Rybka Fruit semantic equivalence of two code sections: throw out all differences until you get only the same bits left. Use this to claim Rybka copied Fruit.
Crafty Fruit semantic equivalence of two code sections: include and emphasise all differences. Use this to claim Crafty did not copy Fruit.
Dual standards. Or cheat as it is better known.
There is no dual standard here. The code Ed pointed out was in Crafty in 2003, LONG before Fruit 2.1 was available. The 2003 code had 4 lines added in 22.2, namely, each line that looked like:
score -= bishop_trapped;
turned into two lines that looked like:
score_mg -= bishop_trapped;
score_eg -= bishop_trapped;
Everything else is IDENTICAL. And pre-dated fruit by 2+ years.
Next, Ed claimed my "trapped bishop scoring" was semantically equal to Fruit's. Sort of hard to make that case when Fruit has MUCH more code for trapped bishops.
In 2003 Crafty considered a bishop trapped if it stood on a7 with an enemy pawn on b6, or if the bishop stood on b8 with an enemy pawn on c7. This was repeated for the other corner, h7/g8. It "reused" that code by flipping the side inside the eval, as opposed to enumerating all 4 sets of corner squares as 4 separate blocks of code.
In 2005, Fruit's eval.cpp shows code that tests a7/b8, AND a6, where the new addition of a6 (compared to crafty) looks for an enemy pawn at b5. This, too, is enumerated for all 4 corner squares.
Only a complete IDIOT would claim that those two DIFFERENT approaches to a trapped bishop evaluation were semantically equivalent. Simple test. place a bishop at a7 with pawn at b6 and compare the evals. Same. Place a bishop at a6, pawn at b5 and repeat. DIFFERENT. NOT semantically equivalent.
Next significant dishonesty. Crafty 19.x existed PRIOR to fruit 2.1. Ergo, if that code (or idea, since the "idea" in fruit is not the same as in crafty) WAS copied, I wonder who had to copy whom?
Third bit of dishonesty. He takes THAT 2003 program, makes the above completely incorrect statements, and then takes a version from 6 years LATER, a version which came out 4 years AFTER fruit, and says "here, the bishop PST is an exact match." It is not exactly an "exact match" in the first place. Can't change the constants in Fruit's PST code to produce my numbers, as you can change them to produce Rybka's numbers. But that's a different point. This "match" is from a program that was released 6 years later than the first. But he implies that this is "all done in Crafty". That way of evaluating trapped bishops has not been used in Crafty since 2008 or 2009. That has changed. That PST was not used in Crafty PRIOR to 2009 or so.
As I said, TOTALLY dishonest. And there are others. In YOUR case, we did not "throw out all differences" in comparing Fruit to Rybka. ALL that was abstracted away was the board representation. Nothing more. Not, according to your hyperbole, "throw out all differences."
Ed and you have absolutely no clue about "semantic equivalence" and what it means. Some of us do. It does not mean "throw out half the code and then claim you have a perfect match with semantic equivalence." That's a lie. And he is becoming quite famous for same. And your support of his dishonesty is letting that impression rub off on you as well.
Carry on. BTW, since you and Ed are obviously "date challenged", please feel free how to explain the fact that the crafty code testing a7/b8 + the other black corner, which was in 2003 released versions of Crafty (BTW the SAME version of Crafty that Vas copied to make 1.6.1 IF you notice, the version of Rybka that existed way before Fruit was released) was somehow copied from a chess program that was not released for two years later? This ought to make for "interesting reading."
No standards at all. Or "lies" as it is better known...
If you were to tell me to stop disagreeing with you, ok, but disagreement with Bob Hyatt != lying.
No projecting, just a simple statement. I am in the midst of pointing out lie after lie by Ed. To the point that I don't have much time to worry with your nonsense at the moment, but when the time is right, or you actually choose to try to make factual statements again rather than just spouting gibberish that can not even be discussed...
First you believe you "have been brought up properly" and by definition that allegedly means that your natural response is "right" and anyone who disagrees is "wrong" or "lying". Unfortunately, the "properly" in the above refers to a Southern USA Alabama upbringing, with history of, well we won't go into that ....
Secondly, you believe you have no need for any externally defined Code of Ethics, not just in this case but in general, because, you have "the Bible", and, having read that, of course, au naturelle, your responses are going to be "right" again etc. as any crusader know.
Thirdly, you believe your ridiculous world view is to be imposed on an international level, when most of the rest of the world regards that with about as much horror as any other fundamentalist world view from any other Ayatolla.
Go back to Alabama, Hyatt and stay there, where perhaps you can do no harm. But get the hell out of computer chess, and take your death fantasies, suicidal death wishes and destructive tendencies with you.
Sort of funny, when you think about it, because I have done ALL of those, and still am.
And I find it funny that helping to expose one of the most significant examples of cheating in CC history is a bad thing. I guess the folks that exposed Madoff were also "rejects from humanity" since they caused him to exit the financial world, enter prison, and broke him financially? You have an extremely twisted view of reality and ethics. But don't let that stop you. Most of "US" are interested in this issue because WE are still competing. YOU haven't done ANYTHING in years, except whine and complain. I wonder who should be considered more seriously?
Yes, we can see that. Only downhill if anywhere. That's why at your age, or my age for that matter, it was better to exit on a high. Just giving you the benefit of my experience. Best if you quit now, too late to salvage your broken reputation, but better late than never.
And you've not exposed any "cheating" by Vas. The jury is still out, but reasonable people (that's not you of course) can see that there's no proof one way or the other, difficult case, shades of grey, and, coupled with your catastrophic process, leaves only one verdict. Not guilty.
> The ONLY debate point I have seen raised was the severity of the punishment.
This is a lie.
> The "jury is not still out" on the evidence.
This is a lie, since it implies that an impartial jury came to a decision. But there was never an impartial jury vote in the ICGA.
> Those that have looked and studied it have been convinced.
This is a lie.
So far, 3 sentences, 3 lies.
> The rest? If they are interested, they can read for themselves rather than letting Ed and you provide them with distorted/incorrect stuff left and right.
Incoherent logic. You ask people to make up their own mind, but then you suggest they do not read certain things so as not to be "provided" with wrong stuff (hinting that people are, in fact, incapable of making up their own mind).
Equivalent to a prosecutor asking the jury not to listen to anything the defense has to say, but instead make up their mind after only listening to the prosecutor. You would be laughed out of court!
The "thing" to read is the ICGA report and the included evidence. THAT is the material that shows that rule two was broken many times, and that code was copied in violation of the license agreements of two different programs... If you want to question a jury's verdict, look at the "trial transcript." Not at the newspapers and such. 9 out of 10 people here that criticize the ICGA have OBVIOUSLY not read the ICGA report. Because they post statements that are directly contradicted by simply reading the report.
In this "trial" the defense did not SAY anything. Don't forget that "small detail."
Makes absolutely no sense.
Has 'being of help' been an issue for Levy anytime in the past? No. I have been at WCCC's several times and I have seen with my very own eyes how badly run this organisation is and what a bad job they are doing in the world of computer chess. No live coverage, no PR, nothing.
The next scandal they produced was their tolerating that Zwanzger, BTW a player of master strength, was allowed to throw a _factually_ drawn game to Shredder, so that this program got into a final tie and actually won against Fritz. All that looked very much like corruption. Never ever I would mention a financial connect between Levy and the Amir Ban creation Junior. I dont want to be painted as, well, kind of absurd, since David lives in Britain as I'm been informed, so that a direct contact can be faithfully excluded.
So there are many examples of the ICGA's behaviour, which (to put it mildly) do not confirm Hyatt's 'high level of fairness and ethics' in this organisation.
Bob, you guys did not have the money or did not want to spend the money to have a panel made up of completely independent and neutral experts writing up unbiased report(s). What you did was good enough for ICGA, but it doesn't nearly meet the general PUBLIC's expectation of fairness. So now instead of looking like the good guys, all of you, especially you and Levy look like a bunch of a.holes and liars. Well done sir.
btw to answer your post of yesterday. The FSF will not make a ruling that is not their purpose. They have already said they will support Fabien. However I think the FSF is quite a small organisation with a backlog of cases. So if it ever goes to court I doubt it will be soon.
- Chris and Ed are very impressive guys, and don't have a vested interest in the outcome. They would be devastating as expert witnesses.
- Congratulations for being the only one in your camp to realize that the FSF never makes findings or rulings. Maybe you also know they have made recoveries in only a few cases against large companies with very deep pockets. More importantly, they are not in the business of recovering money for people using the GPL license, only for their own organization. They are in the business of getting violators of GPL to provide open source. Fabien therefor has zero chance of getting money from any FSF action, however unlikely.
- The FSF's budget is about a million dollars a year. That is a pittance in the legal business. The FSF concentrates on going after companies with deep pockets like Cisco. They certainly won't bother with Vas or even Chessbase. The same economics work against Fabien, and he doesn't even have people doing pro bono work for him.
- Likewise, it's very unlikely that Vas will be taking any action for the libelous press package that was distributed, because neither the ICGA, nor its principals, have deep pockets.
btw Ken has the coolest email address i have ever seen - can't paste it here but it is fab!
I make no comment on what the FSF may or may not be doing as that would be information I can not share.
In any event, despite what you may have been led to believe, obtaining a finding of software copyright infringement isn't trivial in the US (and I suspect it's no easier in Germany or Poland). At the end of the day, you are generally trying to convince non-experts of very technical details, and it is much easier to play defense in these matters.
If I were Vas, I certainly wouldn't be losing any sleep over what the FSF will do. Maybe they will send him some letters. He's already proven to be adept at ignoring this type of distraction. And that will be that, unless they decide to go after bigger fish.
The fact that you and Bob failed to put together an unbiased panel is not my fault so you should not be so hostile towards me. I am not an expert in anything, nothing at all, but I feel flattered that you and Bob keep calling me an expert. My personal opinion though (please note, not expert opinion, just my personal opinion) is that you and Bob are two dishonest a.holes. Good day my friend.
It is not just you that makes this mistake it was the same in the CB article:
> Riis: It really goes without saying that the panel members voted based on
> the findings of the ICGA report...
But the ICGA Secretariat Report was written *after* the panel members voted!
> He posted on the wiki but of course you have not read that.
Ken's contribution on the Wiki:
1) By doing too much sports, Ken, could you have been the creator of the tsunami that then caused the Fukushima reactor to explode?
2) Did you then get the Japan (Nobel) Prize for your magic achievement?
In its shortness (brevity of statements) lies the secret of our wisest men.
Like in the Bible already:
Anyways, what's done is done, let's move on.
> If you have a cycling race, it's preferable not to have one team in charge of conducting a doping test of another team. I am stating the obvious.
It is just staggering that they can't see this, and furthermore that they think there is nothing irregular about the process they used.
Of course they know it, but if you are 2 pieces down in a hopeless position, you might as well sacrifice another two to create some confusion :-).
Any court will crush this case without hesitation, no doubt about it.
Edit: I have suffered financial loss by paying to attend many events and my huge electricity bill :) Not sure I can sue anyone for the money though :(
It's easy enough to make excuses for poor performance. But six years later, Vas still has the strongest engine and a lot of new strong engines have been taking advantage of his work, just as many engines gained strength after the release of Fruit. Guys like Larry are happy enough to reveal that his team has tested out ideas from all of the top engines, keeping what works for their engine. Teams that can't work in this environment are doomed to extinction.
Exactly. So far Uniacke has shown that he doesn't know how to write cluster software (solution: have the ICGA ban cluster software) and to incorporate the new knowledge (solution: ban those income spoilers). When I study Hiarcs' main lines, scores and search behaviour, I can clearly see that Uniacke is trying Robbo/Strelka/Rybka stuff, but also has no clue how to make it work. Therefore the top engines still beat Hiarcs easily by 75-80%.
It must be extremely frustrating for HW to see Uniacke's dominating position in the pre-Rybka times disappear like snow in the sun, but alas, it is the harsh truth.
> Other programmers have families to feed as well. I wonder how much potential income Vas stole from them by cheating?
Here you show true colors; it's not about code copying but frustration. Everybody understands that Vas didn't become the clear #1 for years by copying Fruit. If Rybka would have had the strength of Fruit for years it would have been an insignificant engine and this discussion would never have been raised. All the talk about taking ideas vs copying is just BS. Look at what Critter and Komodo do. They try and test every idea from other engines and if it works they keep it. Would Critter and Komodo have risen to the strength they have today so fast without 'taking many things' from Fruit, Rybka etc? Ofcourse not! Yet they are considered perfectly legit.
> Everybody understands
Powered by mwForum 2.27.4 © 1999-2012 Markus Wichitill