Re: on values

But how can the value of sh:targetNode be prohibited from being a blank node?
I gave an example where this is the case.

peter


On 12/03/2016 10:40 AM, Irene Polikoff wrote:
> The way I read it is as follows:
> 
> A value of a property could potentially be an IRI or a literal or a blank node. This is what follows from:
> 
>>>> The values of a property p for a node n in an RDF graph are the objects of
>>>> the triples in the graph that have n as subject and p as predicate.
> 
> The value of sh:targetNode is either an IRI or a literal.
> 
> Since the second statement “narrows the scope” from a more general definition of what a property value is to a more specific definition of what a value of sh:targetNode is, I do not see any contradictions. 
> 
> However, if this statement is not true and a value of sh:targetNode could be a blank node, I see your point.
> 
>> On Dec 3, 2016, at 1:13 PM, Peter F. Patel-Schneider <pfpschneider@gmail.com> wrote:
>>
>> "Each value of sh:targetNode is either an IRI or a literal." appears to be
>> quite clear, particularly with the definition of value in the doducment.  The
>> problem is that it is not the case that each value of sh:targetNode is either
>> an IRI or a literal.
>>
>> peter
>>
>>
>> On 12/03/2016 09:33 AM, Irene Polikoff wrote:
>>> Peter, I read the entire message first time you sent it.
>>>
>>> I didn't understand what problem you are seeing. This is why I asked for clarification.
>>>
>>> If the last two sentences don't reflect a problem by themselves, then is the problem in that the following doesn't make the subject clear?
>>>
>>> <Each value of sh:targetNode is either an IRI or a literal.>
>>>
>>> or something else?
>>>
>>> Sent from my iPhone
>>>
>>>> On Dec 3, 2016, at 11:44 AM, Peter F. Patel-Schneider <pfpschneider@gmail.com> wrote:
>>>>
>>>> The entire message reads:
>>>>
>>>>
>>>> ***************************
>>>>
>>>> There is a lot of wording like
>>>> Each value of sh:targetNode is either an IRI or a literal.
>>>> in Shapes Constraint Language (SHACL) W3C Editor's Draft 02 December 2016,
>>>> where the relevant definition of value in the document appears to be
>>>> The values of a property p for a node n in an RDF graph are the objects of
>>>> the triples in the graph that have n as subject and p as predicate.
>>>>
>>>> This statement is not universally true, such as in the RDF graph
>>>> _:a sh:targetNode _:b .
>>>>
>>>> Presumably the statement is meant to be interpreted in some context, but
>>>> there is no context given in the neighbourhood of the statement.
>>>>
>>>>
>>>> Peter F. Patel-Schneider
>>>> Nuance Communications
>>>>
>>>> ****************************
>>>>
>>>> The referent of the "This statement" is
>>>> Each value of sh:targetNode is either an IRI or a literal.
>>>> which is not true  in the RDF graph provided.
>>>>
>>>> peter
>>>>
>>>>
>>>>
>>>>> On 12/02/2016 11:02 PM, Irene Polikoff wrote:
>>>>> Peter, could you please explain why you say that the statement is not true.
>>>>>
>>>>> In your example, _:a is a subject (node n), sh:targetNode is a predicate
>>>>> (property p) and _:b is the object (the value).
>>>>>
>>>>>> On Dec 2, 2016, at 10:15 PM, Peter F. Patel-Schneider
>>>>>> <pfpschneider@gmail.com <mailto:pfpschneider@gmail.com>> wrote:
>>>>>>
>>>>>> The values of a property p for a node n in an RDF graph are the objects of
>>>>>> the triples in the graph that have n as subject and p as predicate.
>>>>>>
>>>>>> This statement is not universally true, such as in the RDF graph
>>>>>> _:a sh:targetNode _:b .
>>>>>
>

Received on Saturday, 3 December 2016 21:10:34 UTC