Hi Karen,
Excuse me if I came across as condescending, it was not my intent. I am truly puzzled by those things actually. I do understand the wish to separate constraints from axioms. But I am afraid of the end result if we make this separation. What if we end up with a constraint language (say DSP) that has constraint which can also be expressed in the ontology language (OWL).
Continuing on class disjointness: we certainly want it as a constraint, it's already in our requirements. Should we refrain from putting it in the constraint language, because it's in the ontology language? Or should we try to separate disjointness-constraint from disjointness-axiom?
I'm not trying to outsmart you or whatever. I am really trying to figure out where we will end up because some ontology languages may have gone only half way in the direction of data validation.
We could consider 'constraints' to be connected to closed world assumption, and 'axioms' to the open world one. But I'm really not sure this is a safe path either. The consistency checks in description logics enable some data verification at a level that can considered to be 'structural', with all due quotes.
And here comes my 'explanation' sentence you've quoted. It's extremely difficult to get to a shared understanding of all this, without spending much time discussing. The matter is hard and I'm not even sure of what *I* mean for a couple these words.
As you noted, OWL2 refrains from using some words, and keeps to technical jargon. This is wise from the perspective of OWL2 designers; but that doesn't help the people like us who have to align the technical layer to a more 'business-oriented' one.
(and the picture would become even harder for separating 'ontologies' from other things, if we consider that 'ontologies' in the OWL sense can define data for individuals, i.e. at exactly the same level as what I think are 'descriptions' in DSP)
Antoine
On 10/1/14 11:12 PM, Karen Coyle wrote:
> On 10/1/14, 1:53 PM, Antoine Isaac wrote:
>> Hi Karen,
>
>>
>> I cannot resist sending you another gem of OWL documentation:
>> http://www.w3.org/TR/2012/REC-owl2-rdf-based-semantics-20121211/#Satisfaction.2C_Consistency_and_Entailment
>>
>> I'm going on holidays so won't have time to explain you, but I could, in
>> theory ;-) (I've been trained to understand that formal stuff, a while ago)
>
> Yes, Antoine, I'm sure that I could not understand it without you. :-(
>
> Please try to be less condescending -- I'm not some naive who needs explanations like a child. I understand RDF and OWL as well as most. And this is NOT a contest that one must win -- this is an important discussion about the direction of the group, and I feel strongly about the separation of semantics and constraints.
>
> We must talk separately about constraints and axioms, and be clear that constraining is quite different from inferring. And it is the case that although the word "constraint" may occasionally occur in OWL2 documentation, the language that was once used as "OWL constraints" is now "OWL axioms." We should understand and follow that because the change was made deliberately because of the confusion that was caused by the use of the term "constraint". Even "disjoint" is now defined as an axiom ("the disjointness axiom"). I don't understand your insistence on using "constraint" in a way that 1) it is not used in the current documentation 2) that you brought up as being confusing. Since you think we are confusing things, wouldn't it be better to be clear in the language we use? Your use of "semantic constraints" seems to me to be exactly the sense that OWL2 is trying to avoid.
>
> Oh, yeah, have a good vacation.
>
> kc
>
>
>>
>> Antoine
>>
>> On 10/1/14 8:18 PM, Karen Coyle wrote:
>>> On 10/1/14, 10:32 AM, Antoine Isaac wrote:
>>>> Hi,
>>>>
>>>> We're having serious terminology problems here.
>>>> We can't ask for separating 'constraints' from 'ontology semantics'
>>>> because ontologies (in the regular OWL / Semantic Web sense) may include
>>>> constraints.
>>>
>>> Although the original OWL documentation refered to constraints, it is
>>> that which is a misnomer, and in the OWL2 documentation the term
>>> "axioms" is used [1] for this reason. So we should make sure that we
>>> define our own terminology well, and I would suggest following OWL2
>>> and referring to a formal language for inferencing as axioms.
>>>
>>> The OWL2 primer states:
>>>
>>> "OWL 2 is not a schema language for syntax conformance. Unlike XML,
>>> OWL 2 does not provide elaborate means to prescribe how a document
>>> should be structured syntactically. In particular, there is no way to
>>> enforce that a certain piece of information (like the social security
>>> number of a person) has to be syntactically present. This should be
>>> kept in mind as OWL has some features that a user might misinterpret
>>> this way."
>>>
>>> It also says, in a later section:
>>>
>>> "This is a commonly modeling error: note that a domain (or range)
>>> statement is not a constraint on the knowledge, but allows a reasoner
>>> to infer further knowledge." [2}
>>>
>>> It gives three basic notions:
>>>
>>> Axioms: the basic statements that an OWL ontology expresses
>>> Entities: elements used to refer to real-world objects
>>> Expressions: combinations of entities to form complex
>>> descriptions from basic ones
>>>
>>> Constraints, as named in OWL1, are gone.
>>>
>>> The Stanford document you cite is unfortunately from 1992.
>>>
>>> kc
>>>
>>> [1] http://www.w3.org/TR/2012/REC-owl2-mapping-to-rdf-20121211/
>>> [2] http://www.w3.org/TR/owl2-primer/#What_is_OWL_2.3F
>>>
>>>> I've already said that OWL include constraints.
>>>> Looking at the first definition available for 'ontology' already brings
>>>> "formal axioms that constrain the interpretation and well-formed use of
>>>> these terms"
>>>> http://www-ksl.stanford.edu/kst/what-is-an-ontology.html
>>>>
>>>> Best,
>>>>
>>>> Antoine
>>>>
>>>> On 10/1/14 6:23 PM, Karen Coyle wrote:
>>>>> On 10/1/14, 4:21 AM, Bosch, Thomas wrote:
>>>>>> Hi Tom,
>>>>>>
>>>>>> there has been a lot of discussion of this topic on the W3C RDF
>>>>>> Validation mailing list, with lot's of pros and cons using OWL and
>>>>>> OWL 2 for RDF validation.
>>>>>>
>>>>>> I also see the need for such a requirement.
>>>>>> This requirement is not in the requirements database so far.
>>>>>>
>>>>>> What would be an identifier of such a requirement?
>>>>>> Let me try:
>>>>>> Separate Contraint Semantics and Ontology Semantics
>>>>>
>>>>> How about:
>>>>>
>>>>> Separation of Constraints and Ontology Semantics
>>>>>
>>>>>
>>>>> I think it will need a rather precise definition, and I'm happy to
>>>>> contribute to that. I believe that the bottom line is that the
>>>>> validation constraints must not alter the semantics of the underlying
>>>>> ontology. It's a matter of thinking about both the closed world and
>>>>> the open world at the same time. In fact, I would go so far as to say
>>>>> that "meaning" is entirely the role of RDF/OWL, and that the
>>>>> application profile exercises data creation and management
>>>>> constraints. The things called "constraints" in OWL are actually
>>>>> axioms that permit one to infer meaning from relationships between
>>>>> properties, but do not constrain, in the data quality sense, at all.
>>>>> So it's constraints on the AP side, and semantics on the RDF/OWL side.
>>>>>
>>>>> kc
>>>>>
>>>>>
>>>>>>
>>>>>> Do you have a better one?
>>>>>>
>>>>>> There is a requirements class 'constraint semantics'.
>>>>>> I would assign this requirement to this class.
>>>>>>
>>>>>> We also have requirements which are fairly related:
>>>>>> R-140-SEPARATE-ONTOLOGIES-FROM-VALIDATION-SCHEMAS
>>>>>> R-173-SEPARATE-CONSTRAINTS-FROM-VOCABULARIES-AND-ONTOLOGIES
>>>>>> R-177-DEFINE-SEMANTICS-FOR-CONSTRAINTS
>>>>>>
>>>>>> -----
>>>>>> regarding the meaning between "domain as a means of enabling
>>>>>> inference" (the OWL sense) and "domain as a way to bind properties to
>>>>>> classes"
>>>>>>
>>>>>> I think it is important to choose if you want to use reasoning when
>>>>>> using OWL for RDF validation.
>>>>>> The RDF validator can be used for RDF validation with and without
>>>>>> inferencing.
>>>>>>
>>>>>>
>>>>>> 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 "Thomas Baker [[log in to unmask]]
>>>>>> Gesendet: Mittwoch, 1. Oktober 2014 12:34
>>>>>> An: [log in to unmask]
>>>>>> Betreff: [RDF AP] Re-purposing OWL properties
>>>>>>
>>>>>> On Tue, Sep 30, 2014 at 11:05:08AM -0700, Karen Coyle wrote:
>>>>>>> https://meetings.webex.com/collabs/meetings/playRecording?recordID=12797906&meetingInstanceID=ICWDUC9I93MGCE3CLXLPUZVPSX-JV0D
>>>>>>>
>>>>>>>
>>>>>>
>>>>>> I tried to join yesterday -- it was 23:00 here in Seoul -- but
>>>>>> couldn't
>>>>>> get a WebEx connection after five attempts. I did however get a
>>>>>> chance
>>>>>> to listen to the recording today. I'm sorry I couldn't be there live,
>>>>>> because I'm trying to catch up with the discussion, and apologies in
>>>>>> advance if the points I make have already been discussed and decided.
>>>>>>
>>>>>> Apart from the glitch that addresses do not know people, the DSP demo
>>>>>> was very nice! However, I get very uneasy when I see OWL2 axioms
>>>>>> being
>>>>>> treated as "constraints" in a DSP sense (i.e., interpreted
>>>>>> according to
>>>>>> CWA). I agree with Karen, if I correctly understood her point, that
>>>>>> this is "dangerous territory".
>>>>>>
>>>>>> On the call, that discussion was postponed for a later date, but I
>>>>>> look
>>>>>> forward to having that discussion as soon as possible because I
>>>>>> think it
>>>>>> is fundamental.
>>>>>>
>>>>>> As I see it, the whole question of how a constraint language
>>>>>> relates to
>>>>>> RDF vocabularies and ontologies is one of the most important and basic
>>>>>> _requirements_ for the constraint language itself. The requirement is
>>>>>> that a constraint language not replace (or "hijack") the original
>>>>>> semantics of properties used in the data. I get uneasy, for example,
>>>>>> when OWL cardinality axioms are treated as "constraints" according
>>>>>> to a
>>>>>> closed-world, unique-name assumption, or by using rdfs:domain or range
>>>>>> axioms as if they were expressing mandatory graph patterns.
>>>>>>
>>>>>> In my recollection, the difference in meaning between "domain as a
>>>>>> means
>>>>>> of enabling inference" (the OWL sense) and "domain as a way to bind
>>>>>> properties to classes" was a source of confusion when the Schema.org
>>>>>> vocabulary first appeared, because the early, now-deprecated
>>>>>> representations of the Schema.org vocabulary in OWL translated
>>>>>> Schema.org domains as rdfs:domain, whereas the Schema.org data
>>>>>> model now
>>>>>> makes clear that a much looser definition is intended -- one that has
>>>>>> more to do with documenting intention than with enabling inference
>>>>>> [1]:
>>>>>>
>>>>>> We have a set of properties:
>>>>>>
>>>>>> each property may have one or more types as its domains. The
>>>>>> property may be used for instances of any of these types.
>>>>>>
>>>>>> each property may have one or more types as its ranges. The
>>>>>> value(s) of the property should be instances of at least one
>>>>>> of these types.
>>>>>>
>>>>>> Has it been proposed to express as a requirement, alongside the other
>>>>>> requirements, the notion that the constraint language not impose an
>>>>>> alternative interpretation on existing semantics?
>>>>>>
>>>>>> Tom
>>>>>>
>>>>>> [1]
>>>>>> http://wiki.dublincore.org/index.php/RDF-Application-Profiles/ExamplesFormalConstraints#R-25-OBJECT-PROPERTY-DOMAIN
>>>>>>
>>>>>>
>>>>>> [2] http://schema.org/docs/datamodel.html
>>>>>>
>>>>>> --
>>>>>> Tom Baker <[log in to unmask]>
>>>>>>
>>>>>
>>>>
>>>
>>
>
|