Not logged inRybka Chess Community Forum
Up Topic Rybka Support & Discussion / Rybka Discussion / In retrospect (locked)
1 2 310 11 Previous Next  
- - By Rebel (****) Date 2011-07-14 09:47
I have signed that letter that triggered the IGCA tribunal to investigate the origins of Rybka. I did that because I believed (and still believe) Rybka started it's life as Fruit 2.1 and as such in that state was not allowed to participate in ICGA tournaments. In retrospect I regret that decision for various reasons:

1. The ban of Chris W. from the panel. If there was one active voice speaking in defence for Vas it was Chris W. Chris subscribed but had to undergo a mortifying and humiliating process of identification. When Chris after having showed a lot of patience finally responded in his usual charming way they complained he was being rude and was not allowed to enter. When Chris and I contacted Robert Hyatt about the stalling process (so Hyatt knew Chris was Chris) Hyatt easily could have informed Mark L. and he did not, instead Hyatt decided to play hide and seek. As a result of this scandalous intentional provocation Rybka has been denied the only programmer voice who would have spoken in his defence. A bad start from the beginning, the first victim of the tribunal, not Rybka but a critical voice.

Hyatt on Chris:

Just for the record, Chris applied, there was a 1-2 day lag as each applicant was screened, we did not want anonymous people, nor people with no technical background. During that lag in admission, Chris did his usual and started his complaining. The group generally decided that they could do without his noise. I did not vote to deny him admission, but then I did not object due to the usual flare-up while he was waiting. And in retrospect, it was a + for the panel, Chris had already made a zillion bogus arguments about why vas "might not have actually copied" fruit code, even though they are identical. Those discussions are still available. Might have been here, am not sure, or at CCC perhaps... It would have been about as productive to include Rolf. Wonder if some of you guys would have criticized us had he applied and been turned down???

Hyatt and the tribunal giving the word "independent" a whole new meaning.

2. The life-time ban is way out of control. Not even in professional cycling with all its doping scandals life-time bans are practiced. When a cyclists is caught part of the punishment is the public humiliation via the media. That is taken into account.

3. The IGCA knew (or should have known) Rybka 3 was clean and contained zero Fruit 2.1 Yet in the mainstream media this fact that the current Rybka is clean is not mentioned. It's also not mentioned that Rybka itself was hacked (Strelka / Ippo) and that Vas's legacy (adding 300-400 elo to Fruit 2.1) is now used by various chess programmers because Rybka's secrets are now in the public domain.

4. Related to (3) one could argue since R3 and R4 are clean why there was a need to strip R3 and R4 from its titles.

Taking this all into consideration I wish I never signed the document. It's not about the initial accusation, that still stands, but the lynch attitude as if Vas is Ted Bundy himself makes me regret.

A personal note to Vas, tell the truth, come clean, people will respect you for it. You are probably raised in the belief that the best damage control is to deny and keep a low profile. That's bullshit, the first is better.
Parent - - By sockmonkey (***) Date 2011-07-14 10:14
If you have proof that Rybka 3 is clean, you should present it. Otherwise, there's no reason to believe that.

Your main problems are, therefore, not with the evidence against Vas, but with the process and the punishment.

jb
Parent - - By Graham Banks (*****) Date 2011-07-14 10:21 Edited 2011-07-14 10:30

> If you have proof that Rybka 3 is clean, you should present it. Otherwise, there's no reason to believe that


Whatever happened to "innocent until proven guilty"?
Having said that, didn't Vas say that there was "code from the public domain" in both Rybka 3 and Rybka 4? Guess it depends on whether you regard that as clean or not.
Parent - - By sockmonkey (***) Date 2011-07-14 10:35
Hahahaha. Are you going to start throwing that around, with your history as a "guilty until proven innocent"-guy? One word: IPPOLIT. You are a hypocrite.

But seriously folks: Vas unquestionably lied and defrauded the ICGA, wrt Rybkas pre-version 3. Why should the ICGA spend even 5 minutes to determine if version 3 is clean? The ball is now in Vas' court to prove that he's fixed the problems that the ICGA have demonstrated. Retroactive disqualification is thoroughly justified until he shows compliance with the rules, now that he's been shown to have disregarded them in previous years.

Finally, what "Vas says" is also provably false. What "Vas says" wrt to public domain code is almost certainly wrong, misleading or simply a lie. Probably there is some public domain code, but since he also refers to the code he stole from other developers in Rybkas 1 and 2 as "public domain", what "Vas says" should be taken with a grain of salt. It's "Vas says" against a mountain of evidence that proves that "Vas says" falsehoods.

jb
Parent - - By Graham Banks (*****) Date 2011-07-14 10:37

> Hahahaha. Are you going to start throwing that around, with your history as a "guilty until proven innocent"-guy? One word: IPPOLIT. You are a hypocrite.


If you're an Ippolit adherent, that says it all really.
Parent - - By sockmonkey (***) Date 2011-07-14 10:42
Is that the best you can do? Attempted character assassination by association? About a point that you already know my views on? You are a manipulable, foolish, sad man, Graham.

As you know, I'm not an "adherent" to anything, except to reason and evidence-based discussions. Truth: there is no hard evidence about IPPOLIT's origins. Truth: there is hard evidence about Rybka's origins. Truth: you are a hypocrite.

jb
Parent - - By Graham Banks (*****) Date 2011-07-14 10:46

> Is that the best you can do? Attempted character assassination by association? About a point that you already know my views on? You are a manipulable, foolish, sad man, Graham.
>
> As you know, I'm not an "adherent" to anything, except to reason and evidence-based discussions. Truth: there is no hard evidence about IPPOLIT's origins. Truth: there is hard evidence about Rybka's origins. Truth: you are a hypocrite.


I could post a few unsavoury comments in response, but I'll refrain from stooping to your level.
Parent - - By sockmonkey (***) Date 2011-07-14 10:49
Fair enough. Do you have any comments on the points I made, or are you going to simply ignore them because they are uncomfortable?
Parent - - By Graham Banks (*****) Date 2011-07-14 10:55

> Fair enough. Do you have any comments on the points I made, or are you going to simply ignore them because they are uncomfortable?


I've chosen not to bother with Ippo and its numerous offshoots, purely based on what I've read and what I've been told.
What others choose to believe or do is their choice.
Parent - - By sockmonkey (***) Date 2011-07-14 10:59
My points had nothing to do with IPPOLIT, apart from the point that you're hypocritical.

Here they are again for your reference:

Vas unquestionably lied and defrauded the ICGA, wrt Rybkas pre-version 3. Why should the ICGA spend even 5 minutes to determine if version 3 is clean? The ball is now in Vas' court to prove that he's fixed the problems that the ICGA have demonstrated. Retroactive disqualification is thoroughly justified until he shows compliance with the rules, now that he's been shown to have disregarded them in previous years.

Finally, what "Vas says" is also provably false. What "Vas says" wrt to public domain code is almost certainly wrong, misleading or simply a lie. Probably there is some public domain code, but since he also refers to the code he stole from other developers in Rybkas 1 and 2 as "public domain", what "Vas says" should be taken with a grain of salt. It's "Vas says" against a mountain of evidence that proves that "Vas says" falsehoods.
Parent - By Graham Banks (*****) Date 2011-07-14 11:03
What the ICGA does and decides is its own business. However, I don't like the way in which things were done as I've stated previously.
Parent - - By Prima (****) Date 2011-07-14 17:10 Edited 2011-07-14 17:13

> If you're an Ippolit adherent, that says it all really.


Rybka has been proven to be of an illegal combination of Fruit+Crafty, and even breaking both GPL rules & ICGA rules, based on the lies and deceit of someone. This is fraud. On the other hand, IppoLit has not been shown to be derived from Rybka nor of questionable origin. After all it's open source besides its unknown authors.

So Graham Banks, with all the "proven" lies, deceit & defrauding of the GPL & ICGA associated with Rybka, what does this say about you - a Rybka adherent?
Parent - - By turbojuice1122 (Gold) Date 2011-07-14 20:08

> Rybka has been proven to be of an illegal combination of Fruit+Crafty


Bullshit.  Show me the lines of Crafty code that exist at all in any released Rybka version.
Parent - - By bob (Gold) Date 2011-07-14 20:45
Go to crafty's source.  Look at the macros in chess.h that deal with AttacksBishop() and AttacksRook().  Then go to data.h to see the declarations for rank_attack_bitboards and such.  Then go to init.c, and look at the several hundred lines of code in InitializeAttackBoards().  Vas has already clearly stated that he copied this code from Crafty.

Need more?
Parent - - By turbojuice1122 (Gold) Date 2011-07-14 22:17
I must have missed Vas's "clear" statement that he copied this particular code from Crafty.  Have you found it in Rybka?  Why isn't this in any of the reports?
Parent - By bob (Gold) Date 2011-07-15 01:48
Because, as I explained, when we were doing the investigation, I pointed out that I had given many people permission to copy the rotated bitboard, just as Pradu gave many permission to use his magic move generation stuff.  In light of that, finding the rotated bitboard code in Rybka 1.0 beta and beyond didn't seem to be that important...
Parent - - By Prima (****) Date 2011-07-14 20:57

> Bullshit.  Show me the lines of Crafty code that exist at all in any released Rybka version.


As Bob has already replied to you:

"Go to crafty's source.  Look at the macros in chess.h that deal with AttacksBishop() and AttacksRook().  Then go to data.h to see the declarations for rank_attack_bitboards and such.  Then go to init.c, and look at the several hundred lines of code in InitializeAttackBoards().  Vas has already clearly stated that he copied this code from Crafty."

Tell us (the world) what you find.
Parent - - By turbojuice1122 (Gold) Date 2011-07-14 22:18
Again, I'm not going to try to find that in all of the Rybka code.  I'm looking for someone to find the offending code and show it explicitly here, posting the comparison of the Crafty code and the offending Rybka code.  If it was really this easy, it would have been done in an earlier report.
Parent - - By Prima (****) Date 2011-07-14 22:24
Ah....but this was done in previous undocumented Fruit vs. Rybka reports. However, you can find these in the ICGA documents.
Parent - - By turbojuice1122 (Gold) Date 2011-07-15 14:32
Where?  I have searched...  Tell me the document and page number.
Parent - - By Prima (****) Date 2011-07-15 16:22
Did you go to Crafty's source and reviewed chess.h, data.h, ini.c etc and you didn't find what you were looking for? I urge you to look into the specified files if you haven't. It makes no sense why the rest of the people see these similarities but you (and other pro Rybka-must-remain-strong-at-all-cost) refuse to see it.

If you can't or don't want to see it, there remains nothing else to be done.
Parent - - By Banned for Life (Gold) Date 2011-07-15 16:40
Question: What would be accomplished by looking for chess.h, data.h, ini.c etc in Crafty's source?

Answer: Absolutely nothing.
Parent - - By bob (Gold) Date 2011-07-15 16:59
You would find the code Vas copied to create Rybka 1.0 beta and beyond, through Rybka 3.  Someone asked "what was copied"...  I answered. 

Of course that won't make a point if you don't want it to.
Parent - - By Banned for Life (Gold) Date 2011-07-15 17:29
There is not much significance in saying that Crafty code can be found in a Crafty source repository. Only you and the brown nose that suggested this course of action will take it as proof that Rybka contains copied code.
Parent - - By bob (Gold) Date 2011-07-15 17:43
Dense, denser, densest.

1.  Do you agree that Vas stated publicly that he copied the rotated bitboard code from Crafty?  I can probably dig up the quote if necessary.  But everyone seems to accept that.

2.  If so, I gave a link to the exact code that was copied.  It is not a "dozen lines" or such.  It is several hundred lines of code, perhaps beyond a thousand.  I could count 'em if it is somehow relevant...

It shows _what_ was copied (and not all, by the way, as there are a ton of support functions for bitboards that also had to be copied.  MakeMove(), UnmakeMove(), Swap(), things that are basically bitboard manipulation routines.  It's not just a few lines, as I said...
Parent - - By Banned for Life (Gold) Date 2011-07-15 17:56
1.  Do you agree that Vas stated publicly that he copied the rotated bitboard code from Crafty?  I can probably dig up the quote if necessary.  But everyone seems to accept that.

a. It is well known that Rybka used this code in the past and no longer uses it today.
b. If Rybka is going to be disqualified for using this code due to non-originality, than all other engines using it should also be disqualified. This is obvious to everybody but you.

2.  If so, I gave a link to the exact code that was copied.

Please provide a link to this code in a recent Rybka release.
Parent - - By bob (Gold) Date 2011-07-15 18:33
A. I believe I answered the "how bad was it to copy crafty bitboard code" early in the process.  That is why it was not included in our report, in fact, even after I had verified that the code was present.  Simply because I had allowed others to use it previously, as had Pradu with his magic stuff.  I do not recall ever giving a commercial author permission to copy that code, but we did not consider it a reasonable argument to differentiate between commercial and non-commercial from this particular perspective...

B.  I gave you the link to Crafty's code.  You get to disassemble any Rybka version up through Rybka 3 and locate it.  You talk a good game.  Now let's see you actually do something technical.  It is there...
Parent - - By Banned for Life (Gold) Date 2011-07-15 19:00
This sounds like a much more fitting job for a loser like you.
Parent - - By bob (Gold) Date 2011-07-15 19:28
When the technical arguments are done, the personal remarks start.  90% of your recent posts have been nothing but personal.  Most can figure out why that is...
Parent - - By Banned for Life (Gold) Date 2011-07-15 19:36
Most people understand that when they throw out insults (and nobody throws out more insults than R. Hyatt), they are liable to get incoming responses in the same vein. It's surprising that you've reached your advanced age without figuring this out.
Parent - - By bob (Gold) Date 2011-07-15 19:58
Right.  I _initiate_ the insults, right?  I suppose that to some, facts do insult them since the facts are contrary to their stated position...
Parent - By RFK (Gold) Date 2011-07-15 20:13
I love your one liners - but you would never make it on a debating team.
Parent - By Banned for Life (Gold) Date 2011-07-15 20:26
You are in denial if you don't realize that you are in the 99.99th percentile when it comes to tossing out insults.
Parent - By RFK (Gold) Date 2011-07-15 20:40 Edited 2011-07-15 23:20
Hey Bob, want to meet Jack Shit - go into your bathroom and look in the mirror and say hello to yourself, say, "Hi" you can even wave and Jack will be waving right back at you! But, unfortunately, guy's like you, as soon as you open that medicine cabinet, and the mirror moves aside, you will forget you ever  met or even really know Jack Shit.  Because, you only know you want to see -[edit] what you want to see!
Parent - - By AWRIST (****) Date 2011-07-15 18:36
Bob, excuse me again. Yes, the quotation stuff is elementary in science, it's also basic for law cases, but why are you so obsessed in digging and finding quotes in different forums? Is it for you always a case of absolute and testified truth? You do completely exclude that someone could cheat you on a certain aspect just to keep the truth to himself until a real court case should come up and then he would say the contrary of course then swearing that this now is the truth? And you would react how? That this was unallowed or immoral? I think that it's the same with Vasik's interviews. You take them as if they would be like testimony in court?
Parent - - By bob (Gold) Date 2011-07-15 18:37
If someone claims I made a false or unsupported statement, I try to offer evidence to show it was not false.

Normal debate tactic.
Parent - - By AWRIST (****) Date 2011-07-15 18:45
Yes, and you are famous for that, but you do see the different attitude and its advantages? I guess so.
Parent - By bob (Gold) Date 2011-07-15 18:49
I don't begin to understand your point.  In a debate, to make a point, you often cite a source.  It is the citations that provide credibility to the original statement...
Parent - - By Prima (****) Date 2011-07-15 18:08
Question: What would be accomplished by looking for chess.h, data.h, ini.c etc in Crafty's source?

Answer: Absolutely nothing.


Absolutely nothing?!!! Well then this speaks volumes. Programmers or those familiar with C/C++ language and coding see what has taken place. They sure haven't jump to the same conclusion of "absolutely nothing".

Your refusal to see this truth does not negate the fact that Fruit/Crafty code were copied by Rybka.
Parent - - By Banned for Life (Gold) Date 2011-07-15 18:17
Only a real idiot could take the presence of Bob's code in Bob's code directory as proof that there is copied code in Rybka. I don't have time for this third grade level argument so you join Watchman as the second person on my ignore list.
Parent - By Prima (****) Date 2011-07-15 20:25 Edited 2011-07-15 20:45
You're in the habit of exhibiting childlike temper tantrums and name-calling when your ignorant arguments and lies are refuted and openly disproved.

You asked for proof, it was given to you and for the world to see. Lots of people here and outside this forum see it except for the likes of you who prefer strong-commercial-engines-at-all-costs despite the proven fact it is unethically & illegally tainted. Or should I do everything for you? If you can't or are determine not to study these codes, it's not my problem. Neither is it my problem if you're ethically and morally challenged to understand 'the wrong' Rybka did and why the ban was necessary.

As for the ignore list, who cares? 'Ignoring list' is the norm here & the easy-way-out when defeated in your baseless arguments. Bye.
Parent - - By turbojuice1122 (Gold) Date 2011-07-15 19:41
If you tell me which lines in both Crafty code and in some reverse-engineered Rybka code to observe, then you will be on the right track.  I'm not going to compare some phantom code unless we know exactly what we are comparing in each program and where it is located in Rybka.  All you're doing is saying, "go see this code--it matches some code somewhere in Rybka".
Parent - - By bob (Gold) Date 2011-07-15 20:02
That cat is already "out of the bag."  Vas has admitted to copying the rotated bitboard code from crafty previously.  I could probably dig up the hex addresses for you to look at in the rybka 1.0 beta binary, if you really want it.  But since he admitted to doing that, and since we elected to not make an issue of it since others used the code, I am not sure of the purpose of doing this...
Parent - - By turbojuice1122 (Gold) Date 2011-07-15 20:04
In that case, why the big deal over all of this?  People talk about Rybka 1.0 "stealing" Crafty code and stuff like that, but if this has been done by many programmers, why the tempest in a teacup?
Parent - - By bob (Gold) Date 2011-07-15 22:14
1.  It is a commercial project, and in every case I can remember, excepting for Rybka, the author specifically asked to use that code.  And to further clarify this, remember the claim that ALL versions of Rybka were 100% original except for "snippets that are public domain."   A thousand lines of copyrighted code does not exactly jive with that statement...

2.  It shows a "pattern of behavior" as is often done in actual legal proceedings. 

3.  In the pre-1.0beta versions, there was _much_ more copied than just the bitboard stuff.  Everything was copied.
Parent - - By turbojuice1122 (Gold) Date 2011-07-18 17:11
Sorry, but over the past few months, I have come to realize that I must take such statements as "everything was copied" with a grain of salt.
Parent - - By bob (Gold) Date 2011-07-18 17:37
Look at the "rybka - crafty data."

To wit:

major chunks of eval code were copied, line by line, with no changes of any kind.  Nobody RE'd the entire eval, there was so much found that looking at everything was simply going to be wasted effort.  iterate.c was copied, exactly, with the "edwards tablebug fix included", code that didn't belong there in my code, and code that could not possibly have been written by anyone else, for a bug that had not existed in 10+ years.  The pieces of the various "NextMove()" family.  NextMove(), NextEvasion(), etc.  Using the same "phase stuff" which I have not seen in _any_ other program.  I thought of the idea as a sort of finite-state-machine approach that provides simplicity and high speed.  You can go thru the crafty/rybka information in the ICGA report attachment, While "everything was copied" was not proven, it is a > 99% probability, because things like option.c, even bench.c (the crafty benchmark, using the same positions crafty uses, the same output statements, the same way of setting up the positions, etc).  And bench.c has _nothing_ to do with playing chess, but it is there.  In living color.  And is an identical twin to what is in crafty 19.x

Sort of hard to overlook that kind of evidence.  Did you read through that?  If not, it will be eye-opening, because it does give code, side-by-side, with no effort at comparing "similarities" but in fact, dealing with comparing for identical programming.
Parent - - By Ugh (*****) Date 2011-07-18 18:09
There you go again with the all-inclusive "copy" word.

In this case "copy" means "take Hyatt at his word and use Crafty code verbatim or otherwise, modify it, play with it, test it and learn about chess programming".

So far so good? Can't exactly disagree, can you, but we can expect the usual obfuscating huffin' and puffin' presumably ....

Then, Vas error or mistake or act or carelessness or oversight -he enters the entity, quite legally based on Crafty, into a  basement tournament where it does not perform particularly well and contrary to the expressed wishes of Hyatt.

Well, you could tell him you were disappointed to discover his "betrayal" of your wishes, intentionally or by oversight - but, instead, you saw to it that the ICGA report contained the language "plagiarised Crafty", which is actually very misleading and defamatory.

Stop with the dissembling.
Parent - - By bob (Gold) Date 2011-07-18 18:23
"plagiarising Crafty" is not misleading at all.  It is a gross understatement for version 1.6.1.  A better description would be "complete copy" and then _possibly_ he made some minor changes.  But not any that were apparent in all the disassembly that was done.

There was never any problem with people taking the crafty source and doing whatever they wanted, so long as they stayed within the very explicit license agreement."  Which does exclude giving a copy of said program to _anybody_ for _any_ purpose.  He violated that.  It also explicitly forbids the use of that modified source in any computer chess tournament, _anywhere_ except one run on your own computer by yourself, privately.  He used it in CCT-6, and then sent it to rating agencies to have it included on their public rating lists.  All without complying with the rule that says any public use has to identify the program as "Crafty."

If you call that huffing and puffing, I'd refer you to the Three Little Pigs, because I certainly blew your house of cards down quite clearly with an extremely explicit license agreement, and then the extremely explicit crafty/rybka data included in the ICGA report.  Why don't you try refuting the data, or my claimed license violations, rather than posting such vague and useless stuff?  This only shows why it was better for the overall process that you were _not_ involved.  It kept out a lot of completely irrelevant arguments that are based on fantasy and/or intentional misdirection.

If you don't like my use of "verbatim copying" how about taking the ICGA crafty/rybka evidence and try to refute any single point made therein, with real discussion, rather than "just isn't true."
Parent - - By Ugh (*****) Date 2011-07-21 08:33
You don't have a licence "agreement", if you did there would be a signature, but there isn't, is there?

You have terms and conditions of use, printed as part of the source code. Did Vas say "read and agreed"? Possibly you would not even be able to prove he read it actually. Less enforcable than a signed contract. But the main problem with your terms and conditions has already been identified by Rolf, when he referred to them as sado-masochistic. The other children can play with your new toy tricycle but they can't change gear or go more than 100 metres and they have to hold up a flag "Bob Hyatt is the Greatest" whilst riding and if they turn left they are in breach and you will never let them play in your gang again. In short, the other children can do practically nothing at all and you reserve the right to bend your "terms and conditions" any which way and pursue transgressors for six years with max-power insulting.

Facts are: 1.6.1 was used under your terms and conditions with very trivial breach, it was not published, it was tested which is all fine. The error was the basement tourney entry. Could have been a simple oversight. After 1.6.1 it appears that Vas dumped Crafty, having learnt from it what he needed, and went on to better things. No plagiarism.

"If you don't like my use of "verbatim copying" how about taking the ICGA crafty/rybka evidence and try to refute any single point made therein, with real discussion, rather than "just isn't true.""

Grow up. Re rybka/crafty, you might as well carry out an exhaustive analysis on, let's say, one of your (for the sake of argument copyright free) posts on Open-Chess that gets cut and pasted onto Talkchess. Of course there's a great load of matches. The parallel is that you ENCOURAGED what you now call plagiarism. Stop with the dissembling.
Up Topic Rybka Support & Discussion / Rybka Discussion / In retrospect (locked)
1 2 310 11 Previous Next  

Powered by mwForum 2.27.4 © 1999-2012 Markus Wichitill