In the Mark Watkins document http://www.chessvibes.com/plaatjes/rybkaevidence/EVAL_COMP.pdf section knight mobility, Rybka is awarded an 80% plagiarism score for the alleged match between Rybka's method of scoring knight mobility and Fruit's method, both involve counting pseudo-legal moves and applying a weighting, as advised in "chess programming wiki" as the way to do it. What does Vas think about this "plagiarism" score he has been awarded for this feature?
In the Mark Watkins document http://www.chessvibes.com/plaatjes/rybkaevidence/EVAL_COMP.pdf section bishop mobility, Rybka is awarded an 100% plagiarism score for the alleged match between Rybka's method of scoring bishop mobility and Fruit's method, both involve counting pseudo-legal moves and applying a weighting, as advised in "chess programming wiki" as the way to do it. What does Vas think about this "plagiarism" score he has been awarded for this feature?
In the Mark Watkins document http://www.chessvibes.com/plaatjes/rybkaevidence/EVAL_COMP.pdf section Knight in opposing corner or trapped, Rybka is awarded a 100% plagiarism score for the alleged match between Rybka's method of scoring by a direct kludge hit into the PST and Fruit's similar method. In view of the fact that all programs appear to make special kludge hits in various ways into the PSTs, what does Vas think of the fact that this particular matching PST hit has been specifically selected out from a mass of possible non-matching PST hits and uniquely used, increasing the Rybka "plagiarism" score?
Does Vas consider there is a danger that some of the comparators in the Mark Watkins document have been cherry picked to increase the "guilt" score?
That's the problem. You want to continue to make this appear as if just ideas were copied. But the report doesn't discuss ideas... It discusses semantics which is more concrete...
weight(x) = hyperbolic tangent(y) ^ (N+89.56) - InCraftywedoblablabla + kludgefactor - yetmorelies
I don't get that way of thinking. Everyone needs to know that this is not acceptable, and it won't be tolerated. Otherwise we get to repeat this over and over and over... Got to have a deterrent somewhere. If there was no rule enforcement against cheating, how could you even call it cheating any longer. And obviously, EVERYONE would do it under those circumstances..
To lying, as he did say "look at the first pages of the Zach document, you will see five pages of matching source code"
To lying, as he did say "all the PST tables Fruit and Rybka are identical"
To lying, as he did say "all the PST tables Fruit and Rybka are identical with one multiplier"
You don't think those are issues worthy of resolving?
yes, we did resolve them, but you continued lying again and again.
I don't get that way of thinking. Everyone needs to know that this is not acceptable, and it won't be tolerated.
Otherwise we get to repeat this over and over and over...
Yes, you do keep repeating lies .....
If you can't look at the report and find any identical code, I am not surprised. Because until you open your eyes, you won't see a thing. Again, the first event was Rybka 1.6.1 and Crafty. Any comments on that one or should it just be ignored since in this world, lies about origin/originality are perfectly OK. License agreements are made to be broken and nobody should point out any different?
Is THAT a "lie that I repeat over and over" or is it "the truth, the whole truth, and nothing but the truth?"
Can you recognize the difference is a more important issue...
> To lying, as he did say "all versions of Rybka were 100% original, including the early versions..."
> To lying, as he did say "look at the first pages of the Zach document, you will see five pages of matching source code"
> To lying, as he did say "all the PST tables Fruit and Rybka are identical"
> To lying, as he did say "all the PST tables Fruit and Rybka are identical with one multiplier"
> You don't think those are issues worthy of resolving?
> yes, we did resolve them, but you continued lying again and again.
> I don't get that way of thinking. Everyone needs to know that this is not acceptable, and it won't be tolerated.
> Otherwise we get to repeat this over and over and over...
> Yes, you do keep repeating lies .....
I don't get it? I don't understand what it is that you expect of this man? In all of his life he probably never admitted to committing a grievous error. But for some illogical reason, you expect him to have a "come to Jesus meeting" regarding these certain issues ( especially when it's regarding Vas! )
Have you any idea what his fan base would do to him over at Talkchess? They'd shun him. Call him a Nancy! The ICGA would forever turn their backs on him. He'd be an outcast for telling the truth. It would be better he kept up the blinders and continued in the lie. Think about it! He's already given up his God status for regular membership on this forum-that, in and of itself was a pretty good size kick in the ass. Now you want him to have a "come to Jesus meeting"! If he does have that moment of truth, someone better make sure CPR is on site where he's computing.
Honestly, I think you are wasting your time with this fella!
Only saying this since you're one of the few who should know better...
Obviously you are an individual with your own take on the situation, and it is not in 100% alignment with everyone else on the panel. (It would be strange if you had no dissenting or discordant thoughts.) Unlike Bob, I do not believe you are an energumen, and I believe you acted in good faith and sincerely believe in your conclusions. But as you know, a person can come to fair conclusions with the best of intentions and still go astray. Was it absolutely inevitable and necessary for us to come to the polarized impasse we have today? I submit it was not. Somebody missed the forest on account of the trees.
The ICGA panel had plenty of problems in process. Even if they were inconsequential (like having Bob as a member of the secretariat), they shouldn't have left themselves open to criticism. But the actual people debating on the ICGA wiki were certainly not a part of the problem. It was one of the few places where calm, reasoned debate could possibly occur. Your description of them is misguided, and only serves to heighten the rhetoric in this forum to even more ridiculous extremes.
I am not saying the attackers were without a casus belli. But I am saying they chose to give battle. They chose to surround Rajlich Castle and start lobbing fiery projectiles with their catapults. And now you say this was detrimental to the computer chess community? Yes, it was, you noticed! Congratulations for that insight. Now what to do about it? Just call the whole war off, eh? The attacking army can just march away, leaving the city walls battered, with smoke rising up from inside?
OK, I guess that would be better than continuing the battle. Maybe in several years wounds will have healed, but it will be a long time before memories of the onslaught fade.
> If you are saying this whole affair has been greatly detrimental to the computer chess community then you have to ask yourself who had the initiative? Who was the attacker? Who chose to engage in this way?
Certainly not me, as you're implying. Sure, I've said things that were over the top. It's hard not to get emotionally involved, given the situation. But viewing me as an "attacker" feels wrong to me. I certainly don't wish to attack anyone.
I'm trying to say the blame lay with the collective attitudes of everyone (including myself). Why do you feel the need to place it on only one side? The problem is that people are fighting, instead of trying to find agreement (or even peaceful disagreement). Blaming one side for starting the fight only furthers the discordance.
> I am not saying the attackers were without a casus belli. But I am saying they chose to give battle. They chose to surround Rajlich Castle and start lobbing fiery projectiles with their catapults.
Who is "they"? Can't you see that this rhetoric is just as bad as Bob saying "crimes", if not worse?
> And now you say this was detrimental to the computer chess community? Yes, it was, you noticed! Congratulations for that insight.
It's not an insight. I've been saying this publicly since the beginning, probably well before you were even aware of the situation.
How exactly would one go about "calling the war off"? What more could I do towards that end than I have already? If you really believe in calling the war off, do your part: stop using this kind of rhetoric.
I'll start up a thread with some philosophy on "copyright law", if you want to join in, we'll see where it goes .....
You have been quite anti-copyright in many of your statements. Maybe the community could follow your philosophy, and plagiarizing open source programs would become accepted. Then you will have to accept that some people, such as myself, will not be a part of that community. Your call...
Open source, however, is another matter. The extra restrictions that come with open source licences are in severe danger of breaking the spirit of copyright law, in that they do hamper subsequent progress and re-use. Mix that in with icga rule #2, the attempts by open source programmers (Hyatt in particular) to get others to enforce his licence for him (since he won't do it himself) and we're left with a shambles of a situation where nobody knows where the lines are.
My current proposed solution would indeed treat open source as a free-for-all educational resource (which is what open source programmers appear to claim to want), not to be plundered or plagiarised but to be used in all and every way within the law including copyright law. I would relax, and I would expect courts also to relax, the stringency of their criteria on what actually constituted "copying" by the very fact that the source was voluntarily opened. So I'ld restrict direct code copying, but not be so strict with what you are calling "semantic copying".
Can you live with that?
Everyone likes to assume that when an author opens his source, he should not care if that work is stolen. I think this is a very dangerous viewpoint, and one that I do not agree with at all. When I release open source code, it's with an expectation that it is only used within the license. If someone steals my work, they should be held legally accountable. That's the way the law currently works, and that's the way it should work.
I thought I'ld made it plain that the open source was, in my scheme of things, only to be used within the law and within copyright law. "Stolen" and "steals" would be covered by the normal application of the law and you woudl be protected. Your open source would be protected by copyright law, but you would avoid imposing extra conditions (because these react with existing conditions in unexpected/unintended ways) and your protection would be a little less via lessening of the "semantic equivalance" concept.
Where's the problem, then?
The semantic equivalence stuff is more about trying to prove copyright infringement in a case with imperfect information (i.e., we don't have R1's source code). I would have a different view of the case if I believed that Vas had coded all of Rybka from scratch, and only used Fruit as a source of ideas.
> It is important only for what it says about the Rybka 1 case.
My understanding is that your argument on this is that the existence of Rybka 1.6.1 at the time looking VERY different from Rybka 1.0 Beta, and the quick transition to Rybka 1.0 Beta soon after the Fruit 2.1 code release is a large amount of evidence against Vas, based on the amount of time necessary to create such a new engine from scratch.
However, I think that a different possibility isn't unreasonable: Vas developed a different program with a different framework at or near the same time as Rybka 1.6.1, and then shifted the focus of future development (i.e. eliminating similarities with other programs, cleaning up code, changing this and that, etc.) after the disastrous CCT performance. Then, when Fruit came along, he could have seen the PSTs and evaluation terms there, adapted them, changed some things, and gone forward. Of course, I don't know how Vas actually did this, but this seems like a possibility that could explain the "sudden" jump to Rybka 1.0 Beta: the development of multiple paths simultaneously.
There is a lot more than just evaluation terms that are similar between Fruit and Rybka. Low-level stuff like UCI parsing, time management, and the root search are a near-exact match. Vas had a working UCI parser (based on Crafty's xboard parser). Do you think he rewrote it, then rewrote it yet again after Fruit came out? Similarly, Rybka's board data structure, though it's bitboard instead of mailbox, looks fairly similar to Fruit's (I don't have the listing at hand, so I can't say the exact similarities/differences).
Similarly, there's other signs of copying. The different move generators store moves in different data structures. One is very similar to Fruit's sort_t, while the other uses a different layout (they differ between AOS and SOA), overlaid on the same memory. There's different code for sorting moves in the data structures: one matches Fruit's sort, the other does not. This suggests that some old bitboard code was grafted rather hackily onto a Fruit base.
Of course, you have to weigh in Vas' statements that he never rewrote Rybka, and that the Rybka source code predates all of the Fruit releases (the first one was in March 2004)...
> the root search are a near-exact match
Can you by chance refer us to the relevant place or places in the documents showing this?
Something else, I vaguely remember that prior to Fruit 2.1 there also were earlier Fruit sources?
> Well, I am slightly impressed. It's difficult now to "call the war off" after the publication of the icga report. But it would be possible to park the war, temporarily at least, and change the question. We stop arguing about Vas guilt or non-guilt and consider instead the rules of engagement for program developers in the field of computer chess gaming. This would involved considering copyright law, concepts of plagiarism, the effect of open source and open source licences, the interactions between academic and commercials, tournament rules for "clones" and so on. There's no real reason for such a discussion to become adversarial, unless certain people would want to make it so.
> I'll start up a thread with some philosophy on "copyright law", if you want to join in, we'll see where it goes .....
I can't believe I'm reading this shit. Time to get philosophical! I should have seen it coming with Nelson and Zach. "You're a smart guy."
"Yea, well you are a smart guy also". And "peace, brother." "Peace to you also". And nauseating gets a new definition.
You can come get me now!
The fact is that the damage has been done and there's nothing anyone can do to undo it, short of David Levy issuing a public apology and sending it out across the entire chess community, to include all the media sources that published damaging stories, right up to the New York Times. That simply isn't going to happen.
We have to assess reality and find a way forward from here. If you have new evidence to put forward, or a new argument, then go right ahead. But at the present time it seems the tide of antagonism is subsiding and that isn't a bad thing. Sure, Vas has been treated abominably, but he seems to be handling it.
> The fact is that the damage has been done and there's nothing anyone can do to undo it, short of David Levy issuing a public apology and sending it out across the entire chess community, to include all the media sources that published damaging stories, right up to the New York Times. That simply isn't going to happen.
> We have to assess reality and find a way forward from here. If you have new evidence to put forward, or a new argument, then go right ahead. But at the present time it seems the tide of antagonism is subsiding and that isn't a bad thing. Sure, Vas has been treated abominably, but he seems to be handling it.
I see your point Nelson, but it doesn't make sense. To people like Bob and Zach, it's a gift from heaven. For you- there is a bit more calmness. For Zach and Bob, they don't have to defend any more, and are not under the microscope. You certainly do a lot of big favors.
I'm always amazed by the "Levy comment". Who gives a shit what he thinks or does? He heads up a group who make silly rules for games.
Golly gee! And doesn't even bother to be concerned about enacting and explaining any kind of line said gamer must not cross. I'm like Dann Corbit. He said the Vas verdict was not the issue for him. Rather an arbitrary line with so many gray areas he doesn't even know where it is. And from me, no wonder he cannot see it, it stays on the move and where it stops depends on what evidence ICGA is looking at that day.
I have mentioned this before. Not sure if you have seen it or heard of it. One of the most influential in the anti-Vas group (and do not tell me they were "only truthseekers w/ no biases) said and I quote: "A verdict of guilty for Vas is a given, the vote will merely be a formality". Bad enough on it's own, but how does it grab you when I tell you this statement was made BEFORE THE PANEL HAD BEEN GIVEN ANY EVIDENCE! Tell me you don't have any trouble at all digesting that. If it were me they had in the hot seat, I wouldn't recognize them either.
I hope you are not ever on trial for something serious, and the jury hears that you being guilty is a given- before any statements from either side are made. And the trial has not even started, dammit.
The evidence to clear Vas is either there or it isn't. Playing devil's advocate, let's say it is basically proven he is innocent. Even some of the panel admit it. But Levy won't budge. Good. Let the media, right up to the New York Times, take this story and run with it. Is he going to keep quiet- as he accused Vas of doing? Not in this lifetime. He will try to defend a position that can no longer be defended. Ask someone with only a double digit IQ, and he will even know Levy will have to cave. Only a matter of when.
You are actually worried about rhetoric damaging the other side? Well then just excuse me all to hell. "Zach wants peace". I will just bet he does. "Damage has been done", you say. And so there just isn't anything that can be done. How about if the paperwork done by others proves the panel wrong? Still can't do anything? I would gladly play the sacrificial lamb here, since I am probably the only one on all 3 forums who doesn't give a flying rat turd what ANYONE thinks about me on ANY SUBJECT. I don't take polls before I speak. At any rate, I would love to have the honor of taking this new evidence to Levy, balling it up and lending him a hand in chewing on it.
And this "We have to accept reality" bullshit. Where did you read that line and who exactly made that rule? I don't know about the "We have to accept." Best to change that to "I have to accept." For me, I'm under no obligation to accept anything And FYI, evidence re-studied and new evidence has no place on a damn open forum. You will get run around in circles with "bait and switch". Ask Miguel- he got a good taste of it.
I guess the last 3 or 4 weeks turned out to be "all talk and no show". Why should that surprise me- TALK is CW's forte. Wait, I just noticed in your thread, "The tide of antagonism is subsiding". Jesus Dear God- do you ever just speak English?
I almost closed without saying, "Peace, my brother". If I were through trying to help someone, and myself, I would have said, "Let's just forget the silly shit." I don't care whose mouth speaks it- "politically correct" makes me want to puke. Whether in DC or the Rybka forum.
If the ICGA report holds data that can be misinterpreted and if this can be confirmed by a number of programmers then the ICGA acted in haste and should have waited upon the determination of the FSF before acting.
Another question I have is after looking at the list of people who were considered to be on the ICGA panel-I have to wonder how many understood that they were considered "members" of the ICGA Panel. I'm willing to bet that not just a few joined just to see what was going on with the case and not intending to stand in judgment on Vas. This entire affair stinks to high heaven.
Well, I don't see it that way. I see the other side as individuals, not a group. Zach's recent pronouncements are certainly a step in the right direction. He holds his original opinions, but he wants civility to return to the hobby. Can't we adopt a similar attitude? In every war there comes a time for the killing and bludgeoning to stop and for the rebuilding to begin. It is not necessary to end this war with an unconditional surrender by either side. I believe that the passage of time aids the side of truth, because people do tend to come to their senses over time as emotions become less intense.
I still believe Vas is fundamentally innocent and that the decision that was made is a travesty. I still believe a number of people have a lot to answer for. But I also think it is time for the living to bury the dead.
> All right, peace then. No more flaming arrows from me. A lot of blood has been spilled already.
You have to do what you feel you have to do. Same with me. I'm with some others who will stop when the objective is reached. I have no flaming arrows, and have spilled no blood. But I want to do whatever is necessary to make sure a panel never, ever treats a programmer the way Vas was treated. (I'm sure nothing would make them happier than for everyone to just give up. Aint happening in
this lifetime). If fighting for the truth is akin to spilling blood- so be it.
>Finally we have to mention how much harm this whole matter has done to the hobby. They say, "it was Vas who brought us to this point." I say, "no, it was you vindictive wretches who pushed this witch-hunt to these ridiculous extremes." To satisfy some inner craving to destroy the hobby's leading programmer they have fouled and brought discredit to the hobby, and done so cold-bloodedly and with a sense of overweening and dastardly self-satisfaction.
Thank you Nelson for your kind words.
Regarding the above Chris and I have been thinking along these lines the last few days. It's a mess. We first have to acknowledge that to ourselves. To speak with the words of Dr. Phil, you can't change what you don't acknowledge. Apart from the ICGA-Rybka drama I feel the main damage that is done to our little community is the hack of Rybka 3 and releasing its source code. The cat is simply out of the bag. We already had a clone problem and now it's even bigger.
I see a scenario looming that the IGCA will not accept the current strongest chess engines in the world because of rule no.2 ending in a strong devalued tournament and devalued world title as a result. I see hobby tournaments and rating lists wrestling with what is a clone and what is not. Let's first acknowledge computer chess is in deep trouble. How to proceed?
I have a brainstorm idea, therefore don't criticize me too harsh. Let's be practical. These free sources are not going away. Let's be pragmatic and accept them. But........ in order to be recognized and accepted as a valid engine the clone must first proof itself to have contributed substantial playing strength, say 50-100 elo. That latter can be esthablished by forcing any unknown program to get a certain rating at ChessBase, ICC or FICS. Only then the IGCA will accept them, same for rating list tournaments and other hobby tourments.
What this is all about is defining and securing property rights. This is possible with a strong central government guided by the rule of law. It is not possible in a state of near-anarchy like the one we have now. You can only create such a government if there is broad popular support driven by the general belief that government is needed in order to remedy some social problem. How you accomplish that with the fractious community we have I do not know. Ed, face it, lots of people on both sides of the recent argument are highly intelligent yet a little bit off the rails. How do you deal with such a constituency? They live to argue and fight amongst themselves.
>in order to be recognized and accepted as a valid engine the clone must first proof itself to have contributed substantial playing strength, say 50-100 elo
The validity of an engine should also be based on its ELO contribution and not just on its originality. But, I would raise the ELO contribution threshold to >80. ELO contribution is *more important* than the ability to re-code other people's ideas.
Powered by mwForum 2.27.4 © 1999-2012 Markus Wichitill