Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Bionic vs Crafty, once again

Author: Bruce Moreland

Date: 11:21:01 01/24/99

Go up one level in this thread



On January 24, 1999 at 03:31:48, Ed Schröder wrote:

>Hans, Johan you have my sympathy whatever Bob says in defence.
>Fact is Crafty source code is freeware. Freeware is freeware. If Bob
>(or others) don't like the negative side effects of freeware then don't
>release it as freeware.
>
>If people pick Crafty's sources, make their own changes and give
>the program an own name then that's perfectly legal. Adding all
>kind of demands to the license agreement are not necessarily
>binding.
>
>As far as I understand Dutch law it is perfectly legal to pick
>Crafty's source-code, make changes, build an own GUI and
>sell it. This might differ from country to country I don't know.
>
>The only thing Bob has a right to demand is that it should be
>forbidden to release a 100% exact copy of the freeware
>sources and give it a new name and/or sell it.
>
>I don't understand all the fuss about this topic. Many programs
>are based on the GNU freeware sources. Never saw discussions
>like this. Why is the GNU status different than Crafty status?
>
>As for tournaments, Crafty or GNU clones should be allowed
>from the juridical point of view as simple as that. I can imagine
>organizers might decide otherwise but they are taking a risk
>concerning the juridical point of view. Freeware is freeware. In
>the Crafty / GNU case all ideas behind the program are made
>public so everybody is allowed to use it. You can not publish
>your ideas in the newspaper and say, "I don't want you to
>use it".

I think that your post makes a lot of strong points but I would like to take the
opportunity to respond.

One issue here involves the legality of selling Crafty.  I don't know that
Bionic Impakt is for sale, so absent information that it is, this can be
discussed as a hypothetical case.  Bob apparently doesn't want Crafty to be
sold, and that is enough for me, and my interest doesn't go much beyond that
point.

Another issue involves entrance of derived programs in author-only tournaments.
If someone wants to run a tournament in their garage, with whatever software
they have bought, or downloaded (legally, one would hope) from the net, that is
not an issue that interests me.  What interests me is the tournaments where you
enter your *own* program, meaning one that you wrote.

I think this is a continuum issue.  Obviouisly, it is allowed to enter a program
that you wrote 100% yourself.  And I think just as obviously, it is not allowed
to enter a program that you downloaded and made no changes to.

I talk to Bob, and many others, and we exchange ideas.  I don't think this
should disqualify anyone from claiming sole authorship of a program.

I don't think that anyone should have issue with some limited snipping, either,
or perhaps sharing of opening book or endgame database code or data, although I
can imagine protests raised about this as well, if a well-crafted book finds its
way into multiple programs, or if someone devises a good heuristic approach to
choosing between moves in a drawn endgame database ending, and distributes that
in source library form.

"Starting from Gnuchess" was, I gather, something people commonly did in the
late 80's and early 90's.  Since Gnuchess has stagnated for the last several
years, and since it wasn't very strong in the first place, I think this is a
less threatening situation than someone starting with Crafty and keeping current
on new releases.  If  you have a strong program that is Gnuchess-based, I would
argue that it probably doesn't look much like Gnuchess anymore.  It may have
been an interesting issue to discuss at the time, but I don't think that
Gnuchess is much threat anymore.  (Lest anyone wonder, my own program is *not*
derived from anything.)

The components of Crafty are more thoroughly developed than the components of
Gnuchess.  The search is stronger and more complex, as is the eval function.
The make/unmake move apparatus is more complicated and gives you more
information that you can use in the eval function.

The thing as a whole is much stronger than Gnuchess, and here is a critical
point -- it is stronger than the typical amateur program.  Someone can replace
the evaluation component, and fiddle with some other components, and if any
degree of care is taken, there is not much risk of seriously reducing the
program's strength.

Assuming that this is done, and the program is entered in one of these
author-only tournaments, there are two groups of people who are immediately at
risk of suffering because of this.

The first is Bob himself.  Bob is author of Crafty.  I would argue that Bob is
at very least a co-author of anything that is derived from Crafty.  The
components of Crafty are complicated enough, and required enough work to create,
that anything that starts with Crafty as a base has to have benefited
tremendously from Bob's work.  You could take out everything except move
generation and the make/unmake move apparatus, and there is still a large volume
of work there.  If there is a question of multiple entries, I would argue that
Bob's entry has precedence, and I think I can also argue that Bob, as co-author
of the derived Crafties, could disavow one of these entries and stop it from
competing in an author-entry tournament.  He wrote the program, he should have
some say in whether it competes.  *Especially* since the program in question
here isn't even split off from Crafty, he's still incorporating new
code(parallel search) as Bob writes it.

The other group is the amateurs who do choose to do this on their own, without
starting with a top program as a base.  A lot of these guys have programs that
are weaker than Crafty.  There are lots of reasons for this, which I don't need
to enumerate, but one that I would like to mention is sheer iconoclasm.  I have
often said, with a degree of seriousness, that if I ask someone to tell me which
feature of the program makes them most proud, they will describe to me, in
loving detail, something that is actually making their program weaker, and which
objectively they should remove.  Maybe they will prove me wrong, I would be
happy to see this and I wish them all luck.

I think that this group of amateurs should be encouraged to go their own way,
and I don't think acceptance of entries derived from strong programs such as
Crafty does anything to achieve this.

This is distinct from "taking ideas" from Crafty.  As far as I know, that is
what the program is there for.  "Taking ideas" is a lot different than
"modifying and compiling".  That's a fine thing to do, but I don't think that it
qualifies you for entrance into an author-entry tournament, nor does it
eliminate all of Bob's rights as original author.  Those are my points.

bruce



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.