Re: rdfs:Graph ? comment on http://www.w3.org/TR/rdf11-concepts/#section-dataset and issue 35

To try and stay focused I will stick with test cases, first your two, and then one of mine:


On Sep 7, 2013, at 8:06 AM, Sandro Hawke <sandro@w3.org> wrote:

> TC1 - Does this dataset:
>   PREFIX : <http://example.org/#>
>   PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
>   :g1 :p 1.
>   GRAPH :g1 { :a :b :c }
>   GRAPH :g2 { :a :b :c }
> entail this:
>   :g2 :p 1.
> ?   For you, I expect it does not.

Your expectation is correct. Obviously if someone else wishes to draw that inference, and there may be some signal in the data, even in a standard conformant way, that licenses such an inference, … hmm I am not that happy, but not that unhappy either.


> 
> TC2 - Given this pair of datasets:
> D1:
>   PREFIX : <http://example.org/#>
>   PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
>   GRAPH :g1 { :a :b 1 }
> D2:
>   PREFIX : <http://example.org/#>
>   PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
>   GRAPH :g1 { :a :b 2 }
> Is the pair, taken together, inconsistent?    Or does the pair simply entail
>   GRAPH :g1 { :a :b 1,2 }
> ?

According to the current specs, interpretations are applied to graphs not datasets, so the notion of inconsistent or entail, as in your question is illformed.
But, there does seem to be a set of interpretations I1 consistent with the graph naming in D1 each of which maps :g1 to { :a :b 1 }, and a set of  interpretations I2 consistent with the graph naming in D2 each of which maps :g1 to  { :a :b 2 }, and the sets I1 and I2 are disjoint.

I would like some way of articulating that the semantics should use interpretations consistent with the graph naming.

Here is my test case:

TC3:

A dataset:

:g1 {
   :g1 rdf:type owl:Ontology;
         owl:imports :g2 .
}

:g2 {
   :g2 rdf:type owl:Ontology.
   :c   rdf:type owl:Class .
}


===

Question:
   in the ontology identified by :g1 is there a class identified by :c

The 'identified' concerning the ontology is as in:
http://www.w3.org/TR/2012/REC-owl2-syntax-20121211/#Ontology_IRI_and_Version_IRI

The 'identified' concerning the class is as in:
http://www.w3.org/TR/2012/REC-owl2-syntax-20121211/#Classes

===

Question:
  if the answer to the previous question is that there is insufficient information, what additional information can I add to the dataset to allow the answer to be yes?
[for example, I can add  :c   rdf:type owl:Class . directly to the graph :g1]


==========

My preferred resolution to this test case, is "yes"; however, applications may choose to give a warning if the ontology retrievable from the URL :g2, is not the same as that given in the dataset. (There are a load of issues there, which we should acknowledge but not solve)

Jeremy