Re: on the turtle serialization of SHACL

To return to Peter’s original point, I feel that this particularly comment is somewhat a blast from the past, and not really very helpful. (Unlike many of his comments which I feel are useful critiques).

I think some of the underlying issues Peter is getting at were answered long ago by:

https://www.w3.org/TR/webarch/#namespace-document <https://www.w3.org/TR/webarch/#namespace-document>


"When a namespace URI owner provides a namespace document, it is authoritative for the namespace.”
which I believe is the practice the WG is following. And the nature of “authoritative” is somewhat awkward, as Peter points out.

In particular, there is the possibility of conflict between the ‘authoritative’ namespace document and other authoritative documents; and as in all human endeavors when there is conflict over the rules, there will be lots of discussion.

But this situation is no different than that where a single normative recommendation document is internal inconsistent (I suspect most are). This does not license: ex contradictio quodlibet, but rather suggests an official comment on the public mailing list pointing out the problem, and some solution, eventually. Peter is an expert at this process of public comment and resolution.

I believe that Peter already knows the answers to the specific questions he asks, and I believe the intended readership of these documents is sufficiently skilled to also answer them.

Just for a laugh, ….

> 
> In what sense is this document normative?

In the sense of the Architecture of the World Wide Web

> 
> Would removing the line "rdfs:subClassOf sh:Constraint ;" from the part of
> the document about sh:Shape change anything about SHACL?  


It would reduce the clarity of the following part of the specification:
"Shapes <https://www.w3.org/TR/shacl/#dfn-shape> are SHACL instances <https://www.w3.org/TR/shacl/#dfn-shacl-instance> of sh:Shape and define constraints <https://www.w3.org/TR/shacl/#dfn-constraint>”
and
"sh:Constraint is the SHACL superclass <https://www.w3.org/TR/shacl/#dfn-shacl-superclass> of all constraint types.”

I believe that together this two sentences imply

sh:Shape rdfs:subClassOf sh:Constraint 

but personally I find the triple clearer.

> Would adding
> "sh:PropertyConstraint rdfs:subClassOf sh:Shape." somewhere to the document
> change anything about SHACL?  

I don’t know - it seems to me that this is not within the normal meaning of the recommendation document, nor a licensed inference from the namespace document: so my guess would be yes. However, I am a lot less informed about either than others on this list.


> Would removing "rdfs:domain sh:Shape ;" from
> the part of the document about sh:property change anything aobut SHACL?

No, because I read " Property constraints <https://www.w3.org/TR/shacl/#dfn-property-constraints> are linked from a shape <https://www.w3.org/TR/shacl/#dfn-shape> with the property sh:property. “

> Would changing "owl:" to "rowl:" throughout the document change anything
> about SHACL?  

No: Namespace IRIs and namespace prefixes are not a formal part of the RDF data model. as per https://www.w3.org/TR/2014/REC-rdf11-concepts-20140225/ <https://www.w3.org/TR/2014/REC-rdf11-concepts-20140225/>
> Would changing the document in any way change SHACL?

Yes - the "sh:PropertyConstraint rdfs:subClassOf sh:Shape.” example you gave may be one.

It would also be possible to add axioms that would, for example, make some class a subclass of a has value restriction on sh:property to make large classes of resources be covered by other parts of the shape specification.
Since this does not seem to be in accord with the intent of the rest of the specification, I would hope that the observant reader (such as Peter) would draw attention to this and ask for clarification: either by changing the namespace document, or by changing the specification document.

Personally I believe this e-mail describes normal practice at the W3C and this particular comment from Peter could be adequately addressed with no changes at all.


Jeremy


> On Dec 13, 2016, at 3:50 PM, Peter F. Patel-Schneider <pfpschneider@gmail.com> wrote:
> 
> The current version of the SHACL document contains "The Turtle serialization
> of the SHACL vocabulary is part of the normative specification. However, the
> values of rdfs:label and rdfs:comment in that file are not normative.",
> pointing to a Turtle document available on the web.
> 
> In what sense is this document normative?
> 
> Would removing the line "rdfs:subClassOf sh:Constraint ;" from the part of
> the document about sh:Shape change anything about SHACL?  Would adding
> "sh:PropertyConstraint rdfs:subClassOf sh:Shape." somewhere to the document
> change anything about SHACL?  Would removing "rdfs:domain sh:Shape ;" from
> the part of the document about sh:property change anything aobut SHACL?
> Would changing "owl:" to "rowl:" throughout the document change anything
> about SHACL?  Would changing the document in any way change SHACL?
> 
> 
> Peter F. Patel-Schneider
> Nuance Communications
>