Re: A proposal for entailment tests

On Friday, September 21, 2001, at 07:14  PM, 
jos.deroo.jd@belgium.agfa.com wrote:

>   [ tc:graph g1, g2, g3 ] tc:entailrdf [ tc:graph g4 ].
> describes the RDF entailment of the graph g4 given the
> graphs g1, g2 and g3

Since the set of graphs on both sides of the entailment is 
closed, would not a list be more appropriate for enumerating 
them?

  [ a tc:Graph ; tc:contents (<g1> <g2> <g3>) ]  tc:entailRDF
  [ a tc:Graph ; tc:contents (<g4>) ] .

> and
>   [ tc:graph g1, g2, g3 ] tc:rdfsentail [ tc:graph g4 ].

You use both entailrdf and then rdfsentail. I'd suggest 
mtrdf:entails and mtrdfs:entails to make the separation clearer, 
and also keep model theory statements out of the test cases 
schema.

> The syntax testcases can be described in the same
> manifest file as
> e.g.
>   [ tc:graph g1 ] tc:entailrdf [ tc:graph g2 ].
>   [ tc:graph g2 ] tc:entailrdf [ tc:graph g1 ].

Hmm, this seems less-than-adequate to me because it does not 
represent the key to these test cases, namely that they are in 
different formats... Perhaps something like:

[ a tc:Graph ; tc:file <g1.rdf> ; tc:format tc:RDF-XML ]
   tc:equivalent
[ a tc:Graph ; tc:file <g1.nt>  ; tc:format tc:N-Triples ] .

--
       "Aaron Swartz"      |           Blogspace
  <mailto:me@aaronsw.com>  |  <http://blogspace.com/about/>
<http://www.aaronsw.com/> |     weaving the two-way web