Re: [RIF] Reaction to the proposal by Boley, Kifer et al

Alex Kozlenkov wrote:

> There are a few general concerns we have about the points raised by
> Peter in that RDF/OWL should be first-class citizens in RIF. Firstly,
> the production/reaction rules should play a good part of the standard.
> As I understand it, it has been an industrial requirement all along,
> certainly from our company. PR/RR have their own semantics that is not
> directly compatible with, say, RDF deduction.

In what way? Or for that matter what do you mean by "RDF deduction" in 
that context?

A similar question was asked a few weeks ago in the form "does this mean 
that all RIF processors have to build in the RDF semantics" (where 
"this" = the "RIF must accept RDF data" design constraint). My personal 
response to this was "not necessarily", see:
http://lists.w3.org/Archives/Public/public-rif-wg/2006Apr/0066.html

But in any case, I don't see the fundamental incompatibility between 
specifically RDF and production rules. For example, the Jena software 
includes a simple RETE-style production rule language for processing RDF 
and it seems to do a useful job for quite a few users.

Perhaps you could spell out what specific problems you see and we can 
look at ways of solving those.

Support for RDF and OWL are clear parts of the RIF charter and have also 
been an "industrial requirement all along". At issue is the *way* in 
which we support them, not whether.

> Secondly, tying RIF closely to OWL would effectively assume that
> well-designed OWL ontologies become a prerequisite to do any meaningful
> form of inference. 

I'm sorry but that is a non sequitur. Just because it would be 
*possible* for RIF rules to meaningfully refer to an OWL ontology and 
exploit some part of its semantics in no way implies that all RIF 
applications *have* to use OWL, let alone making its use a prerequisite 
for all "meaningful" inference.

> The reality of large software projects is that it may
> be prohibitively expensive to construct and maintain a consistent and
> complete set of reference ontologies even for a subset of the main line
> of business. 

This is irrelevant to the requirement that RIF be *able* to work with 
OWL. However, even in projects which do use OWL this is not a 
requirement. The semantic web does not call for a complete top down 
formal ontology of everything in your domain. One can perfectly well 
make use of ontologies which cover some small interesting fragment of 
your domain. The semi-structured nature of RDF is a key benefit in such 
situations, it allows you to make flexible trade-offs in the degree of 
axiomization of your domain.

> What I am saying
> is that rules may have to play a pragmatic role than other parts of the
> Semantic Web technologies hence tying them closely with these parts may
> hinder the success of the rules technologies per se. 

You seem to be setting up a false dichotomy between pragmatic and 
semantic web. We routinely use semantic web technologies in highly 
"pragmatic" ways in current business applications :-)

> I was wondering if you think these two observations are making sense.

No sorry, not to me.

Dave

Received on Wednesday, 3 May 2006 14:13:55 UTC