Hi,
what I intended as a starting point for validation levels is to tie a property like spin:validationLevel (we can change the name of the property later on as there is no property with this name within the spin namespace).
When a constraint violation is raised, you will also get the information which validation level has been violated.
Further, you will be able to sort constraint violation according to different validation levels
or you will be able to validate only constraints of a certain validation level.
If we document the validation level for each requirements representing a particular constraint in our database, this information won't be implementation dependent.
Violation triples would look like this:
CONSTRUCT {
_:violation
a spin:ConstraintViolation ;
rdfs:label ?violationMessage ;
spin:violationRoot ?violationRoot ;
spin:violationPath ?violationPath ;
spin:violationSource ?violationSource ;
spin:fix ?violationFix ;
:validationLevel :error }
Cheers,
Thomas
--
Thomas Bosch, M.Sc. (TUM)
PhD Student
GESIS - Leibniz Institute for the Social Sciences
Social Science Metadata Standards
Visitors Address: B2,1, D-68159 Mannheim
Postal Address: P.O.Box 12 21 55, D-68072 Mannheim
Tel: + 49 (0) 621 / 1246-271
Fax: + 49 (0) 621 / 1246-100
Web: http://www.gesis.org
Website: http://boschthomas.blogspot.com/
GitHub: https://github.com/boschthomas/PhD
________________________________________
Von: DCMI Architecture Forum [[log in to unmask]]" im Auftrag von "Karen Coyle [[log in to unmask]]
Gesendet: Donnerstag, 6. November 2014 23:09
An: [log in to unmask]
Betreff: Re: [RDF AP] DSP 2 - validation levels
Eric, yes, I am sure that some of our use cases do require different
schemas for different situations -- in particular DPLA use case has a
very loose schema (as I read it) for raw input, and a different schema
to be applied after their programs have enhanced the data. We should
consider this "different schemas" option to see if it works with all of
our cases, or if something else is needed.
kc
On 11/6/14 1:24 PM, Eric Prud'hommeaux wrote:
> * Karen Coyle <[log in to unmask]> [2014-11-06 11:18-0800]
>> Thomas, we definitely want validation levels. However, I suspect
>> that those will be implementation dependent and therefore the DSP
>> (or whatever comes out) will need to be able to include them. But I
>> don't think we can tie them to constraints themselves. Or is that
>> not what you meant?
>
> One strategy for implementing multiple validation levels over a
> single-level system is to have multiple schemas with different degrees
> of pickiness. Another is to have user-defined messages associated with
> the constraints and simply caveat the way the messages are shown
> (though this can be complex if they trip other errors). A third might
> be to use an extensibily mechnaism.
>
> I don't know whether the WG will design for multiple error levels but
> ideally if they don't, you can test a work-around which we can share
> with others who want multiple levels.
>
>
>> kc
>>
>> On 11/6/14 8:43 AM, Bosch, Thomas wrote:
>>> Hi,
>>>
>>> I propose to define valdation levels for each constraint we want to
>>> define for DSP 2.
>>> That's exactly what have been discussed during the RDF validation
>>> workshop at the DC conference.
>>>
>>> For example, this constraint causes an error, that constraint causes
>>> only warning or even a fail (see requirement
>>> R-205-VARYING-LEVELS-OF-ERROR).
>>>
>>> This requirement will definitely also be part of the W3C requirements,
>>> we could offer a reference implementation ;-)
>>>
>>> When a constraint violation is raised you can immediately see the
>>> severity of the violation.
>>>
>>> That is what we could do now when Karen offers first test cases.
>>>
>>> Just an idea...
>>>
>>>
>>> Thomas
>>>
>>> --
>>>
>>> Thomas Bosch, M.Sc. (TUM)
>>>
>>> PhD Student
>>>
>>> GESIS - Leibniz Institute for the Social Sciences
>>>
>>> Social Science Metadata Standards
>>>
>>> Visitors Address: B2,1, D-68159 Mannheim
>>>
>>> Postal Address: P.O.Box 12 21 55, D-68072 Mannheim
>>>
>>> Tel: + 49 (0) 621 / 1246-271
>>>
>>> Fax: + 49 (0) 621 / 1246-100
>>>
>>> Web: http://www.gesis.org
>>>
>>> Website: http://boschthomas.blogspot.com/
>>> GitHub: _https://github.com/boschthomas/PhD_
>>>
>>
>> --
>> Karen Coyle
>> [log in to unmask] http://kcoyle.net
>> m: 1-510-435-8234
>> skype: kcoylenet/+1-510-984-3600
>
--
Karen Coyle
[log in to unmask] http://kcoyle.net
m: 1-510-435-8234
skype: kcoylenet/+1-510-984-3600
|