- From: Arthur Ryman <ryman@ca.ibm.com>
- Date: Fri, 1 Aug 2014 07:59:58 -0400
- To: public-rdf-shapes@w3.org
Peter, You wrote: "So stating that if you use RDFS and OWL you end up with a lot of baggage and lose on reusability is completely false." I completely agree. The above assertion is false. However, that is what you wrote, not what I wrote. I wrote: "So if you use RDFS and OWL, and include a lot of constraints, then the terms are not very reusable because they carry a lot of baggage in the form of inferences which may not make sense in your application. Note that I am describing the state of the practice. Many ontologies in fact do contain a lot of inferences and this makes them less reusable. This is true even in absence of constraints since W3C OWL doesn't officially handle constraints, just inference rules. The point I was trying to make is that applications often need to impose additional constraints on terms and that these constraints often depend on the context. Let's refocus the discussion on how an application designer would reuse an externally published OWL ontology and impose additional constraints. Let's assume OWL IC semantics. What would the application designer provide to describe a REST API that required exactly one dcterms:title triple in the request body on POST? Regards, ___________________________________________________________________________ Arthur Ryman, PhD Chief Data Officer, Rational Chief Architect, Portfolio & Strategy Management Distinguished Engineer | Master Inventor | Academy of Technology Toronto Lab | +1-905-413-3077 (office) | +1-416-939-5063 (mobile) From: "Peter F. Patel-Schneider" <pfpschneider@gmail.com> To: Arthur Ryman/Toronto/IBM@IBMCA, "Peter F. Patel-Schneider" <Peter.Patel-Schneider@nuance.com>, Cc: public-rdf-shapes@w3.org Date: 07/31/2014 04:55 PM Subject: Re: ShEx relation to SPIN/OWL All these complaints about RDFS and OWL appear to be based on the conception that RDFS and OWL work with a single document containing everything that can ever be said about a particular vocabulary. However, this is a misconception. It is certainly possible in RDFS and OWL to have multiple documents that speak to the same vocabulary. This is done, for example, when OWL ontologies are extended. A core document talks about the core vocabulary and other documents add new vocabulary and add new information about the core vocabulary as well. The exact same thing can (and should) be done with constraints. You can have a document that provides the core information about a vocabulary. You can have multiple documents that provide constraints on this vocabulary for various purposes. When validating some information one can pick which set or sets of constraints to apply. This division can also be done with the axioms of the ontology. There is no need for all the axioms of an ontology to be in the same document, leading to the possibility of having the RDFS portion be in one document and the non-RDFS portion in another document. Other divisions are also possible. So stating that if you use RDFS and OWL you end up with a lot of baggage and lose on reusability is completely false. Peter F. Patel-Schneider On 07/31/2014 01:06 PM, Arthur Ryman wrote: > Peter, > > Apologies for the delayed reply. > > The reason I say RDFS and OWL fail when combining terms from multiple > sources is that the constraints associated with a term often depend on how > the term is used in a particular application. Consider DC Terms. It was > designed to provide a highly reusable vocabulary. There are very few > constraints imposed by DC Terms and this makes it very reusable. For > example, consider dcterms:title. In one application dcterms:title may be a > required property while in another it may be optional. So where do you put > those constraints? They do not belong in the DC Terms vocabulary. You need > another place to define constraints and associate them with the > application. That is what OSLC Shapes does. > > So if you use RDFS and OWL, and include a lot of constraints, then the > terms are not very reusable because they carry a lot of baggage in the > form of inferences which may not make sense in your application. For > example, if you used rdfs:domain or rdfs:range then you infer certain type > triples. I once was looking for terms that defined start and stop dates. I > found dtstart and dtend in http://www.w3.org/2002/12/cal/ical but when I > looked at the RDFS/OWL it had a lot of undesired inferences concerning the > domain. So if I had reused dtstart in my application, then a reasoner > would have inferred a lot of triples that didn't make sense. > > At OSLC we are trying to integrate information from a lot of applications. > We try to reuse terms from established vocabularies. We define new common > terms in a core OSLC vocabulary. We do this to make writing queries > easier. This means that resources contain terms from many vocabularies. > > To make an analogy with natural language, we have words and they are > defined in a dictionary. We also have grammars and they define how the > words are combined. We have the general rules of grammar for ordinary > prose. We also have other types of grammars for specialized texts such as > poetry. There are rules for sonnets, limericks, etc. The dictionary is > analogous to the RDFS or OWL document. The grammar is analogous to a Shape > document. > > Regards, > ___________________________________________________________________________ > Arthur Ryman, PhD > > Chief Data Officer, Rational > Chief Architect, Portfolio & Strategy Management > Distinguished Engineer | Master Inventor | Academy of Technology > > Toronto Lab | +1-905-413-3077 (office) | +1-416-939-5063 (mobile) > > > > > > From: "Peter F. Patel-Schneider" <Peter.Patel-Schneider@nuance.com> > To: <public-rdf-shapes@w3.org>, > Date: 07/10/2014 10:05 AM > Subject: Re: ShEx relation to SPIN/OWL > > > >> From: Arthur Ryman <ryman@ca.ibm.com> >> Date: Thu, 3 Jul 2014 10:58:24 -0400 >> To: "public-rdf-shapes@w3.org" <public-rdf-shapes@w3.org> >> Message-ID: > <OFF14B15B5.802B33E2-ON85257D0A.004C62E1-85257D0A.005240FE@ca.ibm.com> >> >> I'd like to back up a little and discuss the need for something other > than >> OWL and SPARQL. > > [...] > >> Another important requirement is that a constraint language should be >> independent of any vocabulary or ontology since it is often the case > that >> an RDF document combines terms from multiple sources. Both OWL and RDFS >> fail on this count. > > I believe that this is false. Can you provide examples where OWL or RDFS > fail > on this count, particularly where the combination of multiple sources > contributes to the failure? > > [...] > >> Regards, >> > ___________________________________________________________________________ >> Arthur Ryman, PhD >> >> Chief Data Officer, Rational >> Chief Architect, Portfolio & Strategy Management >> Distinguished Engineer | Master Inventor | Academy of Technology >> >> Toronto Lab | +1-905-413-3077 (office) | +1-416-939-5063 (mobile) >> >> > > peter > > > > >
Received on Friday, 1 August 2014 12:00:31 UTC