A couple of typos from the previous and some broader statements on OWL vs. SPARQL vs. SPIN.
The comment for the first query read "# A person can have only one birth father" and should have read "# Functional properties can only have one value". In the third-to last paragraph the first sentence was "SPIN can be used for general-purpose constraint checking, and OWL is really not designed for [consistency] checking." and should have been "SPIN can be used for general-purpose constraint checking, and OWL is really not designed for [constraint] checking. ---- Let me be clear that TopQuadrant is fully committed to OWL standards and provides access to reasoners (including SPIN's OWL 2 implementation) that perform OWL reasoning. TopBraid Suite is compliant with OWL and OWL 2 standards throughout the editor and other TBS features. TopBraid Suite is also fully compliant with SPARQL standards and has already implemented most of the SPARQL 1.1 working draft (final recommendation expected soon). It is more correct to be comparing RDFS/OWL and SPARQL, where SPARQL is a query language for RDF, and RDFS and OWL are specified in RDF. SPIN is fully based on SPARQL. One way to look OWL is that it is a set of axioms that restrict how RDF is used. That isn't a bad thing, as OWL is a nice language for defining classification and consistent models based on FOPL (First-Order Predicate Logic). But it is a statement on expressiveness - OWL is by its very definition less expressive than RDF. SPIN is a family of technologies (see http://spinrdf.org) that includes SPARQL Rules, which uses SPARQL as a rule language. SPARQL Rules are expressive enough to define OWL, and we include a full implementation of the W3C OWL 2 RL profile (see http://www.w3.org/TR/2009/WD-owl2-profiles-20090421/#Reasoning_in_OWL_2_RL_and_RDF_Graphs_using_Rules) in the TopBraid Composer/Live library. OWL thus provides a small subset of what can be expressed in SPARQL, and hence SPIN. Using RDF data structures can make some things much easier to express than in OWL. The modeling required tend to be simpler and more intuitive. Other examples can be provided, but there is a discussion of the Minor League Player example from Allemang and Hendler that I find to be a useful example: http://www.semanticoverflow.com/questions/1786/how-to-do-set-complement-reasoning-in-owl. There are many things that cannot be expressed in OWL. Any calculation, like a person's age, concatenating two strings, etc., cannot be done in OWL. OWL is never enough to create a full application, and must rely on other programming to do the computational tasks that are required by any application. SPARQL, on the other hand, can be used for various data manipulations, and therefore reduces the need to write code. We have complex TopBraid Suite applications that have been created without writing any Java code. The recently announced TopBraid Enterprise Vocabulary Net (EVN) is an example. BTW, this is not to be taken as diminishing OWL's value in any way. The question is on the expressiveness of OWL, and while there are use cases in which OWL is very useful, the fact is that it is not as expressive as RDF/SPARQL, and is in fact a subset of RDF/SPARQL. OWL is not intended for general-purpose computations, instead targeting a subset of FOPL. Anything outside of that domain are things that will require rules, or other computational paradigms, to complete the solution. So, more fuel for the discussion. Let us know if that's hitting the mark at all. What other question need addressing. Etc. -- Scott -- 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
