FSF has not taken the Rybka issue to court, and there has NOT been a judgement on any GPL violation. Yet we have Vas as tried and found guilty and equated to a "thief" and a "murderer". It is as if all he did was just copy Fruit and Crafty and came up with a hybrid program that played a few 100 elos stronger, with absolutely no input from his side. I think since none of us are lawyers, it is a good idea to do some more research on GPL license and what happens if there is a violation. Another thing, it seems that Vas would have an excellent case to sue the ICGA for defamation, and to ask for damages from all those involved. While ICGA might and can decide to give awards or remove awards, what it cannot do is defame a person on its own. That has clearly happened without proper legal proceedings, and I am sure it would be easy to trace whether Levy and other ICGA participants were guilty of defamation as they actively contacted news media. In the case of ICGA I think it would be very easy to prove that this was done in order to gain publicity ... very similar to a tabloid newspaper claiming false and unsubstantiated claims to sell a few more papers. I really hope that Vas follows this up. Here is a good link that has some good data on GPL and if you post on this thread, please try to make it civil and sane. Remember it is ok to have a different opinion other than your own!
>It is unfortunate to see people that were friends and partners in computer chess go at each others throats.
Yes I totally agree. People who you thought decent act like primates "lower" than man. However, I see this as part of the fallout from one man's actions (ignoring the fact we should be able to discuss this as men and not, as so many have done here, as children without real knowledge and understanding of the matter).
>A lot of mud has been thrown around and a lot of animosity has again been allowed to surface between 2 groups ... either for or against Vas
Well put me in a "third" group Majd... I am certainly not "against" Vas.
I merely do not have any appreciation or regard for what he has done. That I am certainly against.
It is not so petty as he broke the rules of some board game like Monopoly (Vas! Saw you take that $500 from the "Bank" now put it back!)
And it is not so heinous as if he was some securities investor / broker and swindled tens or hundreds of millions of dollars, taking people's pensions, lifetime savings etc. thru some elaborate ponzi scheme.
It is what it is, and it stands (in this case falls) on its own merits. We all see the consequences, some of us more aware than others, none of us able to enumerate them all.
>after reading some comparisons to "murder"
Of course that is over the top... dunno I saw Vas compared to a murderer tho. Yep the word was mentioned... that makes it a comparison? Impossible to glean something from a sentence that uses the word? You see the word "murder" and that causes you to stumble so that you cannot determine what is actually being said? Someone uses the word "root beer" and now we suddenly are comparing Vas to a frosty mug of root beer? C'mon.
Now the thief stuff... Some have presented some very good "ideas" that it is entirely possible that Vas could be charged with "stealing".
It is not out of the realm of possibility... certainly stranger things have happened. How many here last summer thought we would be talking about the things we are discussing now?
>To be honest I didn't really know much about GPL
So why oh why are you trying to muddy the waters by bringing up this issue? What ICGA found Vas in violation of has nothing to do with the GPL.
I do not understand you Majd... what is so hard to understand about breaking a rule that you signed by your own hand you would obey?
This happened to Squarknll at last year's WCCC. Don't remember any fuss here about the travesty of that decision...
>This entire BS about "stealing" or "plagiarism" is not correct.
The "plagiarism" is correct! Based on the "setting" or "venue". Majd it has nothing to do with a civil or criminal court so again... why oh why muddy waters by bringing this up...
This idea of going to Lawyers... why do I need a lawyer to tell me if a rule of a game was broken. I do not have the wits and wherewithal to determine or understand something so simple? Again Majd... c'mon..
You play in a FIDE tournament... ok to use computer assistance? Of course not! It is cheating.
You enter an ICGA tournament stating the chess engine is entirely your own work, when it is not. You could have the most detailed readme. Who cares??? You entered the engine as an "original work". If it is not, that is what is referred to as "cheating". It can also be termed "plagiarism". Try to get a grip on the venue and these terms.
What is an Original Work? ICGA wrestles to "fine tune" the definition but there is also an unmistakable idea or definition of what it is (despite however Jeroen or turbo wish to confuse the issue). ICGA has a pretty clear... not whimsical or arbitrary... definition of what it is. And the idea of "original" has to fit within their clearly stated concept of originality.
Besides, let us not feign stupidity or ignorance here (like the child caught with cookie crumbs on his mouth... uh-oh... be careful Rob... now someone will accuse you of comparing Vas to a child caught snitching cookies)
So not feigning ignorance... anyone who has taken a Computer Science course, and has been tasked with writing a program that does "such and such" knows clearly it has to be their own work. If I was stuck and had a problem writing a certain section (which never happened to me, but for sake of argument) it is possible I suppose I would have asked a friend... "hey... I am stuck with how to do such and such. how to do that?"
The friend could say, "oh just use these such and such commands to write this subroutine" "ah ok thanks" and I go off and write my own section of code.
Or he could say, "here... just copy this and implement it" and if I do I have just plagiarized as now my computer program is no longer an original work by me. I took a short cut and copied. I just cheated and did what the professor asked me not to do, because I did not have whatever it would take within me to do what he asked. So, if he "catches" me... I will certainly deserve whatever decision he makes whether or not he should accept my program as completing the assigned work.
So all CS students know what is acceptable and what isn't... how much more should Vas coming from MIT and having a dad whose teaches CS (maybe his dad does not care about it and/or did not instruct Vas on the importance of originality and what plagiarism was... who knows but I do not suspect this).
>Yet we have Vas as tried and found guilty and equated to a "thief" and a "murderer".
You exaggerate and take it way over the top. Nothing different has happened to Vas that hasn't happened to Squarknll and its author (and previous entries like one program? in WCCC 2009 in Pamplona). Except in this case the "violator" was the reigning world champ and had multiple years of "winnings".
GPL study could be useful... but this is just a simple matter of abiding by the rules Majd. And one's word being accurate and truthful. No need to complicate matters further. "Let your word be your bond." That is fairly simple. And yet too difficult for some...
Read ICGA's rules.
Read ICGA's journal and find articles on what they consider "original works" and "cloning".
Then read the ICGA Wiki on Rybka and determine for yourself whether you think Rybka violated ICGAs tournament rules.
And then when you have done all that, maybe then you will have some worthy questions for Dr. Hyatt or mjlef and you won't be wasting their time. You won't look like an "Ignoramus" because you will have done some prerequisite study of your own, to bring yourself up to speed on a "six-month long" investigation.
Sorry, can't put it any clearer than this.
What am I saying… “cast pearls before swine and they will turn to trample them under their feet, then will turn to tear you in pieces.” Yes… foolish of me to expect ‘men’ here”
“ [W]hen a judge calls an argument "ridiculous" or "frivolous," it is absolutely the worst thing the judge could say. It means that the person arguing the position has absolutely no idea of what he is doing, and has completely wasted everyone's time. It doesn't mean that the case wasn't well argued, or that judge simply decided for the other side, it means that there was no other side. The argument was absolutely, positively, incompetent. The judge is not telling you that you were "wrong." The judge is telling you that you are out of your mind.  ”
In this case I think that there would be no problem in saying it!
Researching something I posted!
> “cast pearls before swine and they will turn to trample them under their feet, then will turn to tear you in pieces.”
Those with eyes to see, see the hypocrisy involved and can appreciate the efforts of those who invested their time and efforts to shed the light on this matter. Sadly, selfishness, perverseness, and wickedness are embraced by those refuting the truth.
> Yeah, that must it. The people who would actually like to see some copied code are selfish, wicked perverts! Are you a follower of Jim Jones?
See some copied code?...of what/which engine?
> Copied code from fruit. And not the UCI parser.
Oh, Fruit code in Rybka?
Unless I'm mistaken, has it not been proven factually that copied codes of Fruit is in Rybka? As to the extent of how extensive the Fruit code is in other Rybka versions, well....
The tests you are referencing are not code, but a claim of very close feature similarity, and I have already debunked these using covariance based techniques.
>and I have already debunked these using covariance based techniques.
this! is the very reason I keep coming back... gems like the above.
> There are well over 1,000 posts on this forum regarding ICGA. Is there something preventing people from posting alleged Fruit code in Rybka on this forum? I have not seen this in the ICGA documentation, just people swearing up and down that it's in there.
> The tests you are referencing are not code, but a claim of very close feature similarity, and I have already debunked these using covariance based techniques.
The obvious (and has been stated and proven many times):
Vas 'went through' the Fruit code (and others) forward and backward and Rybka 1 was created. He stated Rybka was/is 100% original at the source code level. Strelka emerges. Vas claimed Strelka to be a Rybka clone and claimed it as his work. Problem is; Strelka bore similarity more to Fruit than to Rybka. Therefore,
if Strelka = Fruit but Vas says Strelka = Rybka, then Rybka = Fruit.
I know, I know, people are going to say "But Vas recanted about Strelka being Rybka..." . Well, in light of the truth, is this surprising why such recantations were made?
NO! Read the ICGA report that quotes what Vas said ...
"We think using Rajlich’s own definition is good, which he wrote during the Strelka
“Generally, code theft is easy to show - just show the two sections of identical
code, side-by-side. There isn't much to debate in such cases. Rybka is of course
original (with some accepted exceptions like bit scans & bit counters, etc).
Strelka contains Rybka code. Whether Strelka also contains Fruit code, I don't
know and don't really care.”"
He clearly states Strelka contained code other than Rybka's, no mention as to what proportion. He clearly stated that where the other code came from he did not care. Therefore there has been no claim that Strelka was entirely Rybka code only that certain elements of it are.
> Prima wrote "Strelka emerges. Vas claimed Strelka to be a Rybka clone and claimed it as his work. Problem is; Strelka bore similarity more to Fruit than to Rybka."
> NO! Read the ICGA report that quotes what Vas said ...
> "We think using Rajlich’s own definition is good, which he wrote during the Strelka
> cloning postings:
> “Generally, code theft is easy to show - just show the two sections of identical
> code, side-by-side. There isn't much to debate in such cases. Rybka is of course
> original (with some accepted exceptions like bit scans & bit counters, etc).
> Strelka contains Rybka code. Whether Strelka also contains Fruit code, I don't
> know and don't really care.”"
> He clearly states Strelka contained code other than Rybka's, no mention as to what proportion. He clearly stated that where the other code came from he did not care. Therefore there has been no claim that Strelka was entirely Rybka code only that certain elements of it are.
YES! Else how do you explain this:
Topic Strelka 2.0 By Vasik Rajlich Date 2008-01-11 12:26
> ".....The author did at first make attempts to hide the Rybka origins, for example by masking the table values in earlier Strelka versions. He also made significant attempts to improve the program. The attempts at improvement are not very original, but they are everywhere. They include PV collection, null verification (and in fact changes to the null implementation itself), some endgame drawishness heuristics, a handful of new evaluation term, a new approach to blending between opening and endgame eval terms, and so on. They also do include various structural changes, such as knight underpromotions, on-the-fly calculations of many tables, the setting of piece-square table values, etc. These changes are extensive and no doubt lead to differences in playing style and perhaps a useful engine for users to have, but they do not change the illegality of the code base.
> In light of the above, I am claiming Strelka 2.0 as my own and will release it in the next few days under my own name. The name of the author with the pen name "Osipov" will be included if he comes forward with hiw own real name, otherwise an anonymous contribution will be noted. The contributions of Igor Korshunov will also be confirmed and noted if appropriate. All usage permissions will be granted with this release."
From this link:
Feel free to read all discussions in relation to the Strelka-Rybka topic discussed back then. Again I repeat, Vas claimed Strelka as his own. Simple deduction.
>and the claim died a quick death
Maybe with Vas!
But Peter... this is precisely what drew "people's" attention at Talkchess that Rybka might be a "clone" of Fruit... since there was so much discussion about Strelka being suspected as a Fruit "knock-off". Vas shot himself in the foot with that one (really worse than "shot in the foot").
Do a post search for the word "strelka" and as author "Bob"
And quit being so logical and reasonable! People like you make me think there is some hope here and keep me coming back for more.
No need Rob, I well remember the controversy at the time. It may well have been a case of "Oops! what have I just said?" but as I am "logical and reasonable" - nice to have that on record, I gave the benefit of the doubt.
> since there was so much discussion about Strelka being suspected as a Fruit "knock-off"
This meant Rybka's makeup & origin was in "danger" of being traced. As to why he recanted later, well....
I think it was just a simple realization that he would be just as wrong in claiming code he didn't write, as he would be right in claiming code he did write. Hard to win in the middle of such a complicated, and completely avoidable mess...
Later he recanted. Unless he realize he was wrong about Strelka, there also had to be a reason for retracting his statement. (2) Else, why would that be?
If this typical knee jacking is the case, could he have been wrong also in the IppoLit issue as well, since no one's seen the evidence against IppoLit from the accuser?
I think his claim was the initial reaction to finding there was something of Rybka in Strelka and most likely it was that which made Rybka special. However if there was also code from another engine or engines it would not be possible to release Strelka as his own.
As for the claimed "derivative engines" of Rybka, I had my doubts as to their Rybka origins because I keep a games record of gross errors in evaluations. This may be due to wrong information or just missing information and it is interesting to see there were certain similarities in the Crafty - Fruit - Rybka chain including Rybka 3 but when the Robbo engine was released, I could find none of these erroneous evaluations in Robbo. This suggested either the author had put all these issues right or the Robbo engine was not intensely Rybka based.
> These changes are extensive and no doubt lead to differences in playing style and perhaps a useful engine for users to have, but they do not change the illegality of the code base.
Interesting how he came to the conclusion that the code base was illegal, when said code was reconstructed from assembly.
It is obvious that the move choice of Strelka is more similiar to Rybka than to Fruit.
Engines may have the same move choice but are their codes the same? There had to be a reason why Vas retracted his initial "Strelka is a Rybka clone" claim.
> Nonsense. Vas never retracted that statement.
Exactly. I don't recall Vas retracting that statement. But this is/was one of the argument used by the Rybka-is-not-a-Fruit+Crafty-supporters, among other non viable excuses.
By the way in case you missed it, Strelka is shown to be closely related to Fruit than to Rybka. Vas realized this and what it implies.
Strelka is shown to be closely related to Fruit than to Rybka. Wrong. Strelka is almost identical to Rybka, and its similarity to Fruit is only slightly above average. I have test this myself.
Vas realized this and what it implies. Wrong. Vas has been entirely consistent on the Strelka matter.
So are all these tests by various individuals showing consistent similarities between Fruit/Strelka vs. Rybka in code analysis, board structures & values etc. wrong? Are everyone else wrong in concluding Rybka is a Fruit/Crafty clone and Vas is right?
So are all these tests by various individuals showing consistent similarities between Fruit/Strelka vs. Rybka in code analysis, The UCI parser and time management code for Rybka 1.0 Beta looks like it was taken from Fruit. The rest of the stuff is a real stretch that wouldn't hold up outside of an ICGA meeting.
Are everyone else wrong in concluding Rybka is a Fruit/Crafty clone and Vas is right? I'm not sure what a clone is, but Rybka 1.0 Beta introduced a set of new ideas that allowed Rybka to lead the other engines by up to 200 Elo and win the great majority of its tournaments. There is also absolutely no indication that R3 or R4 have any code whatsoever from either Fruit or Crafty.
The ban on all Rybka versions resulted from the ICGA's conclusion that they were mislead right from Rybka 1. The reasoning looks to be, if Vas wasn't honest about Rybka 1 and 2 (which has been determined to be Fruit/Crafty clone) when he entered them in competitions, it's likely that he is also misleading them on Rybka 3, 4 & 4.1 unless he proves otherwise.
Powered by mwForum 2.27.4 © 1999-2012 Markus Wichitill