Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Open Source Ideas Into Commercial Programs?

Author: Robert Hyatt

Date: 10:50:07 08/24/05

Go up one level in this thread


On August 24, 2005 at 12:49:55, Rolf Tueschen wrote:

>Basic question:
>
>isn't it absurd that the commercial programs have the open sources for improving
>but that they themselves dont give away new ideas? What indications do you have
>that these secretive commercials have anything original at all and NOT only
>clever applications of known ideas out of public sources?
>
>Or is it for you a convincing situation if a commercial wins repeatedly a
>tournament?
>
>Could you add a few clarifying lines?


Here's my thought.  Over the past 35 years of my computer chess involvement, I
have shared ideas with many computer chess enthusiasts...  They have given me
ideas, I have given them ideas.  Etc.  Every year at the ACM events, the
participants use to sit around and discuss what was new for this year, knowing
no one would try to implement something new in the middle of the event.  And
knowing that even if they did copy that idea for next year, we'd come up with
something new anyway.

So yes, a commercial programmer has likely looked at open-source programs.  Most
are honest enough and admit this freely.  And yes, they take ideas that they
haven't done.  I can think of several "crafty" ideas that showed up in new
commercial programs after the idea was released in Crafty's source.  Outside
passed pawns.  Pawn Majorities.  Etc.

And yes, a commercial programmer is likely to come up with some unique ideas
along the way, and if he doesn't reveal those, they become his "edge".  Until
they are eventually discovered either by hard work or serendipity, at which
point the commercial programmer has to come up with new ideas or else resign
himself to being no better than anyone else...

It's been a 1-way street.  And as I said many times over the past few years, if
enough amateur authors get together to share ideas, before long the commercial
programs will be no better than the best amateurs, and it won't be long after
that until they are actually worse.  We might be beginning to see this happen
today.

Of course, if you choose to not reveal ideas, it is your own personal choice.
To the detriment of computer chess overall.  For example, what 1980's programmer
didn't read the Slate/Atkin chaper in Chess Skill and use that as a basic
frame-work for a program?  Or if they didn't, did they look at GnuChess?

My only issue is that it seems somewhat amoral to take and not give back.
"Giving back" does have its own set of issues.  But this is the norm for all
scientific developments.  Otherwise only Intel could make ICs, only IBM would be
making computers, There would be no Sony since they didn't develop the
transistor, etc...



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.