- From: Bijan Parsia <bparsia@isr.umd.edu>
- Date: Sat, 8 Jan 2005 00:33:36 +0900
- To: jos.deroo@agfa.com
- Cc: www-rdf-logic@w3.org
On Jan 8, 2005, at 12:23 AM, jos.deroo@agfa.com wrote:
> Bijan, no touch of the store
Hmm. There has to be at least for matching the lhs right?
> and lists are done with rdf triples
My personal guess is that you make some assumptions about how the list
triples occur. I'll try to follow up later on this. (Not tonight! For
once, I'm going to bed :)
> actually, we only have for whole engine
>
> public class Euler {
> internal Euler subj = null; // RDF subject
> internal String verb = null; // RDF predicate as absoluteized verb
> internal Euler obj = null; // RDF object
> // some other stuff
How can you have your input then?
> Since 84 or so (waw that is 20 years!)
Never doubt it: You are the spirit of youthfulness incarnate.
> I find prolog terms
> very nifty but since few years I now also find it good that
> if y is function of x then there is relation between y and x
I'm not understanding. I'm referring to having rules whose head is
nff(In, Out) where In and Out get bound to a structure as opposed to
having to assert the term into the store, query for the term, assert
the nnfed term back into the store. (And if you aren't using the store,
how are you using the triple structure? How are you passing things
around? This is, in part, the wellformedness problem!)
Cheers,
Bijan.
Received on Friday, 7 January 2005 15:33:45 UTC