Re: residuation for compositional search?

From: Michael Hanus <mh_at_informatik.uni-kiel.de>
Date: Tue, 14 Jun 2011 18:52:24 +0200

Antonio Regidor García wrote at June 14 (09:46):
> But findall (\x -> partition [1,2,3] =:= x) terminates in one version but not in the other.

"findall" is meta-programming stuff, useful but outside the theoretical
framework (i.e., models or declarative semantics) of FLP.
 
> Anyway, I think it is a BAD THING that the termination of a program dependes on the compiler you use. Maybe the report should be more specific about the evaluation of (&).

I would agree if you would say "successful termination".
In case of failures, I think there are many languages or systems
where the ability to fail instead of running into an infinite
loop depends on the abilities of the compiler. For instance, if you
require that some class of programs should not terminate,
you exclude the development of sophisticated loop detection techniques
that provide useful error messages to the programmer instead of
simply looping.

Best regards,

Michael
_______________________________________________
curry mailing list
curry_at_lists.RWTH-Aachen.DE
http://MailMan.RWTH-Aachen.DE/mailman/listinfo/curry
Received on Di Jun 14 2011 - 18:52:56 CEST

This archive was generated by hypermail 2.3.0 : Sa Dez 07 2019 - 07:15:10 CET