Nickerson, David Phillip wrote: > Hi Matt, > > To repond to your points: > > 1) I think the current state in the simulation metadata specification is that > it is recommended best practice to specify a rdf ID for every simulation, > mainly driven by the example of being able to define graphs for that > simulation. Do you think we need to make it stronger and state that rdf ID's > are required in the simulation metadata specification? > The problem with this is that the simulation metadata specification describes the RDF that should be used for simulation metadata, not the RDF/XML. You could use the same RDF schema to represent the data in another format, such as as a text file containing the triples, or in a relational database (or some data-structure in Zope, or an in-memory data-structure in an application). Therefore, it seems conceptually wrong for the metadata specification to put a mandatory constraint on what RDF/XML you can use for given RDF.
The wording in the current draft for graph metadata: "When describing graph nodes in RDF/XML, it is recommended that the node be given an explicit resource URL, rather than using the anonymous node facilities. This makes it easier for other RDF documents to refer to the graph." The simulation metadata specification does not yet have wording like this, but it probably should. Best regards, Andrew _______________________________________________ cellml-discussion mailing list [email protected] http://www.cellml.org/mailman/listinfo/cellml-discussion
