Proposal for ISSUE-1

Holger,

We have ACTION-26.

I suggest that we accommodate your minimal proposed level of type
inferencing, as well as allowing people to specify matching base
rdf:type triples only.

As I understand it, you want RDFS subclass inferencing in two places.
1. Constrain the type of a value to a class.
2. Scope the application of a shape to a class.

Let's have the following two pairs of properties:

1. No inferencing - match rdf:type directly
sh:valueType
sh:scopeType

2. Follow rdfs:subClassOf triples (match the SPARQL property path
rdf:type/rdfs:subClassOf*)
sh:valueClass
sh:scopeClass

The benefit of this approach is that we avoid defining a new
entailment regime, and depending on the triple store to implement it.

In addition, I suggest we add a new metadata property sh:assumes on
Shapes to signal that the shape assumes an entailment regime. We
identify the entailment regimes using the IRIs defined by SPARQL 1.1.
[1]

[1] http://www.w3.org/TR/2013/REC-sparql11-entailment-20130321/

-- Arthur

Received on Wednesday, 20 May 2015 22:08:45 UTC