(fwd) ICFP programming contest results
Benjamin L. Russell
On Wed, 24 Sep 2008 06:06:05 +0100, in gmane.comp.lang.haskell.general
Malcolm Wallace <malcolm...@cs.york.ac.uk> wrote:
>The ICFP programming contest results presentation:
>http://video.google.com/videoplay?docid=-4697764813432201693
>
>Feel free to pass on this link to any other appropriate forum.
-- Benjamin L. Russell
guthrie
wrote:
> The ICFP programming contest results are out!
>
> >The ICFP programming contest results presentation:
> >http://video.google.com/videoplay?docid=-4697764813432201693
>
The ICFP contest is interesting. This year a bit less than 1/4 of the
entries were even in FP languages, and none of the winners were FP.
What about the contest makes it FP? The idea is to provide a forum to
show the value of FP, but how are any outcomes of the contest actually
used for this? I think it would be useful if somehow good FP solutions
were analyzed "post-game" to show if/how FP methods are useful in
solving such problems.
This could then be a forum for helping students (and others) to
appreciate FP methods more; else rename it to the IC*P contest. :-)
Best,
Greg
ross...@ps.uni-sb.de
>
> The ICFP contest is interesting. This year a bit less than 1/4 of the
> entries were even in FP languages, and none of the winners were FP.
The winner of the lightning division was in ML.
> This could then be a forum for helping students (and others) to
> appreciate FP methods more; else rename it to the IC*P contest. :-)
It's named ICFP programming contest because it is organised by the
annual International Conference on Functional Programming.
- Andreas
Jon Harrop
> On Sep 24, 5:52 am, Benjamin L. Russell <DekuDekup...@Yahoo.com>
> wrote:
>> The ICFP programming contest results are out!
>>
>> >The ICFP programming contest results presentation:
>> >http://video.google.com/videoplay?docid=-4697764813432201693
Are the results available anywhere in a more legible form?
> The ICFP contest is interesting. This year a bit less than 1/4 of the
> entries were even in FP languages, and none of the winners were FP.
The Lightning round was won by a single person using OCaml.
The two worst entrants were probably OCaml and Haskell (judging by their
names).
> What about the contest makes it FP? The idea is to provide a forum to
> show the value of FP,
I think the idea is simply to have fun.
> but how are any outcomes of the contest actually
> used for this?
I would like to be able to study the results. For example, what was the
average score or rank by programming language? How did dynamic vs static
languages fare? What if you divide each score by team size? How much code
was required? Does code size correlate with lighting vs normal (i.e. can we
measure productivity)? Do big teams win regardless of language? What were
the distributions of team sizes for each language? How many Java
programmers does it take to beat an OCaml programmer?
Restricting consideration to the winners only and/or grouping all FP
languages together is just throwing away the data. However, according to
the lecturer on that video, most of the information required to perform
such analyses properly was never gathered and will never be available.
> I think it would be useful if somehow good FP solutions
> were analyzed "post-game" to show if/how FP methods are useful in
> solving such problems.
>
> This could then be a forum for helping students (and others) to
> appreciate FP methods more; else rename it to the IC*P contest. :-)
I would like to see the results from all ICFP's presented in a more easily
assimilated form. People obviously put a lot of work into this contest but
the results barely see the light of day.
--
Dr Jon D Harrop, Flying Frog Consultancy Ltd.
http://www.ffconsultancy.com/?u
Stephen Horne
wrote:
>The idea is to provide a forum to
>show the value of FP, but how are any outcomes of the contest actually
>used for this?
If you aim to show the value of FP, you can't prove your point if you
aren't willing to risk being proven wrong. If you artificially distort
the test to give FP an unfair advantage or to reject non-FP outright,
obviously no-one is going to have any respect for the results.
The problem is that in a fair contest, there's more to winning than
choice of programming language. And even the programming language
choice may be more a matter of familiarity, tools, libraries etc than
the style of the language. So much boils down to popularity.
namekuseijin
wrote:
> namekuseijin <namekusei...@gmail.com> writes:
> > A somewhat related ongoing programming competition is the Euler
> > Project. It comes with some cool statistics which should make
> > functional programmers happy:
> >http://projecteuler.net/index.php?section=statistics
>
> This is indeed a fun set of puzzles. They remind me a bit of the old
> Duke University Programming Contest, which had a set of similar
> problems that you had to solve in a few hours.
>
> I find the ICFP problems much too time-consuming to attempt. I prefer
> small problems and a very short time limit.
It's more math oriented, but much fun indeed. :)
Some of the more advanced problems seem to almost approach ICFP-kind
of difficulty, like 208:
http://projecteuler.net/index.php?section=problems&id=208
to calculate all possible closed one-fifth circular paths a robot
could pursue given 70 moves...
It's also wildly popular, with many people from all over the world
registered ("Within the past day 1542 registered users have visited.")
and an average of at least some 50 people online at any one time!
Solving math/programming puzzles! It's incredible in that respect
alone...
so far, I'm not even a tetrahedron:
http://projecteuler.net/index.php?section=profile&profile=namekuseijin
http://projecteuler.net/index.php?section=scores
:P
Torben Ægidius Mogensen
> On 30 set, 07:30, torb...@pc-003.diku.dk (Torben Ægidius Mogensen)
> wrote:
>> namekuseijin <namekusei...@gmail.com> writes:
>> > A somewhat related ongoing programming competition is the Euler
>> > Project. It comes with some cool statistics which should make
>> > functional programmers happy:
>> >http://projecteuler.net/index.php?section=statistics
>>
>> This is indeed a fun set of puzzles. They remind me a bit of the old
>> Duke University Programming Contest, which had a set of similar
>> problems that you had to solve in a few hours.
>
> Some of the more advanced problems seem to almost approach ICFP-kind
> of difficulty, like 208:
> http://projecteuler.net/index.php?section=problems&id=208
> to calculate all possible closed one-fifth circular paths a robot
> could pursue given 70 moves...
>
> so far, I'm not even a tetrahedron:
I have solved only a handful of the early puzzles (using Haskell) plus
#205 (http://projecteuler.net/index.php?section=problems&id=205),
which is a shoe-in for Troll:
count (sum 6d6)<(sum 9d4)
It took 0.004s to calculate the answer. :-)
Torben