Not only Bob is a narcist.... you are having the same problems.
not worth the time to spend a word to.
> You've got to be f******* kidding?! Too bad you can't defend yourself.
> One day it may make sense to you, Vas and everyone else. It's not something I should need to explain.
Then one day it may make sense to you, Bob, and everyone else. It's not something that Vas should need to defend.
You're only making yourself look silly by posting half-thoughts and not explaining or defending them. An intelligent person can explain thoughts in a clear, coherent way, and the more intelligent the person, the more resources he is able to pull in order to do this more eloquently because the more he truly understands the subtleties of the topic and how to get around these. Read any of the papers by the great scientists, and you'll see what I mean--a person not in the field might not understand the jargon and might be missing some background, but the logical connections are still there for all to see, and that is one of the things that makes such scientists great. On the other hand, someone who has a lot of information about something, but doesn't have the intelligence to really understand it, has trouble explaining his thoughts to others--he forms opinions based on this wealth of information, but isn't able to make the necessary logical connections to properly explain or defend his opinions under scrutiny.
Now, while I know that you're not really part of this second group, you're currently making yourself out as though you are. Thus, why don't you try to show us that you're actually part of the first group?
Interviewer: Have you been nervous or tense lately?
Patient: No, I got a head of lettuce.
Interviewer: You got a head of lettuce? I don’t understand.
Patient: Well, it’s just a head of lettuce.
Interviewer: Tell me about lettuce, what do you mean?
Patient: Well…lettuce is a transformation of a dead cougar that suffered a relapse on the lion’s toe. And he swallowed the lion and something happened. The…see, the…Gloria and Tommy, they’re two heads and they’re not whales. But they escaped with herds of vomit, and things like that.
I want to give a really big thanks to Ed Schroder, Soren Riis, Chris Whittington, Nelson Hernandez, Nick Carlin, Jeroen Noomen and Alan Sassler for their superb efforts in defending me against the accusations that I have broken ICGA tournament rules. Soren did a great job detailing the shenanigans pulled during the ICGA's investigation, from stacking the jury to premature public accusations to a comprehensive fabrication of evidence.
I also appreciate the tenacity of Chris Whittington and especially Ed Schroder in digging through the mountains of documents and answering them point by point.
Finally, I greatly appreciate the support that I have received from countless others. This support has been touching and uplifting and I really appreciate it. Thanks guys!
In other news, I'm working on Rybka. Rybka 5 will be ready sometime this year and the remote Rybka renting will be launched sometime after that. We'll make announcements on our web site and in our forum when we have more information. It should be a fun year.
Best wishes to everyone for a great 2012!
He could have and should have defended himself. Dr. Levy practically begged him to do so! He didn't, and had the audacity to snub the whole process. He denigrated the ICGA.
He was in no position to do this. If he were honest and had nothing to (hide) fear he would have behaved like an adult and address the whole matter like an adult.
That is why I took such umbrage with his irresponsible and shameless post. The Computer Chess World against Vas. Ridiculous! He's conned all of you. It's that simple. Played all of you like a violin!
The FSF on the other hand could sue Vasik. Most people who came to the aid of Vasik are actually making it worse. Chris has no credibilty. None. I'll leave it at that. It's a pointless discussion. Injustice...really...that's rich.
Sure they can. But they won't. And if they did, they wouldn't win. And if a miracle occurred, and they sued and won, they wouldn't have an easy time collecting, and couldn't possibly collect enough to pay for the effort. But keep us posted on this. Bob has already promised us that this will be happening any day now. After all, they did get $400K from Cisco, right?
But seriously, the FSF is interested in enforcing their GPL license. Nothing more. When they go after companies, their interest is in having stuff released under GPL. They are not going to be a collection agency for Fabien (as Bob suggested), nor are they interested in going after monetary damages. In fact, as they state on their web page, if they had gone to trial with Cisco, it would have been the first time (and the $400K settlement with Cisco was mentioned in Cisco's financial disclosure statements, but the amount was in the noise). This is not surprising given that the FSF relies on pro bono work from attorneys that wouldn't take kindly to going after groups who are unlikely to have deep pockets (e.g. Vas and possibly Chessbase).
> You show me the lies.
Well, first he says that Rybka 2.3.2a was shown to be "clearly a derivative" of Fruit 2.1. Unless he is using a very different definition in that taking a few ideas and recoding suddenly makes it a derivative, this is a lie. Yes, this is ideas because the "proof" is that on a conceptual level, Rybka 2.3.2a contains many of the same evaluation features as Fruit 2.1, at least when compared with completely out-of-date programs that don't even play a strong game of chess compared with some of the humans on this forum. (Relevant current programs of the time, which almost certainly had far more similarities given their Rybka-style of play, were not used for comparison.) He then strongly implies that Rybka 1.0 Beta's release is copyright infringement, something that I think can be considered to be disproven, given the lawyers who have taken a look at this. (This is notable also because copyright infringement is a much more objective way of measuring "lack of originality" than anything the ICGA decided.) Later, Levy says that Vas copied Fruit's evaluation function--this is a total lie, as noted in many other locations. Vas took some ideas from Fruit's evaluation, recoded them, and used them in a different way in Rybka, something that is perfectly okay and is practiced by most programmers in all fields. He later says that 16 people voted "yes", when in fact only 14 out of 34 voted "yes", and at least 1 has since backtracked, and we also know of a couple of people who didn't know they were considered as part of the panel who would have voted "no" if they had known--this might be a lack of correct memory, but this would seem unlikely if Levy has actually been involved in the investigation.
Anyway, that should be enough...
You would also expect some sort of "order of code" to remain, but that is not there either.
You'ld expect code to at least come from the "same regions", ie be somehow connected together, but it doesn't. This is most glaring in the pawn code which was recently begun to be discussed. The Rybka pawn sections code which Zach claims side by side in the report, and which looks somehow contiguous at first sight, is in fact obtained by extracting bits of code from utterly varied sites in the Rybka. Hundreds of lines of other code separate the functions grouped under "pawn code". Whereas Fruit has them in one block. I rather doubt that one copies code, spreading bits of it all around one's program in other places. That's what happens when you write the code, from ideas, yourself, in your own way.
If I take my Vas is innocent glasses off, and look at the code and reports of "similarities" in an unbiased way (yes, I can do that), then there really is no way one can say "copied". That's the bottom line and more and more people are seeing just that.
As for the Rybka 2 series stuff, it's not about "what I believe", but about what has been established. Very little has been established concerning the Rybka 2 series, and this was the reference of my comment. Part of the problem might be that we're going for multiple points of attack at once: that Rybka 1.0 Beta wasn't really copying of anything that is significant, and that the Rybka 2 series doesn't contain anything more than a few ideas and is quite different from the original beta. These are two completely separate arguments because the Rybka 1 series never participated in the WCCC, and the only thing shown with the Rybka 2 series is similarity in some evaluation features. It is a complete lie for Levy to say that Rybka 2 "coped Fruit's evaluation function". Perhaps we should all just focus on Rybka 2 so that there is a very clear refuter to everything in the ICGA investigation: that Rybka 1 is irrelevant, and that Rybka 2 contains no problems. (I think that Rybka 1 is okay from an objective perspective, but perhaps the ICGA can legitimately say that the ideas taken therein were "too much"--but since this engine is irrelevant, the point is moot.) The similarity in evaluation function was found to be suspect because it was weighed against a bunch of other weak, pre-Fruit programs, when the only real meaning in stating such statistics would be to weigh Rybka-Fruit with comparisons of other strong, post-Fruit programs that are "accepted". This is what Kai has done, with results being that there are a number of "accepted" engines out there that are playing their best games producing moves that are more similar to Fruit or Rybka than what occurs when Fruit and Rybka play their best games and are compared with each other. (By "best games", I mean the naturally tuned parameters, i.e. without attempting to tune them to produce similarities or differences.) Of course, the drivel out of Levy's mouth is, "Rybka copied Fruit's evaluation function".
The Rybka 1 stuff is more debatable, and that has been the bulk of the discussion, but this is not an engine that ever took part in a WCCC, so even though we can argue to and fro about it, the result is meaningless--we should be focusing on Rybka 2 series, and Levy intentionally made several very incorrect statements about this, statements that are contradicted by the facts.
> Anyway, if you admit Rybka 1.x is derivative of Fruit you'd have to deny Rybka 2.x is derivative of Rybka 1 to avoid Rybka 2.x being derivative of Fruit. Good luck with that.
No, this is false. Being a derivative is not transitive.
We have engine A.
We create engine B by taking half of engine A and adding an original second half. Then B is (partly) a derivative of A.
Then we create C by taking the original half of B and adding another original half. Then C is (partly) a derivative of B.
However, C and A have nothing in common, and C is not a derivative of A.
But, you hit upon an important point. The ICGA logic throws around words like "derivative" and "has its origins in", without assessing the actual extent of similarity between the particular engine in question, Rybka 2.3.2 and Fruit 2.1. This is flawed and lazy reasoning on their part.
For instance, they say that "too much" of Rybka 1.0 beta is similar to Fruit 2.1. Then they say that the change in evaluation function between Rybka 1.0 beta and Rybka 2.3.2 is small. But you cannot conclude from this that Rybka 2.3.2 is also "too similar" to Fruit 2.1. It may be that this "small" change is enough to make the similarity no longer "too big". Also, it ignores all the other changes between Rybka 1.0 and 2.3.2 that possibly made 2.3.2 even less similar to Fruit.
Finally, to nail the flawed reasoning of ICGA: We know that Rybka 3 and 4 have practically nothing in common with Fruit 2.1. But they are all derivative of the previous version. Should we conclude that Rybka 4 is a derivative of Fruit 2.1 even though it has nothing in common with it? Then you make things more a matter of genealogy than actual engine comparison.
There will be code to look at. Not side-by-side as it would be too wide. But I am sure that intelligent people can deal with a vertical presentation? Block of original code, followed by block of copied code. Repeated until one gets tired of reading it? Our plan is to make it understandable by anyone. Everyone can then form their own conclusion, including the ICGA.
This never happened. Instead, we found out in July, after the investigation was already over, Bob still thought that the PST code in Zach's document under the 'Rybka' label was actually Rybka code. Of course it wasn't. It was actually Zach saying that the Fruit like code could have generated the Rybka PSTs. Bob's comment was:
This is a PDF link: http://icga.wikispaces.com/file/view/ZW_Rybka_Fruit.pdf. Page one is introduction. Page 2 has piece square tables _and_ some side-by-side code. The next several pages _all_ have "side-by-side code analysis." Did you just read the "introduction" and quit? that would explain a lot...
I didn't know it was _that_ hard to find it...
So somehow or other, the big chunks of copied code in released versions of Rybka just melted away, until we reached a point where no copied code at all could be shown. Of course this wasn't easy to explain, but this didn't stop Bob from posting 5,582 times in 11 months, endlessly repeating the mantra that although nobody could show any examples of copied code, examples were everywhere.
Of course, in order for him to violate ICGA rules, one would have to show that Rybka bitboard code translates into Fruit mailbox code for Rybka 2.0 and Rybka 2.3.2a (or Rybka 3), something which nobody has been even close to doing.
initialise the bishop mobility score to zero
mobility = 0;
add in the legal move count, captures and non-captures are hard coded to be treated the same way
mobility = mobility + [sum of pseudo-legal moves];
multiply by a Rybka specific weighting factor
mobility = mobility * [weighting factor specific to Rybka];
add mobility to the evaluation score
(the value of mobility here will be quite specific to Rybka and different to Fruit)
eval = eval + mobility;
initialise the bishop mobility score to some negative value, this effectively reduces the bishop value compared to the knight, it's a chessic difference
mobility = -BishopUnit;
add in the legal move count, captures and non-captures are hard coded to be treated a different way to each other
mobility = mobility + [sum of pseudo-legal moves];
multiply by a different weighting factor
mobility = mobility * [weighting factor specific to Fruit];
add mobility to the evaluation score
(the value of mobility here will be quite specific to Fruit and includes an offset value which reduces the value of a Fruit bishop compared to its other pieces)
eval = eval + mobility;
Basically, the only similarity is that both programs use the concept of mobility, using the definition of mobility given in the chess wiki but treated and scored in a different way, and add this mobility to the evaluation score. As do countless number of other programs.
There is no code equivalence, everything is done differently.
There is absolutely no EXACT same code as Hyatt nonsensically claims.
There is no literal copying.
There is no non-literal copying.
And, by the time we abstracted (by force) to this level, our degree of freedom of expression is close to one, and we are discussing ideas and not code.
Quite what "mobility" is doing in any listing of "side by side" code, is quite beyond me, and should be quite beyond anybody able to read. There is no comparison.
I recommend treating this example, the mobility code, as a template for all the code throughout the evaluation. It shows clearly how the comparison process has moved/abstracted into the realm of ideas and how comparison on the base of "code" has failed to show up any matches.
Powered by mwForum 2.27.4 © 1999-2012 Markus Wichitill