On 7/16/07, Bulat Ziganshin <bulat.ziganshin@gmail.com> wrote:
> Topcoder certainly isn't about benchmarking. Undoubtedly, it would be
> absolutely awesome to be able to use Haskell in topcoder... but it
> wouldn't say anything about speed. My guess is that practically no
> topcoder submissions fail by exceeding the allowable time limit. The
> competition (the alg one, which is the only one anyone really cares
> about) is about solving problems quickly (in programmer time) and
> accurately.
that's ideal for haskell. like ICFP, if they will allow haskell code,
then all winer solutions will be written using it
Careful. Although writer's solution (the "reference" solution) must run in under 1 second in Java, many problems run really close to the 2-second limit in practice. That means if your language is inherently 30% slower, you may fail the harder problems.