Re: [UCR] RIF needs different reasoning methods

Bijan Parsia wrote:
> When you say, "have the same declarative semantics", I read, "has the
> same semantics", and thus, "justifies the same answers to queries".
This is not quite so, I think, because the queries have to be specified.
Assume some rules specify how much taxes one should pay. One query can
be: "how much taxes should Annsa pay?". Another query could be: "does
Anna pay as much taxes as she should?". This difference in usage
(pragmatics) leads to different rule processing methods.

> Pragmatics might help the use understand what's going on, but, in
> general, they shouldn't change the answers returned. I.e., they are
> like comments.
My understanding is that, eg in the example above, this is not quite so.
>
> Now in your prior message you mentioned that termination might be
> different. Does this mean that if I set the rule type bit and the
> proof procedure bit that, in order to process the rule set
> *correctly*, I must return exactly the same answers as that proof
> procedure? Is it ok *not* to actually use the specified proof
> procedure so long as I produce those answers? (Does answer order matter?)
I just meant that very often forward (resp. backwards) rule processors
do not terminate on rulesets thought for a backward (resp. forwqard)
rule processor.

Regards,

François

Received on Wednesday, 8 March 2006 16:14:03 UTC