- From: Antoine Zimmermann <antoine.zimmermann@emse.fr>
- Date: Wed, 27 Feb 2019 14:57:26 +0100
- To: public-linked-json@w3.org
Le 27/02/2019 à 14:39, James Anderson a écrit : > [skip] > >> >> In this case, it is using RDF to describe a named graph pair (n,g). This could be thought of as an abstraction of the fact that the URI n has been set up at a certain point in time by someone to return a representation of g when the URI is looked up. In this case, the metadata apply to the association of the URI to the graph, which can have a creation date and so forth. >> >> I neither support nor disapprove this interpretation, > > but, in any case, this reads as if you agree that interpretations exists, that is, it is a case where rdf has been used to describe an rdf graph. Again, there are different ways of interpreting the graph name in a named graph pair (n,g). One option is to interpret n as g. In this case, one cannot make assertions about different occurrences of the same graph. It is a problem for metadata. Another option is to interpret n as (n,g). You can distinguish different occurrences of the same graph on the Web and describe them separately. There are limitations of this last approach (how to distinguish between different versions of the graph document?). Other options exist. All these options have their use cases. I don't very much support one over the others. I hope it's clearer now. --AZ > >> but I must say it works well for the "graph metadata" use case. Carroll et al. in their "named graph papers" mention these use cases and insist that the name should identify the pair, not the graph alone. > > -- Antoine Zimmermann Institut Henri Fayol École des Mines de Saint-Étienne 158 cours Fauriel CS 62362 42023 Saint-Étienne Cedex 2 France Tél:+33(0)4 77 42 66 03 Fax:+33(0)4 77 42 66 66 http://www.emse.fr/~zimmermann/ Member of team Connected Intelligence, Laboratoire Hubert Curien
Received on Wednesday, 27 February 2019 13:57:49 UTC