Re: What do the ontologists want

> >1.  I think you over-estimate the fraction of the data in the universe
> >    that is more than ground facts.
> 
> But there are many ground facts that cannot be put into simple 
> triples, notably negations. Ground facts can get arbitrarily 
> complicated in their syntax in some languages: check out ground DAML, 
> for example.
> 
> And, while it may be true that the bulk of the total information is 
> ground facts, (maybe even ground atoms) if you just count symbols, it 
> is often true that these ground facts are only of use because they 
> can be processed by a smaller collection of non-ground facts (ie if 
> 'rules' can be applied to them).

Yeah, maybe I should switch sides of this argument.  I like the idea
that the ground facts are just raw observations/inputs and everything
else is derived by rules and need not be stored.  The bulk is probably
still in ground facts, but certainly not the interesting parts.

> >2.  RDF is not necessarily verbose: the RDF syntax in the current W3C
> >    spec is verbose, but other RDF syntaxes are much less so (eg n3,
> >    as Jos de Roo pointed out).
> 
> I agree: the verbosity arises chiefly from XML rather than RDF 
> itself. Has anyone suggested Lexical_XML? 
> <capletter>I</capletter><letter>t</letter><space> 
> </space><letter>l</letter><double-letter>oo</double-letter><letter>k</ 
> letter><letter>s</letter> like that. Its a really neat universal 
> notation: you can describe it in itself!  (The proof is too long to 
> fit in this message, however.)

<laugh_out_loud>
  :-)
</laugh_out_loud>

      -- sandro

Received on Friday, 18 May 2001 15:43:26 UTC