Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Bionic vs Crafty, once again

Author: Bruce Moreland

Date: 10:36:18 01/25/99

Go up one level in this thread



On January 25, 1999 at 08:47:11, Enrique Irazoqui wrote:

>Aside from legal and moral issues, I find confusing the meaning of freeware or
>public domain. Opening books are built from human theory and used by every
>program. I have never seen opening lines in computer games quoted "as in the
>game Karpov-Ivanchuk, Linares '92". Tablebases are widely used too by
>programmers that had nothing to do with their development. These are two quick
>examples of "public domain" used freely (as in free-ware?) by everybody.  I
>guess I am being naive.

A challenging case involving computer openiing books is when someone constructs
a book, and the book is used by more than one program.  There are rules against
multiple authorship, and the question is whether they should apply in this case.
 Personally I would argue that they do, but I think I would lose.

The case involving endgame databases is also interesting.  The databases come as
a package, you get the data and some code that can perform a lookup on a
particular position.

It is an interesting exercise to do this yourself, and if the rule were changed
so that you have to do it yourself, it would benefit me, but I don't think that
I am willing to argue that the rules should make you do it yourself.
Reconstruction of all of the endgame databases by everyone would not do much to
advance the field, I think.  The outcomes of all of those endings are known,
they are solved perfectly enough, it's probably not useful to make everyone else
do it again.

Now, if we're talking about a system that lets you decide which drawn move to
make, in order to maximize winning chances, I think that should come under the
multiple authorship rule.  That's significant code, and there is more than one
method of doing that, and there should be incentive to allow these methods to
compete, rather than letting everyone take the first one that is published and
incorporate it.

>There is a famous case in literature. When Thomas Mann published his "Doktor
>Faustus", he was immediately accused of unethical behavior because he put in the
>mouth of Leverkün, one of the main characters,  the musical theories of
>Schönberg and of the Frankfurt School, notably Adorno, without ever mentioning
>his sources. Had he given credit to Adorno and Schönberg, no one could have said
>anything about Mann's use or development of their theories. Does this apply to
>the discussion in this thread? Imagine that a programmer uses your code as the
>basis for his new program and introduces some modifications here and there, in
>evaluations functions, wherever. Then he gives you credit for the source and
>enters the new hybrid in a tournament for testing purposes. All programmers want
>to see how successful is their baby. Would this be allowed? If not, what are the
>limitations of freeware (free?) and what is the practical use of freeware if not
>"take it and develop it in any way you want" (free)?

If Bob competes in these tournaments I think it is extremely clear that he
should have precedence, and that if he is given credit, that makes him a
co-author in my opinion, and the rule against multiple entries comes into
effect.

I think that Bob should always have precedence in these cases, because to do
otherwise is completely silly.  To say that Bob can't compete because someone
else has already entered a slightly modified Crafty wouldn't make any sense.

If Bob doesn't compete, I think things get a little stickier.  Personally I
think that Bob should have right of approval of these entries.  Bob has sent
representatives to tournaments in the past, I don't see any problem with his
sending a representative who has made changes to the code.  But several of them,
no way.  One Crafty per tournament is enough.

>I understand that all this thread started as a discussion about Bionic. I read
>that Hans gave you due credit. I am not questioning Bionic or Hans. I am only
>trying to understand what is "freeware" in general and as a learning/developing
>tool.

http://www.impakt.be/bionic/index.html

If you look at that web page, most references to Crafty are as a punching bag,
and I didn't see any reference at all to Bob.  He certainly wasn't listed as an
author or co-author.

I am not so much interested in this as I am in making sure that the ICCA
tournament fields are as diverse as possible.  I don't want a return to the days
when companies entered their programs several times.

bruce

>
>Enrique



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.