On Wed, 6 Dec 2000, Patrik Stridvall wrote:
> I think we have different definitions of heuristics.
Definitely.
> But that is not an heuristic. If it it referenced or recusive it is
> always correct to not inline it. There is no incorrect cases and
> therefore it is not an heuristic it is an algoritm and i'm not
> possitive that it is even a non-optimal algoritm.
Making a distinction between heuristics and algorithms does not really
make sense here. Heuristics are not (black) magic, when they are part
of a program, they are algorithms as well.
> Then we have a little missundering with the word heurstics.
> There are four kinds of algoritms.
>
> 1. Optimal algoritms
> 2. Non-optimal algoritms
> 3 Algoritms buggy by design
> 4. Algoritms buggy by misstake
>
> As far as I'm concern 3 is heuristics,
> but you seem to wish to include 2 as well.
I have been dealing with heuristics of various kinds, but "buggy by
design" fits none of those.
Many problems simply do not lend themselves to a general and optimal
strategy, that's where heuristics come into play. (Even Quicksort is
usually implemented with heuristics to choose the "mean" value, and
you won't call Quicksort buggy by design, will you?)
And yes, I do have scientific papers in referreed conferences that
deal with heuristics! :-)
On Wed, 6 Dec 2000, Ove Kaaven wrote:
> (heuristics should be limited to what can be optimized, which is
> perfectly okay, since that can never generate incorrect code).
Yup: An optimizer employing heuristics does not generate incorrect code
due to this (unless it has a bug). It may, however, generate suboptimal
code in some cases -- that's where the heuristics may fail.
Gerald
--
Gerald "Jerry" [EMAIL PROTECTED] http://www.dbai.tuwien.ac.at/~pfeifer/
