Participants

Corey, Michael, Daniel, Kai

Minutes

[17:14] * chrpr_ helping scribe
[17:15] <@chrpr_> Architecture list is high traffic
[17:15] <@chrpr_> Starting DCAM Revision
[17:15] <@chrpr_> Determined need for issue tracker
[17:15] <@chrpr_> Kai's taken steps to rethink DCAM in RDF as something similar to SKOS – an added layer. RDF can be used to represent many DCAM constructs.
[17:16] <@chrpr_> Can do most everything w/ RDF, but not easy to relate to for non semweb geeks.
[17:18] <dgarijo> http://wiki.dublincore.org/index.php/DCAM_Revision_Tech
[17:19] <@chrpr_> Kai's architecture revisions & edits will reflect goals of DC Provenance TG
[17:21] <@chrpr_> Wanting description set to be a class so that statements can be assigned to it.
[17:23] <@chrpr_> Michael: DCAM is a recipe / cookbook, or a rigorous semantic document
[17:23] <@chrpr_> With cookbook – can connect to RDF more easily and be flexible.
[17:25] <@chrpr_> TBL Messages about what happens when you de-ref a named graph.
[17:25] <@chrpr_> Start of thread: http://lists.w3.org/Archives/Public/public-lod/2012Jan/0006.html
[17:25] <@chrpr_> TBL's messages:
[17:25] <@chrpr_> http://lists.w3.org/Archives/Public/public-lod/2012Jan/0018.html
[17:25] <@chrpr_> http://lists.w3.org/Archives/Public/public-lod/2012Jan/0026.html
[17:26] <@chrpr_> This could be easier with Desc. Sets than Named Graphs.
[17:26] <@chrpr_> Making this a more rigourous semantics doc – Micheal's point that we'd be recreating RDFs & OWL, which none of us have the formal logic training to do successfully.
[17:31] <@chrpr_> Description set contains a thing, that thing is a statement, it becomes *very* hard to say this in RDF.
[17:36] <@chrpr_> Having to wait for RDF1.1 / RDFnext to handle relationships–probably similar to TRIG.
[17:36] <@chrpr_> Open question whether this will help DCAM.
[17:38] <@chrpr_> Kai – Doing this without RDFnex. Strict RDF-DCAM combined with LOD Best Practices.
[17:38] <@chrpr_> A description set is a URI. Resolving that URI directs you to an RDF representation of all included statements.
[17:38] <dgarijo> that sounds good to me
[17:54] <@chrpr_> My question about relation to w3c prov. They're looking the something provenance of anything.
[17:55] <@chrpr_> DCprov isn't in their use case, so the RDF provenance discussion isn't in their use cases.
[17:56] <@chrpr_> DCAM & dc prov has a close relationship to RDF Next Steps group, less so with w3c prov.