Folks,
We're trying to do a "Haskell vs anything-else (preferably C)" study. We're
interested both in their size and structure, and in their performance. We
intend to make quite detailed comparative measurements of their performance.
Do you have programs we could have a look at?
We are most interested in "real" programs that you actually used for
Real Work. The ideal would be programs that you "prototyped" in
Haskell, then rewrote into C (or the other way around :-).
The functional programs can be anything Haskell-ish; the "anything
else" can be anything; the "realness" can be limited; the `match'
between the two versions can be tenuous... We'll look at anything at
this point!
What's in it for you? We hope to learn more about why functional
programs behave as they do, so we can give you a better Glasgow
Haskell compiler.
Many thanks to those who have supplied us with Haskell-program fodder
over the past few years. You get partial credit for GHC's
ongoing improvements.
Simon
PS. We already have, or know about:
- the Simple code
- the Salishan problems
- the Cowichan problems
- "pseudoknot"
