Re: ISSUE-61 (was: Comments on SHACL, especially regarding compatibility with OSLC Resource Shapes 2.0)

Holger,

I disagree with you statement:
"direction of the current sh:nodeShape is already covered by rdf:type"

rdf:type relates a resource to a class.
sh:nodeShape relates a resource to a shape.

A shape is not in general a class. That's why we have sh:nodeShape.

However, sh:scopeNode, which relates a shape to a node, is simply the
inverse of sh:nodeShape. Given that we are allowing focus nodes to be
literal, then that leads us to using sh:scopeNode since RDF demands
that the subject of a statement be a non-literal. Someone pointed that
out at the 2015-10-22 telecon.

I am therefore +1 for sh:scopeNode. The spelling is consistent with
sh:scopeClass.

-- Arthur

On Mon, Oct 12, 2015 at 7:02 PM, Holger Knublauch
<holger@topquadrant.com> wrote:
> On 10/13/15 5:12 AM, Jim Amsden wrote:
>
> 2.1.1: I prefer the shape pointing to the thing it constrains - this keeps
> the constraints out of the vocabulary and allows the vocabulary to be easily
> reused in different contexts with different constraints for different
> purposes.
>
>
> This is an old and still unresolved ticket:
>
>     http://www.w3.org/2014/data-shapes/track/issues/61
>
> There are indeed arguments in both directions, and I could live with
> something like sh:scopeNode that points from a shape to a node because the
> direction of the current sh:nodeShape is already covered by rdf:type.
>
> Question: where would those triples live - would they still be in the data
> graph or in the shapes graph?
>
> Would be good to have this syntactic issue resolved before more and more
> people (and test cases etc) use our current sh:nodeShape vocabulary.
>
> Holger
>

Received on Thursday, 29 October 2015 17:04:28 UTC