Page 1 of 3

Fruit, Rybka, Strelka final conclusion! (I hope)

Posted: Fri Jul 13, 2007 11:08 pm
by Michael Sherwin
Fruit is not a very original work. Even Fabien has said that there is nothing new in Fruit!

Therefore there is nothing in Fruit that can be protected under the FSF GPL. Nothing that can be licensed.

Copyright law is the only protection that Fabien has to protect Fruit from cloners.

Since, Rybka is only derived from fruit and not copied. It is totally legal.

Rybka is a derivation and not a clone.

Strelka 1.0 is only suspect because of the data tables that it used from Rybka. Data is data and is not copyrightable no matter where it came from. Just because Vasik gathered the data does not mean that others can not use the data. The data already existed in the database. That is why it is called a 'data'-base. Vasik did not create the data.

Strelka is a derivation and not a clone!

END OF STORY! :D ! :D ! :D

Re: Fruit, Rybka, Strelka final conclusion! (I hope)

Posted: Fri Jul 13, 2007 11:31 pm
by GenoM
nice try :D :D :D

Re: Fruit, Rybka, Strelka final conclusion! (I hope)

Posted: Fri Jul 13, 2007 11:35 pm
by George Tsavdaris
Michael Sherwin wrote: Since, Rybka is only derived from fruit and not copied. It is totally legal.
How do you know that Rybka is derived from Fruit?

And what exactly is the meaning of: "Engine-A is derived from engine-B"?

Re: Fruit, Rybka, Strelka final conclusion! (I hope)

Posted: Fri Jul 13, 2007 11:36 pm
by gerold
GenoM wrote:nice try :D :D :D
I agree Geno. NIce try Mike. :) :) :)

Still the fellow spoke to soon. If he would have kept his
mouth shut it would have been hard to prove that
it is a clone.

Gerold.

Re: Fruit, Rybka, Strelka final conclusion! (I hope)

Posted: Fri Jul 13, 2007 11:48 pm
by Onno Garms
Michael Sherwin wrote:Fruit is not a very original work. Even Fabien has said that there is nothing new in Fruit!

Therefore there is nothing in Fruit that can be protected under the FSF GPL. Nothing that can be licensed.
My irony detector gives unclear results on this statement, so let me assume you mean above seriously.

Certainly Fabien meant that there are no new algorithms in Fruit. (Algorithms are not GPL'able anyway.) Still Fruit are his lines of code and hence certainly licensable and protected by copyright.
Copyright law is the only protection that Fabien has to protect Fruit from cloners.
Everybody can choose between accepting the GPL or rejecting it and being limitted by copyright law. This is always the case not just for Fruit.
Since, Rybka is only derived from fruit and not copied. It is totally legal.
Rybka is a derivation and not a clone.
"derived from" is confusing. Derivative work in the sense of a modified copy is protected by the GPL. Strelka author says that Rybka were a rewrite of Fruit. This would not be derivative work in the sense of the GPL and hence would be legal. I personally believe that the Strelka author says the truth, but that is only my personal oppinion.
Strelka 1.0 is only suspect because of the data tables that it used from Rybka. Data is data and is not copyrightable no matter where it came from.
I don't know.

Re: Fruit, Rybka, Strelka final conclusion! (I hope)

Posted: Sat Jul 14, 2007 12:09 am
by Alessandro Scotti
Michael Sherwin wrote:Strelka 1.0 is only suspect because of the data tables that it used from Rybka. Data is data and is not copyrightable no matter where it came from. Just because Vasik gathered the data does not mean that others can not use the data. The data already existed in the database. That is why it is called a 'data'-base. Vasik did not create the data.
I would not be so sure about that, things are far more complicated.

For a start, in the digital era everything is data.

Now take a commercial song, make an MP3 out of it, encrypt it and swap even and odd bytes. Looks like random data eh? But it's not, and you'll be infringing copyright by distributing it.

Or consider the football championship. You have friends in the stadiums that tell you how games are going in real-time, by phone. You use this data to update and internet site, also in real-time. Should be fine, eh? After all it's just data, collected independently, about something that is happening in the world. Knock, knock... cops at the door! (YMMV, might be country-dependent).

One last... just look at your arm and go deeper: skin, tissue, cells, DNA, genes, patents. Ops! Those are hardly original are they? Nature anyone? Yet some genes are patentable under current law, incredible!

Vasik must have found a way to compute efficiently coefficients and tables to use for evaluating a chess position. Maybe he started from public data, I don't know. But nonetheless I would think twice before assuming that what he got is not copyrightable.

Re: Fruit, Rybka, Strelka final conclusion! (I hope)

Posted: Sat Jul 14, 2007 12:12 am
by Michael Sherwin
GenoM wrote:nice try :D :D :D
Thanks! :D

Re: Fruit, Rybka, Strelka final conclusion! (I hope)

Posted: Sat Jul 14, 2007 12:19 am
by Michael Sherwin
Alessandro Scotti wrote:
Michael Sherwin wrote:Strelka 1.0 is only suspect because of the data tables that it used from Rybka. Data is data and is not copyrightable no matter where it came from. Just because Vasik gathered the data does not mean that others can not use the data. The data already existed in the database. That is why it is called a 'data'-base. Vasik did not create the data.
I would not be so sure about that, things are far more complicated.

For a start, in the digital era everything is data.

Now take a commercial song, make an MP3 out of it, encrypt it and swap even and odd bytes. Looks like random data eh? But it's not, and you'll be infringing copyright by distributing it.

Or consider the football championship. You have friends in the stadiums that tell you how games are going in real-time, by phone. You use this data to update and internet site, also in real-time. Should be fine, eh? After all it's just data, collected independently, about something that is happening in the world. Knock, knock... cops at the door! (YMMV, might be country-dependent).

One last... just look at your arm and go deeper: skin, tissue, cells, DNA, genes, patents. Ops! Those are hardly original are they? Nature anyone? Yet some genes are patentable under current law, incredible!

Vasik must have found a way to compute efficiently coefficients and tables to use for evaluating a chess position. Maybe he started from public data, I don't know. But nonetheless I would think twice before assuming that what he got is not copyrightable.
Then we are destined to ride this merry-go-round forever! :evil:

But, in strelka 1.8 Juri has done away with the data tables and replaced them with functional equivilents so the point is now mute.

Strelka 1.8 is a derivitive of Fruit and Rybka and is not a clone!

Re: Fruit, Rybka, Strelka final conclusion! (I hope)

Posted: Sat Jul 14, 2007 12:26 am
by Michael Sherwin
George Tsavdaris wrote:
Michael Sherwin wrote: Since, Rybka is only derived from fruit and not copied. It is totally legal.
How do you know that Rybka is derived from Fruit?

And what exactly is the meaning of: "Engine-A is derived from engine-B"?
It is just rather obvious. Vasik admitted to taking many things from Fruit.

Which are you asking about, the legal or common definition of derivitive as applies to the GPL?

Re: Fruit, Rybka, Strelka final conclusion! (I hope)

Posted: Sat Jul 14, 2007 12:36 am
by CRoberson
You miss the point.

Decoding a binary and taking knowledge out of it is theft unless
permission is given.