Re: Interpretation of RDF reification

* John F. Sowa <sowa@bestweb.net> [2006-03-26 09:03-0800]
> Dan,
> 
> Common Logic is still in the FCD stage at ISO.  Some
> people are developing tools based on it, but I don't
> know of any that are commercially available right now.
> 
> DB> Sounds good to me. Where can I down load a Common Logic
> > database to play with? I've a few legacy files I'd like
> > to import...
> 
> However, the Common Logic Interchange Format (CLIF) has
> a large overlap with the Knowledge Interchange Format (KIF),
> which has been widely used since the mid 1990s.  Almost
> anything done in KIF can be moved over to CLIF with minor
> modifications (and some with no modification at all).
> 
> I know a couple of groups who have used Prolog to translate
> RDF and OWL into other formats.  At VivoMind, we have used
> Prolog to handle very large RDF and OWL files that cause
> Protege to choke.  Prolog is lightening fast for reading
> such files and converting them to other formats (such as
> CLIF and CGIF).  Our tools for using the results, however,
> are still in alpha and beta stages.
> 
> One of the primary tools we have been developing at VivoMind
> is the Flexible Modular Framework (FMF).  See the following
> paper for the basic idea:
> 
>    http://www.jfsowa.com/pubs/arch.htm
> 
> Our plans are to make an open source version of the FMF
> freely available and to make a business of developing and
> licensing modules to plug into the FMF.  However, we are
> still working on the commercial version.

Thanks for the pointers, I'll take a look around.

It sounds like we're at an awkward stage; RDF and OWL are 
'legacy' (from your perspective) but the tools to properly exploit
CLIF aren't quite ready to hit the mainstream yet (opensource or 
not). I appreciate that KIF got some traction, but it didn't win 
enough hearts, minds and budgets to stop XML taking over the world. Is 
CLIF expected to occupy a bigger niche? 

Having FMF opensourced sounds like a good step towards bridging 
that gap...

> 
> The short answer to your question is that I don't know
> anybody who develops stuff to play with, and the people
> I do know are up to their ears in work from paying
> customers.

Ah, I think Topic Maps suffered from that too. All work and no play...

If Common Logic is to go mainstream, and to really make RDF and OWL
(and XML?) obsolete, getting multiple opensource toolkits out there will be a 
big part of that. If a would-be paying customer says to me "hey,
I heard RDF is kinda 70s, shouldn't we be using Common Logic instead 
of RDF, OWL and SPARQL"? While I might agree in the abstract, the 
tool and market situation doesn't quite there yet. Well, it depends 
what the problem is. If the problem isn't heavily tied up with 
wide-area data sharing, CL tools might still solve the problem better
than RDF/OWL ones, even if they're relatively obscure. 

I'm entirely comfortable with the idea that CL may turn out to be the next 
great leap forward, but there's got to be work from someone on getting tools
into the playful and creative hands of ordinary tech developers if that's going 
to happen. And you're right; paying for that effort isn't going to happen 
by magic...

Dan


> John

Received on Sunday, 26 March 2006 17:48:50 UTC