Re: Re: using classes to control constraints

> Michel's request is going farther in saying that if a range of
> property values in a constraint is defined as members of a superclass,
> data that uses members of its subclasses for the property values
> should also pass validation.

ah, I didn't get that.. thanks for clarifying!

simon

---
DDipl.-Ing. Simon Steyskal
Institute for Information Business, WU Vienna

www: http://www.steyskal.info/  twitter: @simonsteys---
DDipl.-Ing. Simon Steyskal
Institute for Information Business, WU Vienna

www: http://www.steyskal.info/  twitter: @simonsteys



Am 2015-02-12 18:09, schrieb Irene Polikoff:
> I don't think so.
> 
> Everything said so far, confirms the view that constraints stated on a
> superclass must hold on subclasses.
> 
> Michel's request is going farther in saying that if a range of
> property values in a constraint is defined as members of a superclass,
> data that uses members of its subclasses for the property values
> should also pass validation.
> 
> On Thu, Feb 12, 2015 at 12:40 AM, Simon Steyskal <ssteyska@wu.ac.at>
> wrote:
> 
>> Hi!
>> 
>> I guess that would contradict the general notion of inheritance,
>> i.e. all constraints stated on a superclass must hold on its
>> subclasses too and there should be no constraints that are not
>> passed to a superclass' children.
>> 
>> simon
>> 
>> -------- Ursprüngliche Nachricht --------
>> Von: Michel Dumontier <michel.dumontier@gmail.com>
>> Datum: 12.02.2015 05:54 (GMT+01:00)
>> An: Irene Polikoff <irene@topquadrant.com>
>> Cc: "Peter F. Patel-Schneider" <pfpschneider@gmail.com>, Holger
>> Knublauch <holger@topquadrant.com>, public-data-shapes-wg@w3.org
>> Betreff: Re: using classes to control constraints
>> 
>> Hi,
>> 
>> i would like to have shapes to be compatible with OWL entailment.
>> For instance, if I place a superclass in a constraint, i would like
>> to validate positive where i have a subclass in the data. But I see
>> that as a choice that should be specified with the shape, as I could
>> imagine that you might also want to validate with only the specified
>> class.
>> 
>> m
>> 
>> On Feb 11, 2015, at 8:16 PM, Irene Polikoff <irene@topquadrant.com>
>> wrote:
>> 
>> There is no interaction with entailment or querying. The data is
>> what it is.
>> 
>> Constraints describe what the data should be in order to pass the
>> validation. They are used to validate the data that is available.
>> They don't change the data.
>> 
>> On Wed, Feb 11, 2015 at 7:38 PM, Peter F. Patel-Schneider
>> <pfpschneider@gmail.com> wrote:
>> -----BEGIN PGP SIGNED MESSAGE-----
>> Hash: SHA1
>> 
>> On 02/11/2015 04:16 PM, Irene Polikoff wrote:
>>> When is it supposed to be checked?
>>> 
>>> When constraint checking/data validation is invoked
>> 
>> Only then? What is the interaction with entailment? And querying?
>> 
>>> What reporting needs to be done?
>>> 
>>> As I recall, there has been a discussion about what should be
>> returned
>>> and a few people provided examples of the kind of reporting they
>> want. It
>>> has been captured in the LDOM document.
>> 
>> That was for explicit invocation of validation. If type
>> assertions can be
>> made to shapes then I think that much more needs to be done.
>> 
>>> Why are you asking?
>> 
>> Because, explicit typing to shapes needs to be integrated into the
>> rest of
>> RDF, RDFS, and SPARQL.
>> 
>> peter
>> 
>>> 
>>> On Wed, Feb 11, 2015 at 4:17 PM, Peter F. Patel-Schneider
>>> <pfpschneider@gmail.com <mailto:pfpschneider@gmail.com>> wrote:
>>> 
>>> So there is in an error in an RDF graph. How is that supposed to
>> work?
>>> When is it supposed to be checked? What reporting needs to be
>> done?
>>> 
>>> peter
>>> 
>>> 
>>> On 02/11/2015 01:08 PM, Irene Polikoff wrote:
>>>> It is intended for validation and works over data that exists.
>> So, if
>>>> ex:a is not ex:p ex:q, there is an error.
>>> 
>>>> On Wed, Feb 11, 2015 at 1:49 PM, Peter F. Patel-Schneider
>>>> <pfpschneider@gmail.com <mailto:pfpschneider@gmail.com>
>>> <mailto:pfpschneider@gmail.com <mailto:pfpschneider@gmail.com>>>
>> wrote:
>>> 
>>> 
>>> 
>>>> On 02/11/2015 10:42 AM, Irene Polikoff wrote:
>>>>> <What would it mean to assert that an object belongs to a shape
>> via
>>>>> an rdf:type link?>
>>> 
>>>>> I believe it would mean that constraints defined for the shape
>> apply
>>>>> to the object.
>>> 
>>> 
>>>> So I can infer things from this assertion? For example, if
>> ex:shape
>>>> requires that the value of ex:p be ex:q then does ex:a rdf:type
>>>> ex:shape . imply ex:a ex:p ex:q .
>>> 
>>>> peter
>>> 
>>> 
>>> 
>>> 
>>> 
>> -----BEGIN PGP SIGNATURE-----
>> Version: GnuPG v1
>> 
>> iQEcBAEBAgAGBQJU2/X1AAoJECjN6+QThfjztwYH/RGKnTIPGMQrrJb9OtignRAT
>> MNcGm2fkh39D8IpUkoE85JAKzG9NJcvdI74748JJppdUnrJPbCwXWlX9HnNDNOW4
>> lbgTK8Y3eiDr7liavMsK+7ZbuF/QAocAXaWU9dPbwdrCXHFY1jmfY6y1H0KlfvST
>> vvyAh12zhzHFxgksALkxKEvnSaGL6rHlZUoNh6Ke/8gZKn5Z2B0yQJZvkJdVU5sa
>> j1P/BrzLd5QNIUgiSQJklQecXN8sTZt5Cd96ePGlGD6hn9aLnVUKgbNH5BvpMchw
>> z51tUAaXAQFK1RtoRec+PYiJxaXRQ3UK3ZZQ1JsWSIq5350vx16j7jXgyc5+4eg=
>> =8Iey
>> -----END PGP SIGNATURE-----

Received on Thursday, 12 February 2015 18:21:59 UTC