Re: [computer-go] KGS Tournaments: Uniqueness of players

[Don Dailey <drd@xxxxxxx>]

> I think the big things you want to prohibit are plagiarism and random
> people entering copies of publically available programs.  


I agree very much.


> My guess is that there actually won't be that many cases that are
> actually in question.

I think your guess is a bad one :-)   Here's why I say that:

Practical lessons from computer chess indicate that this is very
likely to be a big problem.  Don't ask me why, but people love to take
public domain works like popular chess programs, rename them, register
them on servers and wrap their ego's around them.   I guarantee this
will happen with Computer Go.   

I think at some point soon, especially with popularity of the GTP
protocol and program that use it, you will find this happening a great
deal.

It may be possible to handle the most flagrant cases of this by
requiring a more complicated registration process.  Once an author is
trusted, it shouldn't be a problem.  For instance if David Fotland
enters a tournament we could consider him a certified author and trust
him.  But if John Doe enters a tournament, we might ask for some
validation process, such as sending copies of his binaries to a
trusted validator, such as the tournament directory.  The binary could
be checked for similarity to popular public doman (or otherwise)
programs or could simply be tested.    The fact that you have to 
send a binary will serve as a significant deterrent for most. 

Similar to GPG encryption,  trust could be passed along to other
potentially new authors.   In other words, a new author might
be vouched for by other author(s).     

Their might be serveral ways to validate an author or entry to at
least minimize the problem of having 6 GnuGo's enter a tournament.

I do agree that derivative works should be allowed.  If someone
modifies GnuGo in interesting and original ways, it is good for
computer go and should be allowed.  The bottom line is that some
author should have put substantial effort into some project in order
to compete.

- Don






_______________________________________________
computer-go mailing list
computer-go@xxxxxxxxxxxxxxxxx
http://www.computer-go.org/mailman/listinfo/computer-go/