If you know that you will use a tableau (DL) reasoner, you should keep your model(s) in DL. And you should probably develop them in such a way as to optimize the DL reasoning. For example, due to the nature of tableau algorithm it likes a lot of disjoint statements, not so with rules based reasoners. Tableau is good about figuring out subsumption, that also has implications on the style of modeling. Tableau reasoners are slow and do not scale to reasoning over largish data sets.
Therefore, the plan to use a DL reasoner typically means that either there is a small amount of information or the information is partitioned into several graphs making it possible to run the DL reasoner only over select small graphs. In this case, the overall graph does not need to be in DL, only the graphs that you plan to use DL reasoner with should be in DL. Other OWL reasoners do not care if ontology is in OWL-DL. In fact, some of the modeling style choices optimized for DL reasoners will have an opposite effect on other reasoners. So, one way or another, it is important to decide what reasoning (if any) you will use as it will have modeling implications. For these reasons, with OWL 2 the conversation has moved from talking about OWL DL and OWL Full (and you never hear about OWL Light anymore - it was not a very grounded idea) to talking about different profiles where each profile is suited for a particular reasoning technology. Btw, foaf, for example, is OWL Full. Because some of the datatype properties it uses (such as YahooID, etc.) are inverse functional. Regards, Irene On Jan 26, 1:31 pm, "Bradley Shoebottom" <[email protected]> wrote: > Scott, > > To put it simply, I was told to keep the ontology in OWL-DL so as to do > reasoning better. I think it had to do something with in OWL Full that a > Class and an instance could be the same making it difficult to isolate out > the instances. > > Bradley Shoebottom > Information Architect - R&D, Innovatia Inc. > Tel: (506) 674-5439 | Skype: bradleyshoebottom | Toll-Free: 1-800-363-335 > begin_of_the_skype_highlighting > [email protected] |www.innovatia.net| Follow us on Twitter > > -----Original Message----- > From: [email protected] > [mailto:[email protected]] On Behalf Of Scott Henninger > Sent: January-26-11 10:12 AM > To: TopBraid Suite Users > Subject: [topbraid-users] Re: Syntax Checker? > > Hello Bradley; I'm curious why you believe it is important to make > sure your model is not OWL Full. I believe Holger's point is that > this tends to be much ado about nothing. A case in point is that the > simple kennedys model is OWL Full. But it runs without problems in > any reasoner. > > So I'm curious whether there are real use cases in which a OWL Full > construct causes real (not theoretical) problems for a reasoner. I've > search the web periodically for such an example, but always come up > empty-handed. Can someone provide such examples? > > I'll also make a bold claim that you will get all of the OWL reasoning > you really need from OWL 2 RL, which can run on OWL Full. Again, if > there are counterexamples, we are very interested in hearing them so > we can suggest either modeling constructs or rules that will fill in > any gaps, perceived or real. > > BTW, I usually have trouble getting to that validator site. > > -- Scott > > On Jan 26, 6:28 am, "Bradley Shoebottom" > <[email protected]> wrote: > > There is a OWL Species checker here at Manchester University. > > >http://owl.cs.manchester.ac.uk/validator/ > > > I use it to ensure that when I make changes in my ontology or after I > > have received a populated ontology form text mining that we have not > > accidentally go to OWL-Full form OWL-DL (It happened once.) > > > Bradley Shoebottom <http://www.linkedin.com/companies/innovatia> > > Information Architect - R&D, Innovatia Inc. > > Tel: (506) 674-5439 | Skype: bradleyshoebottom | Toll-Free: > > 1-800-363-335 begin_of_the_skype_highlighting > > [email protected] > > <mailto:[email protected]> |www.innovatia.net > > <http://www.innovatia.net/> | Follow us on Twitter > > <http://www.twitter.com/InnovatiaInc> > > <http://www.innovatia.net/> > > > From: [email protected] > > [mailto:[email protected]] On Behalf Of Holger Knublauch > > Sent: January-26-11 3:25 AM > > To: [email protected] > > Subject: Re: [topbraid-users] Syntax Checker? > > > Hi Leonard, > > > we used to include an OWL species checker in earlier versions of TBC, > > but the underlying component had been discontinued by the Jena project. > > So right now there is nothing out of the box. One strategy is to simply > > run an OWL DL inference engine such as Pellet and observe whether it > > will complain. > > > The broader question that I always ask is why is the OWL DL distinction > > relevant in your particular case. From our experience, there are lots of > > users out there who believe that going beyond OWL DL is "dangerous" but > > don't really know why. OWL DL is a two-side sword and getting religious > > about pure OWL DL may prevent you from exploiting the real power of > > RDF-based languages. There are plenty of very useful OWL Full (or even > > RDFS) ontologies out there, and very little success stories of OWL DL > > reasoning. > > > In my personal experience I have not seen many projects that required > > OWL DL compliance. Just use OWL RL implementations and other reasoners > > that are able to just ignore the OWL Full bits. > > > Holger > > > On Jan 26, 2011, at 4:51 AM, Leonard Jacuzzo wrote: > > > Hello, > > > I am wondering if TBC has a built-in syntax checker that will determine > > whether or not a model is OWL-DL compliant. > > > Any help will be greatly appreciated. > > > Best wishes, > > > LFJ > > > -- > > You received this message because you are subscribed to the Google > > Group "TopBraid Suite Users", the topics of which include TopBraid > > Composer, > > TopBraid Live, TopBraid Ensemble, SPARQLMotion and SPIN. > > To post to this group, send email to > > [email protected] > > To unsubscribe from this group, send email to > > [email protected] > > For more options, visit this group > > athttp://groups.google.com/group/topbraid-users?hl=en > > > -- > > You received this message because you are subscribed to the Google > > Group "TopBraid Suite Users", the topics of which include TopBraid > > Composer, > > TopBraid Live, TopBraid Ensemble, SPARQLMotion and SPIN. > > To post to this group, send email to > > [email protected] > > To unsubscribe from this group, send email to > > [email protected] > > For more options, visit this group > > athttp://groups.google.com/group/topbraid-users?hl=en > > > image001.gif > > < 1KViewDownload > > > image002.gif > > < 1KViewDownload > > > image003.gif > > 3KViewDownload > > -- > You received this message because you are subscribed to the Google > Group "TopBraid Suite Users", the topics of which include TopBraid Composer, > TopBraid Live, TopBraid Ensemble, SPARQLMotion and SPIN. > To post to this group, send email to > [email protected] > To unsubscribe from this group, send email to > [email protected] > For more options, visit this group > athttp://groups.google.com/group/topbraid-users?hl=en -- You received this message because you are subscribed to the Google Group "TopBraid Suite Users", the topics of which include TopBraid Composer, TopBraid Live, TopBraid Ensemble, SPARQLMotion and SPIN. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/topbraid-users?hl=en
