Re: on the turtle serialization of SHACL

As you know very well, the formal resolutions are usually just a summary 
of the previous discussions. They cannot possibly repeat every little 
detail. See

https://lists.w3.org/Archives/Public/public-data-shapes-wg/2016Feb/0005.html

for an example to highlight that rdfs:domains and rdfs:ranges were 
discussed at the time.

Holger


On 15/12/2016 9:34, Peter F. Patel-Schneider wrote:
> It appears the rdfs:domain and rdfs:range were added to the Turtle document on
> 20 April 2016.  The Turtle document was not a part of the SHACL document as of
> the last public working draft of 14 August 2016.
>
> There was some discussion of domains and ranges in the file back in November
> 2015 but I cannot find any resolution on this point   There was a version of
> the document created by Arthur Ryman as a response to ACTION-31, but this
> version appears to have disappeared.
>
> The only relevant resolution I can find is
> 2015/11/19 - Close ISSUE-87 with two files: shacl-vocab.ttl and
> shacl.shacl.ttl as per Arthur Ryman's proposal
> http://www.w3.org/mid/CAApBiOn9eBvt99Eyu%253DjGUL9FxGHB%252B4r6%253DmPrUrwzCAHjmsQpSA%2540mail.gmail.com
> I don't see any reference to rdfs:domain or rdfs:range there.
>
>
> So where did the working group decide to include rdfs:domain and rdfs:range
> in the document?
>
>
> Peter F. Patel-Schneider
> Nuance Communications
>
>
> On 12/14/2016 02:44 PM, Holger Knublauch wrote:
>> The working group had already discussed this as part of ISSUE-87 in 2015 and
>> decided to include rdfs:range and rdfs:domain statements as suggested by
>> Arthur Ryman. Among others it simplifies interaction with editing tools that
>> populate input forms based on ranges and domains. According to your logic we
>> also couldn't include rdfs:subClassOf triples, because someone may expect us
>> to do RDFS inferencing.
>>
>> Holger
>>
>>
>> On 15/12/2016 8:14, Peter F. Patel-Schneider wrote:
>>> The problem is that readers may be misinformed into believing that this domain
>>> and range information actually relates to anything that happens in SHACL.
>>>
>>> The subclass information might have some relationship to what happens in
>>> SHACL, so there might be some utility to include it in an official SHACL
>>> document.  Since the domain and range information doesn't, including it in an
>>> official SHACL document seems to me to only be misleading.
>>>
>>> peter
>>>
>>>
>>> On 12/14/2016 01:34 PM, Irene Polikoff wrote:
>>>> Something like:
>>>>
>>>> sh:property rdfs:range sh:PropertyConstraint
>>>>
>>>> is simply a declaration.
>>>>
>>>> It does not require that inferencing must happen. Just like other published
>>>> vocabularies (e.g., SKOS) do not require their users to perform inferencing.
>>>>
>>>> Whether the reader of the document chooses to only be informed by these
>>>> statements or do inferencing is the readers choice not the publisher.
>>>>
>>>> So, I don’t really see what problem the domain and range statements in the
>>>> SHACL vocabulary would create.
>>>>
>>>> Irene
>>>>
>>>>> On Dec 14, 2016, at 10:47 AM, Peter F. Patel-Schneider
>>>>> <pfpschneider@gmail.com <mailto:pfpschneider@gmail.com>> wrote:
>>>>>
>>>>> However then the document has to reflect the actual
>>>>> situation with respect to the SHACL vocabulary.  This does not appear to be
>>>>> the case.  There are lots of occurrences of rdfs:domain and rdfs:range in the
>>>>> document.  As SHACL doesn't do RDFS reasoning these are only creating false
>>>>> impressions.
>>

Received on Wednesday, 14 December 2016 23:59:12 UTC