From: "MR Keenan" <mike.keenan@aspentech.com>
Path: news.mathworks.com!newsfeed.mathworks.com!WebX
Newsgroups: comp.soft-sys.matlab
Subject: Re: Scoring Algorithm
Message-ID: <eeb4a65.63@WebX.raydaftYaTP>
Date: Tue, 12 Nov 2002 15:57:44 -0500
References: <eeb4a65.-1@WebX.raydaftYaTP> <eeb4a65.9@WebX.raydaftYaTP> <eeb4a65.10@WebX.raydaftYaTP> <eeb4a65.22@WebX.raydaftYaTP> <eeb4a65.59@WebX.raydaftYaTP>
Lines: 73
MIME-Version: 1.0
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: 8bit
Xref: news.mathworks.com comp.soft-sys.matlab:142694


I think this is not correct.  Of course, the programming is poorly
documented and inefficient.  But, incredibly, both the documentation
and efficiency increases as the contest goes along.  What folks like
Lucio have done is an excellent example of proper coding techniques. 


One problem of course is the extreme difficulty of the problem.  PhD
theses could be written on the solution to it (or at least a good
masters thesis).  Once again, I congratulate the Mathworks team on a
great problem selection. 


Some of the other suggestions below may be good.  But the system
doesn't appear to be broken to me.


Stefan Stoll wrote:
> 
> 
> Kevin Spiteri wrote:
>>
>> score = result + exp(-1.95 + 0.05*runtime)
>>
> 
> The current top entry contains 1771 lines
> of poorly documented M-code, making it al-
> most impossible to grasp what's being changed
> or added from one entry to the next.
> 
> I admit that all the submitted "spaghetti
> codes" are very good at folding the proteins
> from the contest suite, but I can't escape
> the feeling that most GPRs (good programming
> rules) are completely neglected. To put it
> a bit sharper:
> 
> The contest discourages good programming.
> 
> Don't misunderstand me. I am enjoying the
> contest and facinated by the zoology of
> modules in the current code!
> 
> So what can be improved?
> - Add the length of the p-code of an entry
> to its score. This will favour generic
> and compact solutions.
> - Disclose code of new entries only when
> they are a few hours old. This helps
> participants to stick to their own ideas
> at least for a while before copying the
> tweaks from others into their codes.
> - Let participants submit entries for
> testing purposes only. The scoring
> result will be sent back to the parti-
> cipant per e-mail. So one has time to
> work on an idea before it gets "open
> source". Every fifth contribution (or
> so) of a participant gets included into
> the official ranking, and its code
> disclosed.
> - If the contest should encourage team
> work, why not set up teams that can
> register and share their private ranking
> for a while? One could combine the best
> codes from all private ranking lists into
> an official ranking every couple of
> hours.
> 
> Just ideas...
> 
> Last not least:
> Please provide an entry download button in
> addition to the edit button!
> 
> sTefan
>