tiger wrote:Alex your analysis is very interesting.
However I believe that the word "clone" in the title of your message could be misleading and the discussion could degenerate because of this.
You are comparing Fruit 2.1 and Rybka 1.0 outputs, and I think that nobody who has seriously compared the two programs could say that they are clones.
Rybka is clearly not a clone of Fruit, so the word "clone" in your message title is not going to help looking at the matter in an objective way.
I wanted to point out this because in earlier discussions where I was talking about the similarities between Fruit and Rybka, people immediately thought I was bringing back a CLONE accusation.
Actually, what I believe is that Rybka 1.0 started its life as Fruit 2.1. I mean that at some time T=0, probably in the middle of 2005, the source code of Fruit has been taken as the base and has been little by little modified until it became Rybka 1.0 as we know it.
Siginficant and brilliant additions have been made to that code. The end result is a program that is much stronger than the original program.
So in my opinion:
- Rybka 1.0 is not a clone of Fruit 2.1, it is somehow different and better
- Rybka 1.0 started as the source code of Fruit 2.1
I think your analysis tends to confirm this opinion. It is a piece of the puzzle. Of course more analysis, in particular of the sources codes, is necessary to confirm it completely.
// Christophe
Getting soft Christophe,I wonder what's the reaon
The reason is that the previous discussion has degenerated and some people have been distorting my words.
I have never used the word clone and if you read what I wrote before I have never even meant that X was a clone of Y.
What I have said before is that I think X started its life as being the source code of Y. It's totally different and has different implications and consequences.
Somehow the word "clone" seems to ring a bell and drive everybody crazy. I don't want people to put this word in my mouth when I have NEVER said it, and NEVER wanted to imply it.
tiger wrote:So in my opinion:
- Rybka 1.0 is not a clone of Fruit 2.1, it is somehow different and better
- Rybka 1.0 started as the source code of Fruit 2.1
We have a different opinion what clone means. Starting with the source code would violate the GPL even if every single line is changed. So I would call this cloneing.
But as I stated several times before, it must not be illegal in case the original author agrees.
Alex
There is a word for what you describe (starting from source code and modifying it). It is called "derived work" in the GPL v2 licence that protects the source code of Fruit 2.1:
tiger wrote:Alex your analysis is very interesting.
However I believe that the word "clone" in the title of your message could be misleading and the discussion could degenerate because of this.
You are comparing Fruit 2.1 and Rybka 1.0 outputs, and I think that nobody who has seriously compared the two programs could say that they are clones.
Rybka is clearly not a clone of Fruit, so the word "clone" in your message title is not going to help looking at the matter in an objective way.
I wanted to point out this because in earlier discussions where I was talking about the similarities between Fruit and Rybka, people immediately thought I was bringing back a CLONE accusation.
Actually, what I believe is that Rybka 1.0 started its life as Fruit 2.1. I mean that at some time T=0, probably in the middle of 2005, the source code of Fruit has been taken as the base and has been little by little modified until it became Rybka 1.0 as we know it.
Siginficant and brilliant additions have been made to that code. The end result is a program that is much stronger than the original program.
So in my opinion:
- Rybka 1.0 is not a clone of Fruit 2.1, it is somehow different and better
- Rybka 1.0 started as the source code of Fruit 2.1
I think your analysis tends to confirm this opinion. It is a piece of the puzzle. Of course more analysis, in particular of the sources codes, is necessary to confirm it completely.
// Christophe
Getting soft Christophe,I wonder what's the reaon
The reason is that the previous discussion has degenerated and some people have been distorting my words.
I have never used the word clone and if you read what I wrote before I have never even meant that X was a clone of Y.
What I have said before is that I think X started its life as being the source code of Y. It's totally different and has different implications and consequences.
Somehow the word "clone" seems to ring a bell and drive everybody crazy. I don't want people to put this word in my mouth when I have NEVER said it, and NEVER wanted to imply it.
// Christophe
If I remember correctly, Vas stated that he studied Fruit very carefully and that, as a consequence, he was able to strengthen Rybka by about 30 points..................
chrisw wrote:Well, if the report of Vas's comment (did not look at Fruit UCI output) is true, you have two possibilities
The comment itself is true, in which case the matches are perhaps curious.
The comment is false, in which case he quite reasonably used Fruit as a development target.
I would prefer to imagine he fibbed over which engines he concentrated his firepower on rather than the alternative assumption that he copied code. It's not entirely unimaginable he might have wanted to keep quiet over his development procedures and targets, is it not?
No idea. I try to avoid getting into "what was he thinking" and such as there is no scientific way to prove anything. However, if there are inconsistencies is lots of statements, then it does tend to leave an impression that all is not as it should be... even if it is speculation...
tiger wrote:Alex your analysis is very interesting.
However I believe that the word "clone" in the title of your message could be misleading and the discussion could degenerate because of this.
You are comparing Fruit 2.1 and Rybka 1.0 outputs, and I think that nobody who has seriously compared the two programs could say that they are clones.
Rybka is clearly not a clone of Fruit, so the word "clone" in your message title is not going to help looking at the matter in an objective way.
I wanted to point out this because in earlier discussions where I was talking about the similarities between Fruit and Rybka, people immediately thought I was bringing back a CLONE accusation.
Actually, what I believe is that Rybka 1.0 started its life as Fruit 2.1. I mean that at some time T=0, probably in the middle of 2005, the source code of Fruit has been taken as the base and has been little by little modified until it became Rybka 1.0 as we know it.
Siginficant and brilliant additions have been made to that code. The end result is a program that is much stronger than the original program.
So in my opinion:
- Rybka 1.0 is not a clone of Fruit 2.1, it is somehow different and better
- Rybka 1.0 started as the source code of Fruit 2.1
I think your analysis tends to confirm this opinion. It is a piece of the puzzle. Of course more analysis, in particular of the sources codes, is necessary to confirm it completely.
// Christophe
Getting soft Christophe,I wonder what's the reaon
The reason is that the previous discussion has degenerated and some people have been distorting my words.
I have never used the word clone and if you read what I wrote before I have never even meant that X was a clone of Y.
What I have said before is that I think X started its life as being the source code of Y. It's totally different and has different implications and consequences.
Somehow the word "clone" seems to ring a bell and drive everybody crazy. I don't want people to put this word in my mouth when I have NEVER said it, and NEVER wanted to imply it.
// Christophe
I support that too, I have never seen you using the cl___ word.
But didnt you attack Vas at all? Yes. When you wrote that you could also give Tiger a 200+ infusion if you did what Vas had taken from Fruit but you didnt want to do it. Was that reasonable towards a commercial collegue? The message appeared in a moment when the "wrong" of Vas has been discussed. Today we know, he didnt do anything wrong at all.
The second case when you insulted everybody pro Vas as ignorant who only had found a new God like substitute. That was a bit too much with ignorant. This is all deleted and therefore you can continue to ask what you had done at all.
-Popper and Lakatos are good but I'm stuck on Leibowitz
tiger wrote:So in my opinion:
- Rybka 1.0 is not a clone of Fruit 2.1, it is somehow different and better
- Rybka 1.0 started as the source code of Fruit 2.1
We have a different opinion what clone means. Starting with the source code would violate the GPL even if every single line is changed. So I would call this cloneing.
But as I stated several times before, it must not be illegal in case the original author agrees.
Alex
There is a word for what you describe (starting from source code and modifying it). It is called "derived work" in the GPL v2 licence that protects the source code of Fruit 2.1:
I believe that talking about "derived work" is much more appropriate because it seems to describe better what happened.
// Christophe
Christophe,
From the licence link above, clause 0:
0. This License applies to any program or other work which contains a notice placed by the copyright holder saying it may be distributed under the terms of this General Public License. The "Program", below, refers to any such program or work, and a "work based on the Program" means either the Program or any derivative work under copyright law: that is to say, a work containing the Program or a portion of it, either verbatim or with modifications and/or translated into another language. (Hereinafter, translation is included without limitation in the term "modification".) Each licensee is addressed as "you".
Note a "work based on the program" is a "work containing the Program or a portion of it".
So, hypothetically, if a programmer starts with Fruit (let's say) and gradually modifies ALL of it with his own work, he replaces the movegen with his own movegen and so on with all functions, then this work does not "contain the Program or a portion of it" is is not covered any longer by the GPL licence.
It looks ok to take the code and re-write it bit by bit, using it, if you like, as a test bed for development as the programmer goes along, until it is completely turned into something else. Such a development process seems to breach no licence. Or?
Alexander Schmidt wrote:OK, I will name it like that in the future.
It is of course a thin line between reusing concepts and/or little parts of an open source program and then making or finalizing your own work, or first copy the open source stuff and then modify it incrementally and maybe ask money for it. I can not tell what has been done with Rybka. Maybe it it is in the margin of the borderline of violation. I would just like to here the view of Vas on this matter. The guy is brilliant, no doubt, but maybe just maybe he crossed the line somewhere. For all security: this is about the process and not the person.
Alexander Schmidt wrote:OK, I will name it like that in the future.
It is of course a thin line between reusing concepts and/or little parts of an open source program and then making or finalizing your own work, or first copy the open source stuff and then modify it incrementally and maybe ask money for it. I can not tell what has been done with Rybka. Maybe it it is in the margin of the borderline of violation. I would just like to here the view of Vas on this matter. The guy is brilliant, no doubt, but maybe just maybe he crossed the line somewhere. For all security: this is about the process and not the person.
Best regards,
Kaj
You must willy nilly enter the Rybka forum. He doesnt write here. And I think it cant be expected after the actual campaign. IMO he has already made enough statements. Therefore the actual movement in CCC is absolutely unmotivated unless people want to harm the Rybka 3 sales who just begun.
H9ow can someone decent accuse a collegue for something what he potentially has done himself. See the displayed values in Tiger 12,13,14. CT simply ignores the question.
-Popper and Lakatos are good but I'm stuck on Leibowitz
PLEASE stop using Quotes when you reply to rather offending posts.
I have seen yet more insinuations in this thread and I deleted it, There were complaint reports about the clone insinuations (yet again).
About 8 posts followed up using direct Quotes of the offending texts but making reasonable posts...such as posts from Uri, Chris, Bob and Zach.
I'll quote their posts in the follow up.
Moderation tools, available to moderators, is really poor.
