PROV-ISSUE-583 (equivalent-instances-in-bundles): Questions concerning what it means for applications to treat equivalent instances 'in the same way', particularly in bundles. [prov-dm-constraints]

PROV-ISSUE-583 (equivalent-instances-in-bundles): Questions concerning what it means for applications to treat equivalent instances 'in the same way', particularly in bundles.  [prov-dm-constraints]

http://www.w3.org/2011/prov/track/issues/583

Raised by: James Cheney
On product: prov-dm-constraints

A sub-issue of ISSUE-576.

>From Antoine Zimmermann's email:
http://lists.w3.org/Archives/Public/public-prov-comments/2012Oct/0004.html



"""
Applications [...] should treat equivalent instances or documents in the same way.
"""

Does it mean that an instance in a bundle can be safely replaced by another equivalent instance inside the bundle?  This is important because it may be understood that a bundle identifier identifies the exact set of PROV statements given in a PROV document, or that the bundle identifier simply identifies the logical content of the bundle, or even any content that logically "implies" the content of the bundle in the said document.
The definitions of equivalence and inference suggests that it is the later option. Note that this is something that may have to be understood when writing PROV documents as RDF datasets, given that RDF 1.1 is not specifying any particular relationship between a graph IRI and the content of a named graph.

Received on Thursday, 25 October 2012 16:23:17 UTC