Re: shapes-ISSUE-199 (Constraint component IRIs): [Editorial] Constraint Component IRIs need to be defined before use [SHACL Spec]

On 19/11/2016 1:56, Karen Coyle wrote:
> No, Holger, I am saying the opposite. You cannot assume that the 
> person has read and fully understood sections 2 and 3. The spec needs 
> to function as a reference document, and people will enter into it at 
> what they think is the appropriate point without having perfect memory 
> of something they may have read at another time.

To me this is the reason why such documents contain hyperlinks. The 
first sentence of section 4 is

     This section defines the built-in SHACL Coreconstraint components 
<#dfn-constraint-component>that/must/be ...

Holger


>
> kc
>
> On 11/17/16 10:30 PM, Holger Knublauch wrote:
>> Hi Karen,
>>
>> I believe this is already sufficiently covered. Our assumption is that
>> in order to understand section 4 of SHACL, the reader has already been
>> through section 2 and 3. Search for the term
>> "sh:sourceConstraintComponent". In addition to various example Turtle
>> snippets, you will see
>>
>>
>>         3.4.5 Constraint Component (sh:sourceConstraintComponent)
>>
>> For validation results produced as a result of a constraint component
>> <#dfn-constraint-component>, the
>> property |sh:sourceConstraintComponent| must have as its object value
>> the IRI of the constraint component <#dfn-constraint-component> that
>> caused the result. For example, results produced due to a violation of a
>> constraint based on a value of |sh:minCount| would have the
>> value |sh:MinCountConstraintComponent|.
>>
>>
>> And in the intro to section 4 we already have
>>
>>
>>     4. Core Constraint Components
>>
>> This section defines the built-in SHACL Core constraint components
>> <#dfn-constraint-component> that /must/ be supported by all SHACL Core
>> processors. Each constraint component is identified by an IRI that is
>> referenced in the validation results via |sh:sourceConstraintComponent|.
>>
>>
>> I believe this should be sufficient - it's pretty much the sentence that
>> you want. Maybe you have reviewed an older version of the spec, or why
>> is the sentence above not sufficient?
>>
>> Holger
>>
>>
>> On 18/11/2016 8:48, RDF Data Shapes Working Group Issue Tracker wrote:
>>> shapes-ISSUE-199 (Constraint component IRIs): [Editorial] Constraint 
>>> Component IRIs need to be defined before use [SHACL Spec]
>>>
>>> http://www.w3.org/2014/data-shapes/track/issues/199
>>>
>>> Raised by: Karen Coyle
>>> On product: SHACL Spec
>>>
>>> The section on Core Constraint Components has a Constraint Component 
>>> IRI for each constraint component. These are shown in the section on 
>>> Validation reports, but not defined in the section where they 
>>> appear, which could be confusing. This could be simply a matter of 
>>> adding in introductory sentences of section 4: "Constraint component 
>>> IRIs specific to each constraint component are used in the 
>>> Validation report to identify the specific constraint that is being 
>>> reported." Or something like that.
>>>
>>>
>>>
>>>
>>>
>>
>