what can we learn from Rybka/IPPLITE

Discussion of chess software programming and technical issues.

Moderators: hgm, Rebel, chrisw

liuzy

what can we learn from Rybka/IPPLITE

Post by liuzy »

Now we know the secret of Rybka, but what can we learn from it. Can we use the code of IPPLITE directly in SF or TG.
Dann Corbit
Posts: 12540
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: what can we learn from Rybka/IPPLITE

Post by Dann Corbit »

liuzy wrote:Now we know the secret of Rybka, but what can we learn from it. Can we use the code of IPPLITE directly in SF or TG.
We have learned that if people are too lazy to think for themselves, we can steal from those who are smarter than we are.
Gian-Carlo Pascutto
Posts: 1243
Joined: Sat Dec 13, 2008 7:00 pm

Re: what can we learn from Rybka/IPPLITE

Post by Gian-Carlo Pascutto »

Dann Corbit wrote:
liuzy wrote:Now we know the secret of Rybka, but what can we learn from it. Can we use the code of IPPLITE directly in SF or TG.
We have learned that if people are too lazy to think for themselves, we can steal from those who are smarter than we are.
I would have stated "program" instead of "think". There was never anything wrong with stealing ideas.

Stockfish can probably improved by the ideas of Ippolit. But you won't be able to reuse code directly (doing so would be copyright infringement anyway).
Fguy64
Posts: 814
Joined: Sat May 09, 2009 4:51 pm
Location: Toronto

Re: what can we learn from Rybka/IPPLITE

Post by Fguy64 »

Gian-Carlo Pascutto wrote: ...
I would have stated "program" instead of "think". There was never anything wrong with stealing ideas.
...
I was gonna say. It's a grey area. But thank God for the ideas of others, there's no way I would have worked out alpha-beta pruning on my own.

So when when does an idea, or a technique, become public property? I don't have an answer. I do have a lot of respect for the real creations of others though. And if I ever come up with an original idea that others use, I'd be thrilled. I've certainly used lots of theirs, so It's the least I can do to give back.
bob
Posts: 20943
Joined: Mon Feb 27, 2006 7:30 pm
Location: Birmingham, AL

Re: what can we learn from Rybka/IPPLITE

Post by bob »

Dann Corbit wrote:
liuzy wrote:Now we know the secret of Rybka, but what can we learn from it. Can we use the code of IPPLITE directly in SF or TG.
We have learned that if people are too lazy to think for themselves, we can steal from those who are smarter than we are.
Careful, that can cut both ways in _this_ discussion (who stole from who, for example). :)
adieguez

Re: what can we learn from Rybka/IPPLITE

Post by adieguez »

Gian-Carlo Pascutto wrote:
Dann Corbit wrote:
liuzy wrote:Now we know the secret of Rybka, but what can we learn from it. Can we use the code of IPPLITE directly in SF or TG.
We have learned that if people are too lazy to think for themselves, we can steal from those who are smarter than we are.
I would have stated "program" instead of "think". There was never anything wrong with stealing ideas.

Stockfish can probably improved by the ideas of Ippolit. But you won't be able to reuse code directly (doing so would be copyright infringement anyway).
but the ideas are almost everything. What you do when examining a source code? even if thinking maybe: I will take a note of the general ideas and redo the implementation details, it is still a grey area, because, is everyone going to do the same or not? and what is a general idea? I try to be positive but getting inspiration looking directly source code has never been the hobby as I like it. btw obviously just board rep, language programming etc doesn't matter that much, if a program is 2 or 3 times faster than another, that is not a so dramatical difference in strength alone.

I have an idea. Let forget about ippolit source, and let its authors explain some of his ideas here :) that's more normal, everybody happy. Ok, lets wait for Mister Pescatore :)
jwes
Posts: 778
Joined: Sat Jul 01, 2006 7:11 am

Re: what can we learn from Rybka/IPPLITE

Post by jwes »

Fguy64 wrote:
Gian-Carlo Pascutto wrote: ...
I would have stated "program" instead of "think". There was never anything wrong with stealing ideas.
...
I was gonna say. It's a grey area. But thank God for the ideas of others, there's no way I would have worked out alpha-beta pruning on my own.

So when when does an idea, or a technique, become public property? I don't have an answer. I do have a lot of respect for the real creations of others though. And if I ever come up with an original idea that others use, I'd be thrilled. I've certainly used lots of theirs, so It's the least I can do to give back.
Ideas are always public property. Copyright protects a particular expression of an idea, patents protect implementations of an idea, and trade secrets protect against certain ways of discovering an idea. Currently software is patentable in some countries including the USA, but there is a case before the Supreme Court that may change this.
Fguy64
Posts: 814
Joined: Sat May 09, 2009 4:51 pm
Location: Toronto

Re: what can we learn from Rybka/IPPLITE

Post by Fguy64 »

jwes wrote:
Fguy64 wrote:
Gian-Carlo Pascutto wrote: ...
I would have stated "program" instead of "think". There was never anything wrong with stealing ideas.
...
I was gonna say. It's a grey area. But thank God for the ideas of others, there's no way I would have worked out alpha-beta pruning on my own.

So when when does an idea, or a technique, become public property? I don't have an answer. I do have a lot of respect for the real creations of others though. And if I ever come up with an original idea that others use, I'd be thrilled. I've certainly used lots of theirs, so It's the least I can do to give back.
Ideas are always public property. Copyright protects a particular expression of an idea, patents protect implementations of an idea, and trade secrets protect against certain ways of discovering an idea. Currently software is patentable in some countries including the USA, but there is a case before the Supreme Court that may change this.
Fair enough jwes, but I suggest that this kind of begs the question. I'll change my question to what is an idea? Is a programming technique an idea? Is an algorithm an idea? An optimization strategy? What about a compression format? When is something an invention and not a discovery? I find these sorts of questions very interesting, and the answers are not so clear.

regards.