Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: suggestion for a rule for exposing source code in world championship

Author: Uri Blass

Date: 01:48:13 02/28/04

Go up one level in this thread


On February 27, 2004 at 22:39:24, Robert Hyatt wrote:

>On February 27, 2004 at 18:31:18, Telmo Escobar wrote:
>
>>On February 27, 2004 at 10:22:56, Robert Hyatt wrote:
>>
>>
>>>
>>>How will this help?
>>>
>>>I win the even.  Five years later I say "no way".  What is the remedy for that?
>>>I have had the title for five years, even if I cheated.  There is no way to go
>>>back and eradicate five years of publicity.
>>
>> If, after five years of publicity, you say "no way", everybody would assume you
>>cheated and that means the rest of your life of negative publicity. That would
>>prove fatal if you were a young programmer trying to get a name in the
>>profession.
>>
>>  Telmo
>
>Perhaps, perhaps not.  There are plenty of "fly-by-night" one-year wonders
>around.  And even ignoring that, who could be sure that what I release 5 years
>into the future was the actual program that won 5 years ago?  Hard to get the
>same hardware, etc...
>
>No way to make this work 5 years hence.  It is hard to make work today...

I guess that it is possible to protect the program by a long passward and give
it in the beginning of the tournament when only after 5 years you tell people
the passward so they can get both the source code and the exe file of the
program.

You should also give the exe file of the program to someone that you trust not
to give it to other people and if it turned out that the exe are different or
that the program cannot play the same moves in almost all cases then it is clear
that you cheated(parallel programs are not 100% deterministic but you can still
see based on the moves if the thing is totally different).

Uri



This page took 0 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.