Re: Properties v classes in validation

On 9/4/2015 0:57, Karen Coyle wrote:
>
>
> On 9/2/15 11:02 PM, Holger Knublauch wrote:
>> On 9/3/15 3:52 PM, Karen Coyle wrote:
>>>
>>>
>>> On 8/31/15 10:47 PM, Holger Knublauch wrote:
>>>>
>>>> SHACL can certainly express all this, but maybe not with its Core
>>>> Vocabulary. It's still SHACL though.
>>>
>>> Note that this is included in Use Case #1 as a needed feature:
>>>
>>> "3.1 UC1: Model validation
>>>
>>> There is a general need to validate that the instance data matches the
>>> models that have been defined in RDFS or OWL. The primary validation
>>> requirement is to ensure that the appropriate information is given for
>>> each property (or class) in the model. As examples, one could require
>>> that each property must have a domain,
>>
>> rdf:Property
>>      a sh:ShapeClass ;
>>      sh:property [
>>          # Each property must have a domain
>>          sh:predicate rdfs:domain ;
>>          sh:minCount 1 ;
>>      ] .
>>
>>> or that classes must be explicitly stated in the instance data.
>>
>> ex:MyShape
>>      a sh:Shape ;
>>      sh:scope [ a sh:AllSubjectsScope ] ;
>>      sh:property [
>>          # all "instances" must have a type
>>          sh:predicate rdf:type ;
>>          sh:minCount 1 ;
>>      ] .
>>
>>> Input to this case is the RDF representation of an RDFS (or OWL)
>>> ontology.
>>>
>>> Summary: Requires the ability to check whether certain information is
>>> given/available for a property or class."
>>
>> These examples can already be covered. What else is missing?
>
> Cardinality for instances of a class is one. Dependencies (if instance 
> of classA then must/must not also exist instance of classB) See the DC 
> requirements here:
>
> http://wiki.dublincore.org/index.php/RDF_Application_Profiles/Requirements 
>
>
> And we have use cases that have further information which we will 
> develop as test cases for SHACL.

We had tried to give you some examples of how this can be expressed, e.g.

https://lists.w3.org/Archives/Public/public-data-shapes-wg/2015Sep/0005.html
https://lists.w3.org/Archives/Public/public-data-shapes-wg/2015Aug/0164.html

Instead of continuing with requirements expressed in (possibly 
ambiguous) English, maybe your group would like to try things out with 
the current SHACL draft and see how far you get. It would be best to 
have an RDF file where we can fill in the missing bits and thus identify 
the details of what appears to be missing.

>
>>
>>>
>>> And also that this is the second requirement that has been brought
>>> forward from the library/archive community as a strong requirement
>>> that seems to be dismissed even though we included it in the use cases
>>> for SHACL.
>>
>> Which one was the other?
>
> The issue brought up by Phil Archer:
>
> https://lists.w3.org/Archives/Public/public-data-shapes-wg/2015Aug/0007.html 
>

This had been recorded as

     http://www.w3.org/2014/data-shapes/track/issues/80

and will be discussed in due course.

Holger

Received on Thursday, 3 September 2015 22:30:51 UTC