33(0)4 72 43 61 74 - Fax: +33(0)4 72 43 87 13
Lecturer at:
Institut National des Sciences Appliquées de Lyon
20 Avenue Albert Einstein
69621 Villeurbanne Cedex
France
antoine.zimmermann@insa-lyon.fr
http://zimmer.aprilfoolsreview.com/

Received on Tuesday, 28 June 2011 10:16:48 UTC