Re: [Caml-list] Rule based language [was: productivity improvement]

[Oleg <oleg_inconnu@myrealbox.com>]

On Wednesday 24 July 2002 06:31 pm, sajuma@utu.fi wrote:
> I think you misunderstood the specification of the language.
> (It was not very clear). The meaning of "a and b" should not
> be "a is reachable and b is reachable" (additive and), but
> "a and b are true at the same time" (multiplicative and).
> Of course I could be mistaken too, but the multiplicative case
> is more interesting.

I did not misunderstand. I use multiplicative AND. All three programs give 
equivalent output when they all finish for all cases I looked at. 

However, your and Alex's programs, for examle, fail to finish processing this 
file containing 9000 rules with preconditions of lengths 1 to 10, 10 goals 
and 10 dataset points. (My patience ran out after 72 and 45 minutes of 
waiting for your and Alex's programs, respectively):

http://www.columbia.edu/~ot14/rules_test_long.input.gz (152 kB), 

while mine takes only 4 seconds. Something to think about [1]

> Here is a question: in C you can hack in all the object
> oriented features, so why are you using C++? Many claim that
> OOP in C is better than in C++, so what would you say to these
> people?

I'd ask them if they were on any special medication.

Cheers,
Oleg

[1] As I said, I certainly do not blame O'Caml for this. Just poor choice of 
algorithm.
-------------------
To unsubscribe, mail caml-list-request@inria.fr Archives: http://caml.inria.fr
Bug reports: http://caml.inria.fr/bin/caml-bugs FAQ: http://caml.inria.fr/FAQ/
Beginner's list: http://groups.yahoo.com/group/ocaml_beginners