rybka 1.0 and plaggerism: has the case been compiled yet

[geots]

I know there were some folks who were building a case that rybka 1.0 involved plagersism of fruit code. My understanding is that this case was to be presented to vas so he could respond. I assumed the case would have a detailed description of all (or at least many) of the acts of plaggerism found in rybka 1.0.

Has this case (or letter) been sent to Vas yet, and if so, can it be made public?

I am curious as to what the last month of decompiling work has yielded

best
J

I read about a 1/4 of the reply posts and stopped! Seems like the answer to your question Ozz is that nothing has been done in any way constructive.

Seems to me (a newbie and proud of it) that the only one who should be concerned about this is the auther of Fruit and Rybka. Thats it!

I also want to add that I enjoy the Talkchess forum and have gotten a lot of great ideas for my own engine. Its great to be able to ask many of the really talented "experts" for their opinions and ideas. I log on to this forum almost every nite to see what questions have been posted and to see the responses. Its fun!

I then realize that I really should open my wallet and donate some of my hard earned money to support this forum because I do enjoy it. Then I run into "another" series of posts like this BS and I put my wallet away. Is my money going to help support this crap? No! I don't think so!

I am sorry but I finally had to say something about this. I will not do it again. You people should really grow up.




........A concerned fan of the Talk Chess Forum

AMEN TO THAT! Regardless of your age, you are wise beyond your years. And such a fresh common-sense voice to have here. What a welcome change! Please hang around.

Best,

[Terry McCracken]

What I have seen so far:
There has been enough evidence presented to prove without a shadow of a doubt that Vas studied the UCI parser in Fruit. His is very similar, but not the same. The use of strtok() inline is a good idea, and somewhat unusual.

There are some similar eval terms (of course nearly every engine has some similar eval terms).

There is some similar search functionality (of course every engine has some similar search functionality).

Of course, there is nothing wrong with that.

Perhaps there is some other important evidence I have missed.

I guess that the crimes of which people are raising the red flag there is some merit:
Clearly, Vas has learned a lot from Fruit.

But (of course) every chess programmer has learned a lot from other chess programmers. Has Vas done more than they have?

Trying to stage a trial in the court of public opinion is a bad idea in my eyes.

Even in a court trial, the person performing the examination is not allowed to call for a conclusion on the part of the witness.

I think that if the possibly injured parties are interested, they should be the ones to perform the inquiries, and they should do it through proper channels.

Now, I am not against publication of evidence and allowing us to form our own opinions, as long as it is extremely clear that what we are viewing is not misrepresented.

But proclamations of guilt or innocence are best left to a judge unless the evidence is very, very clear.

And (buy the way) plagiarism clearly does not apply in this case. We are not talking about fraudulent publication of academic or literary works that were copied without permission or acknowledgement.

Copyright violation is a possibility. I am not convinced (in either direction) that something bad has taken place {or has not taken place}.

I think that there is something very sad about this whole process.

I also think that those making charges often have a lot at stake and so it may be impossible for them to be truly neutral, even though they do not know it and really do intend to be fully neutral.

Identical Blocks of Code is Plagiarism.
Dan you're just adding gasoline to the fire. You've given the trolls a green light! Of course you're unaware you've done this.


http://dictionary.reference.com/browse/Plagiarism

[Terry McCracken]

I know there were some folks who were building a case that rybka 1.0 involved plagersism of fruit code. My understanding is that this case was to be presented to vas so he could respond. I assumed the case would have a detailed description of all (or at least many) of the acts of plaggerism found in rybka 1.0.

Has this case (or letter) been sent to Vas yet, and if so, can it be made public?

I am curious as to what the last month of decompiling work has yielded

best
J

I read about a 1/4 of the reply posts and stopped! Seems like the answer to your question Ozz is that nothing has been done in any way constructive.

Seems to me (a newbie and proud of it) that the only one who should be concerned about this is the auther of Fruit and Rybka. Thats it!

I also want to add that I enjoy the Talkchess forum and have gotten a lot of great ideas for my own engine. Its great to be able to ask many of the really talented "experts" for their opinions and ideas. I log on to this forum almost every nite to see what questions have been posted and to see the responses. Its fun!

I then realize that I really should open my wallet and donate some of my hard earned money to support this forum because I do enjoy it. Then I run into "another" series of posts like this BS and I put my wallet away. Is my money going to help support this crap? No! I don't think so!

I am sorry but I finally had to say something about this. I will not do it again. You people should really grow up.




........A concerned fan of the Talk Chess Forum

AMEN TO THAT! Regardless of your age, you are wise beyond your years. And such a fresh common-sense voice to have here. What a welcome change! Please hang around.

Best,

[swami]

Forgot to mention this: The original poster gave moderators a permission to edit the thread title, added "1.0" next to Rybka.

Chris Whittington has posted the Sticky notice that "Discussion is fine, accusation not" So, please, honour this request, and in future, Just add 1.0 next to the engine name whenever you start a whole new thread.

Regards.

[bob]

What I have seen so far:
There has been enough evidence presented to prove without a shadow of a doubt that Vas studied the UCI parser in Fruit. His is very similar, but not the same. The use of strtok() inline is a good idea, and somewhat unusual.

There are some similar eval terms (of course nearly every engine has some similar eval terms).

There is some similar search functionality (of course every engine has some similar search functionality).

Of course, there is nothing wrong with that.

Perhaps there is some other important evidence I have missed.

I guess that the crimes of which people are raising the red flag there is some merit:
Clearly, Vas has learned a lot from Fruit.

But (of course) every chess programmer has learned a lot from other chess programmers. Has Vas done more than they have?

Trying to stage a trial in the court of public opinion is a bad idea in my eyes.

Even in a court trial, the person performing the examination is not allowed to call for a conclusion on the part of the witness.

I think that if the possibly injured parties are interested, they should be the ones to perform the inquiries, and they should do it through proper channels.

Now, I am not against publication of evidence and allowing us to form our own opinions, as long as it is extremely clear that what we are viewing is not misrepresented.

But proclamations of guilt or innocence are best left to a judge unless the evidence is very, very clear.

And (buy the way) plagiarism clearly does not apply in this case. We are not talking about fraudulent publication of academic or literary works that were copied without permission or acknowledgement.

Where on earth does _that_ come from?

Plagiarism is the unauthorized use or close imitation of the language and thoughts of another author and the representation of them as one's own original work.

Within academia, plagiarism by students, professors, or researchers is considered academic dishonesty or academic fraud and offenders are subject to academic censure. In journalism, plagiarism is considered a breach of journalistic ethics, and reporters caught plagiarizing typically face disciplinary measures ranging from suspension to termination. Some individuals caught plagiarizing in academic or journalistic contexts claim that they plagiarized unintentionally, by failing to include quotations or give the appropriate citation. While plagiarism in scholarship and journalism has a centuries-old history, the development of the Internet, where articles appear as electronic text, has made the physical act of copying the work of others much easier, simply by copying and pasting text from one web page to another.

Plagiarism is not copyright infringement. While both terms may apply to a particular act, they are different transgressions. Copyright infringement is a violation of the rights of a copyright holder, when material protected by copyright is used without consent. On the other hand, plagiarism is concerned with the unearned increment to the plagiarizing author's reputation that is achieved through false claims of authorship.

_either_ is bad. Both are related. And one could make the case that if code from fruit was copied, _both_ terms are applicable...

Copyright violation is a possibility. I am not convinced (in either direction) that something bad has taken place {or has not taken place}.

I think that there is something very sad about this whole process.

I also think that those making charges often have a lot at stake and so it may be impossible for them to be truly neutral, even though they do not know it and really do intend to be fully neutral.

[Peter Aloysius]

I have a lot of respect for those involved, but I abhor the way they went about their project.

Just curious, what exactly did we/they do?

It seems to me that whatever genuine efforts anybody had in this were drowned in an ocean of inane argumentation. IOW, neither side was fully to blame for what ensued.

You should build a watertight case and then present it rather than making accusations beforehand.

You do realize that part of "building the case" was done _right_ here. As were prior clone discussions about other programs such as Le Petite, Voyager, Bionic, to name some that were copied from my program. This is the right place to discuss this. It is a "discussion forum" where lots of technically competent people participate... So discussions should not happen on a discussion forum until after the discussions are completed???? How, then, can that happen. How can people join in and participate if they don't know there is something to participate in?

This angle of the discussion is a bit hard to digest...

The difference is that you're dealing with an engine that has gone commercial, so you need to much more careful in how you go about things.
You could be seen as damaging the product as well as the author's reputation. Such aspersions can therefore be seen as legally dubious and therefore against the CCC charter.
That's why only a watertight case should be presented as the first action.
Regards, Graham.

And that's ALREADY done in first place, a part of Rybka source is quite convincing to be similiar to Fruit.

The prove is ALREADY presented. And YOU still reject, claiming that variable name is different, or just concidence, or Vas able to memorize Fruit source code.

You said that watertight case should be presented, but some folks here doing whatever they can to keep that from happen. You guys here make personal attack to people who were trying to present those "watertight case" !!! AND if those watertight case is presented, you still reject it anyway.

I'm not a programmer Peter, but a watertight case means absolutely no doubt whatsoever.
Such a case was not presented as far as I'm aware.

Regards, Graham.

And as I said before, presented those watertight prove is IMPOSSIBLE here, because some folks trying to make personal attack to person who trying to present those prove.

And some other easily refute those watertight with some ridiculous reason. You guy not a programmer, but you think that you were smarter than Bob who is a professor in computer science and already teach programming for 40 years.

If someone here provide those watertight prove, everybody here will scream
"hey, the variable name is different, so it's not watertight" even though changing variable name is the first thing that a cloner do.

See my point?

[Tony Thomas]

Forgot to mention this: The original poster gave moderators a permission to edit the thread title, added "1.0" next to Rybka.

Chris Whittington has posted the Sticky notice that "Discussion is fine, accusation not" So, please, honour this request, and in future, Just add 1.0 next to the engine name whenever you start a whole new thread.

Regards.

Doesnt everyone already know that they are discussing the free version of Rybka? Why add the 1.0? If that's the case, the whole name should be used. Rybka 1.0 beta 32bit/64bit. I am glad the name of the engine wasnt one of the longest english words..or the version number isnt 9999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999.98.

[Uri Blass]

Here are few ideas that you can easily memorize(without talking about code).
1)average between opening and endgame evaluation.
2)evaluating mobilities based on number of squares that the piece can go into them.
3)giving higher weight for mobility of minor pieces.

People can clearly memorize more than it and they can also memorize code patterns(without memorizing the exact source).

ummm, isn't that common sense? Are you gonna say that queen should valued higher than knight is based on Fruit too? Is queen valued 900, rook 500, knight and bishop 300, based on Fruit too?

You can say that if somebody memorize code patterns he is not allowed to use them.
I will express no opinion about it but you have no way to prove that rybka started from fruit if this is what you try to convince us.

I believe that memorizing every line of fruit is clearly impossible for humans but nobody claims that it happened and even some similiar codes that you posted were clearly not identical.

Uri

What I say is, IF, someone find a lot of similiarities code between Fruit and Rybka, folks here WILL say that Vas able to memorize the whole fruit source code, or just coincidence, or variable names is different, or whatever ridiculous reason to defend Rybka.

Well, they did that already.
See : http://64.68.157.89/forum/viewtopic.php ... =&start=20

There's no way they let Rybka and Strelka get verdict as not original, whatever the evidences are, since they were too strong.

And by the way, I DO NOT try to convince you that Rybka is based on Fruit. I just DO NOT like the way you defend Rybka, I don't like all ridiculous reason you made just to defend Rybka, and I don't like personal attack to Bob, Christophe, and other decompiler.

1)It is not obvious to do average between opening evaluation and endgame evaluation based on the stage of the game by common sense.

I did not do it before fruit.

It is also not obvious that you should give knight and bishop mobility bigger weights and I did not do it before reading fruit.

I never evaluated rook as 500 and knight and bishop as 300.

My evaluation of bishop and knight are bigger to prevent bad trade of bishop and knight for rook and pawn.

People also talked about safe mobility so it is not obvious not to evaluate the question if the square that you control is safe or not safe for your pieces.

2)No time to read all this thread but
I do not like the way that some people use to attack Vas and imply that he is quilty with no convincing evidence.

Nobody suggested that Vas was able to memorize everything in fruit
and you suggest that people are going to defend Vas by saying that he remember everything from there.

It is clearly a distortion of the words of people who disagree with you
to claim that people will say that Vas able to memorize the whole fruit source code.

Nobody suggested that Vas was able to memorize the whole fruit source code and I only suggested that it is possible that he was able to memorize some parts.

Distorting my claims to make them absurd and
saying that people who disagree with you make absurd claims is not a good way to convince other people.

Uri

[Uri Blass]

most probably, whatever prove they gonna have, no matter how convincing it, you guys still reject it.

I think people will be responsive to the evidence. You are making an assumption here..



J

in this case, surely not. You guys gonna defend Rybka and Strelka at all cost. Saying it's just purely coincidence, or vas able to memorize all Fruit source code, the decompiler just jealous, or whatever creative reason that might come into your mind.

See people reaction here? Personal attack to Bob, Christophe, Zach, and others?

Rybka is not identical to fruit so no need to memorize everything.
Memorizing some patterns may explain similiarity.

If you read and understand fruit then you can clearly remember some ideas even if you write from scratch.

Here are few ideas that you can easily memorize(without talking about code).
1)average between opening and endgame evaluation.
2)evaluating mobilities based on number of squares that the piece can go into them.
3)giving higher weight for mobility of minor pieces.

People can clearly memorize more than it and they can also memorize code patterns(without memorizing the exact source).


You can say that if somebody memorize code patterns he is not allowed to use them.
I will express no opinion about it but you have no way to prove that rybka started from fruit if this is what you try to convince us.

I believe that memorizing every line of fruit is clearly impossible for humans but nobody claims that it happened and even some similiar codes that you posted were clearly not identical.

Uri

It is _way_ beyond time to stop this line of reasoning. Why don't you take some free-source programs, and look to see how much similarity you find? See if you find identical arrays of evaluation constants. Large blocks of code that are identical. Even an identical bug here and there. You keep saying it is very likely. You should be able to prove that, there are dozens of programs to choose from, so the similarities should be rampant if your supposition is anywhere near reality...

No time for doing it.

Reading and understanding one chess program takes a lot of time
so I have no time to do it for many free source program.

I also do not know if there are identical arrays of evaluation constants for rybka and fruit.

You imply that fruit and rybka have identical arrays of evaluation constants but no proof for it.

I remember that the only array that I suspected based on looking at fruit and strelka could be some piece square tables(did not verify it so I may be wrong) but based on my memory piece square table in these programs are derived from some formula so practically it is less numbers than the number of numbers in the tables.

I remember that most constants are clearly different.

Uri

[ozziejoe]

And some other easily refute those watertight with some ridiculous reason. You guy not a programmer, but you think that you were smarter than Bob who is a professor in computer science and already teach programming for 40 years.

This is an appeal to authority, which is a type of falacious argument. You could also cite the authorities who disagree with bob. Or you could argue that Bob is somehow more of an authority than others like Fabien. No matter how you argue it though, the appeal to authority provides no evidence for a claim


[/quote]