Hi Nikolaos,

this is certainly an interesting question, yet I don't see how to solve this. As soon as inferencing is activated, and you have a class hierarchy such as

rdfs:Resource
    ex:RootClass   rule1
        ex:ChildClass   rule2 overrides rule1

and ex:Instance a ex:ChildClass, then an inference engine would add ex:Instance a ex:RootClass, and things become really messy to check. I believe rules and constraints should only ever narrow down the semantics.

It would help me to see a real-world example of such scenarios. Maybe there are work-arounds using multiple inheritance/multiple rdf:types.

BTW this is an interesting discussion in regards to the ongoing W3C Shapes Working Group, that quite possibly creates something close to a "SPIN 2.0", called SHACL [1]. In that spec, take a look at the proposed sh:scope attribute that would allow you to place pre-conditions in front of constraints.

Thanks
Holger

[1] https://w3c.github.io/data-shapes/shacl/


On 4/3/15 11:11 PM, Nikolaos Beredimas wrote:
This feels very close to what I was thinking, because it fits very well with my use case.

However, the question remains (because one can start imagining a lot of use cases):
Wouldn't it be more powerful/beautiful if it could be something like

:myClass spin:ignore :rule1 (eg. just for :myClass, but not its subclasses)

or

:myClass spin:retract :rule1 (for :myClass and all its subclasses)

or

:myClass spin:override [ :rule2 spin:replaces :rule1] (override/replace rule for :myClass and subclasses)

or
:myClass spin:extends [ :sparql_fragment spin:optional :rule1] (and a OPTIONAL graph matching pattern to the to the sparql query that gets executed)

and for all these, the SPIN engine (and not the developer) would be responsible to automatically concstruct/decide the new sparql queries to be executed.

I understand, this would not be a trivial feature to add, but maybe I could persuade you :)

On Fri, Apr 3, 2015 at 3:49 PM, Scott Henninger <[email protected] <mailto:[email protected]>> wrote:

    OK, try this instead.  Add a property, let's name it
    :doNotProcessRule.  For the classes you do not want to be
    processed, add (one ore more) rule name, such as "rule1". The use
    the following graph pattern.

    WHERE {
        ?this ...
        OPTIONAL {
            ?this a ?cls .
            ?cls :doNotProcessRule "rule1" .
        } .
        FILTER (!bound(?cls)) .
    }

    If a class has a value of "rule1" for :doNotProcessRule, its
    instances will not be processed by the SPIN rule.  Of course you
    can have values other than "rule1", but it's just a matter of
    making a match on a class property to tell the reasoner to skip
    its instances.


    Basically, SPARQL, and maybe a bit of modeling provides a great
    deal of flexibility for controlling SPIN execution.

    -- Scott


    On 4/3/2015 8:16 AM, Nikolaos Beredimas wrote:
    Yes I understand that.
    But anything like this feels like a custom, one-off, solution,
    and that entais some disadvantages.
    For example, it ties the SPIN rule to the specific ontology. In
    another ontology, you'll have to reimplement the rule.
    It, also, ties this information to the rule definition, when in
    reality it should belong to the class definition.
    Inferencing also comes into play and you have to account for that
    too.

    That's why I was asking for a generic, SPIN defined solution.
    Judging from the answer I understand there is probably not a way
    to achieve this in a generic, ontology-independent, way, at least
    currently

    Do you think this is a feature that we could hope for in a newer
    SPIN revision?
    In a nutshell, it would extend the object-oriented notion of SPIN
    rules/constraints inheritance with the ability to
    override/shadow/retract specific rules/constraints for given
    subclass(es)


    Τη Παρασκευή, 3 Απριλίου 2015 - 2:41:56 μ.μ. UTC+3, ο χρήστης
    Scott Henninger έγραψε:

        Nikolaos; One approach would be to FILTER out the class
        instances. I.e.:

        { ?this :prop ?value .
          ...
          ?undesiredInstance a :SubClass1 .
          FILTER (?this != ?undesiredInstance)
        }

        -- Scott

        Sent from my iPhone

        On Apr 3, 2015, at 7:05 AM, Nikolaos Beredimas
        <[email protected]> wrote:

        Quoting from http://www.w3.org/Submission/spin-modeling/ :

            In other words, SPIN class descriptors can only "narrow
            down" and further restrict what has been defined further
            up in the class hierarchy. In this spirit,
            /global/ class descriptions are those that are attached
            to the root class |rdfs:Resource| or its OWL equivalent
            |owl:Thing|.



        So if you define a rule/constraint/... for owl:Thing, it is
        a global rule. It applies to every subclass of owl:Thing
        automatically.

        I have come across some use cases where this isn't the
        desirable behavior. I was wondering, therefore, if there is
        a stanard, SPIN way, of annotating a class, or the top class
        of a sub-hierarchy of classes, to exclude a specific
        rule/constraint or a collection of rules/constraints from
        applying to this class/classes.
        Ofc, I could edit said class and annotate it with a custom
        property, and edit the spin rule/constraint accordingly to
        check for that annotation
        or, I could edit the classes hierarchy and move said class
        to a different hierarchy
        or any other custom solution.

        But I was wondering if there was a SPIN specified way of
        doing this.

        TL;DR
        How to exclude specified SPIN rules from applying to
        specified subclasses, using only the SPIN vocabulary

        PS. Apologies if the answer is too obvious
-- You received this message because you are subscribed to the
        Google Group "TopBraid Suite Users", the topics of which
        include Enterprise Vocabulary Network (EVN), Reference Data
        Manager (RDM), TopBraid Composer, TopBraid Live, TopBraid
        Insight, SPARQLMotion, SPARQL Web Pages and SPIN.
        To post to this group, send email to [email protected]
        ---
        You received this message because you are subscribed to the
        Google Groups "TopBraid Suite Users" group.
        To unsubscribe from this group and stop receiving emails
        from it, send an email to [email protected].
        For more options, visit https://groups.google.com/d/optout.

-- You received this message because you are subscribed to the
    Google Group "TopBraid Suite Users", the topics of which include
    Enterprise Vocabulary Network (EVN), Reference Data Manager
    (RDM), TopBraid Composer, TopBraid Live, TopBraid Insight,
    SPARQLMotion, SPARQL Web Pages and SPIN.
    To post to this group, send email to
    [email protected]
    <mailto:[email protected]>
    ---
    You received this message because you are subscribed to the
    Google Groups "TopBraid Suite Users" group.
    To unsubscribe from this group and stop receiving emails from it,
    send an email to [email protected]
    <mailto:[email protected]>.
    For more options, visit https://groups.google.com/d/optout.

-- You received this message because you are subscribed to the Google
    Group "TopBraid Suite Users", the topics of which include
    Enterprise Vocabulary Network (EVN), Reference Data Manager (RDM),
    TopBraid Composer, TopBraid Live, TopBraid Insight, SPARQLMotion,
    SPARQL Web Pages and SPIN.
    To post to this group, send email to
    [email protected]
    <mailto:[email protected]>
    ---
    You received this message because you are subscribed to a topic in
    the Google Groups "TopBraid Suite Users" group.
    To unsubscribe from this topic, visit
    https://groups.google.com/d/topic/topbraid-users/yxgHQe0P5Fk/unsubscribe.
    To unsubscribe from this group and all its topics, send an email
    to [email protected]
    <mailto:[email protected]>.
    For more options, visit https://groups.google.com/d/optout.


--
You received this message because you are subscribed to the Google Group "TopBraid Suite Users", the topics of which include Enterprise Vocabulary Network (EVN), Reference Data Manager (RDM), TopBraid Composer, TopBraid Live, TopBraid Insight, SPARQLMotion, SPARQL Web Pages and SPIN.
To post to this group, send email to [email protected]
---
You received this message because you are subscribed to the Google Groups "TopBraid Suite Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected] <mailto:[email protected]>.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Group "TopBraid 
Suite Users", the topics of which include Enterprise Vocabulary Network (EVN), 
Reference Data Manager (RDM), TopBraid Composer, TopBraid Live, TopBraid Insight, 
SPARQLMotion, SPARQL Web Pages and SPIN.
To post to this group, send email to [email protected]
--- You received this message because you are subscribed to the Google Groups "TopBraid Suite Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to