Latest Updates:
Page Index Toggle Pages: [1] 2 3 ... 6
Topic Tools
Very Hot Topic (More than 25 Replies) Rybka disqualified and banned from WCCC (Read 33004 times)
Tullius
Junior Member
**
Offline


I Love ChessPublishing!

Posts: 71
Joined: 05/03/11
Re: Rybka disqualified and banned from WCCC
Reply #81 - 10/04/13 at 12:34:48
Post Tools
Ed Schroeders point of view is already known and so his arguments are not new when you have followed computer forums. A long time ago they have already seen very their emotional discussions.

What Schroeder does not mention is that the programmer of Rybka did not defend himself against the accusation and did in the judicial proceedings an extremly poor job or got terrible advice.

Before the court you have a right to keep your silence but the other side of the coin is that you can have then no chance for a testimony. And when you or you lawyer want challenge any expertise you have to do it in the court and before the judgement is announced and even that was not done.

Moaning after that is wasting time.
  
Back to top
 
IP Logged
 
Vass
God Member
*****
Offline


I Love ChessPublishing!

Posts: 1105
Joined: 06/22/11
Re: Rybka disqualified and banned from WCCC
Reply #80 - 10/04/13 at 06:45:39
Post Tools
GMTonyKosten wrote on 10/04/13 at 00:35:20:
I've just read an interesting article about Rybka entitled "greatest injustice perpetrated in computer chess history" which pretty much speaks for itself.
You can see it here: http://www.chessvibes.com/rybka-case-greatest-injustice-perpetrated-in-computer-...


Thank you, Tony!
Great article, indeed.
As we all know, the truth has always two sides, ...at least.
Nobody would argue that Vasik R. & Robert H. are great in their profession, I suppose. They both started to develop their engines stepping on the hard rocks of  such exceptional open-source engines as Fruit & Robbolito, but then greatly improved them. As long as I know, it's a matter of question about the legal rights, licenses and similar stuff.. Anyway, what they did is already written in chess history!  Wink
  
Back to top
 
IP Logged
 
GMTonyKosten
YaBB Administrator
*****
Offline


Mr Dynamic?

Posts: 3093
Location: Clermont-Ferrand
Joined: 12/19/02
Gender: Male
Re: Rybka disqualified and banned from WCCC
Reply #79 - 10/04/13 at 00:35:20
Post Tools
I've just read an interesting article about Rybka entitled "greatest injustice perpetrated in computer chess history" which pretty much speaks for itself.
You can see it here: http://www.chessvibes.com/rybka-case-greatest-injustice-perpetrated-in-computer-...
  
Back to top
IP Logged
 
Keano
God Member
*****
Offline


Money doesn't talk, it
swears.

Posts: 2891
Location: Toulouse
Joined: 05/25/05
Gender: Male
Re: Rybka disqualified and banned from WCCC
Reply #78 - 02/21/12 at 16:11:10
Post Tools
Chessbase dragging this can of worms up again with a David Levy interview and some disagreement to follow, to be honest I have lost any interest in it as I suspect most chess fans have, but for the specialized group it involves it obviously will fester on.
  
Back to top
 
IP Logged
 
Harvey Williamson
Full Member
***
Offline



Posts: 193
Location: Media City, UK
Joined: 01/04/03
Gender: Male
Re: Rybka disqualified and banned from WCCC
Reply #77 - 01/13/12 at 23:46:23
Post Tools
We are happy to agree on a trusted 3rd party for him to submit his code to but he will not do this.
  
Back to top
WWW  
IP Logged
 
fling
God Member
*****
Offline


I Love ChessPublishing!

Posts: 1580
Joined: 01/21/11
Gender: Male
Re: Rybka disqualified and banned from WCCC
Reply #76 - 01/13/12 at 23:22:34
Post Tools
SWJediknight wrote on 01/10/12 at 16:13:50:
Having read the arguments and counter-arguments in full, I think the case against Rybka is very strong (maybe not 100% certainty, but at least 99%).  The main points from the Watkins report that led me to this stance are:
Quote:
At any rate, one concludes that the Rybka/Fruit data point was an outlier at 6-8 standard deviations in a pool of around 30 comparisons, or more than a 1 in a million chance of occurring at random.

Quote:
The Fruit code also includes some idiosyncratic redundancies, which curiously re-appear in Rybka.

The above looks like pretty damning evidence to me.  I could understand people independently coming up with the same ideas, and drawing inspiration from each others' ideas to help shape their own, but both having the same redundant pieces of code strikes me as highly unlikely to be coincidence.

Given the terms of the GPL a legitimate Fruit-derived engine would have to be free, plus credit Fruit for helping out with some of the code.


Well, as I have understood it, the problem is that nobody has the source code for examination. If Rybka is based on many of the same ideas in the evaluation as Fruit, which seemed to be the revolutionary part, it will maybe also show up together in the tree. It doesn't mean that they are based on the same code, as I understood it, just that they evaluate positions similarly.

It is pretty weird that the same bugs occur, but I have no knowledge of how this could or couldn't happen if the code is not similar (I haven't yet seen what bug it is, for that matter).

This discussion would not be if Rajlich could actually show the code to somebody. But that won't happen, I guess, unfortunately  Angry
  
Back to top
 
IP Logged
 
Harvey Williamson
Full Member
***
Offline



Posts: 193
Location: Media City, UK
Joined: 01/04/03
Gender: Male
Re: Rybka disqualified and banned from WCCC
Reply #75 - 01/13/12 at 23:11:05
Post Tools
GMTonyKosten wrote on 01/10/12 at 10:13:22:
I'm curious to know the human background to this whole affair. Do all the different programmers get on well with each other, or is there a lot of excessive rivalry? Are there different groups who work together, but avoid other groups? Did the fact that Rybka was the undisputed number 1 for so long (and presumably was the most successful commercially?!) create a lot of tension (jealousy?!) with its rivals? Undecided
Perhaps Harvey can shed some light on this as he has been involved with Hiarcs for so long, and must know the environment very well.


Hi Tony,

An interesting question. I have attended several of these events and i must say that I have never had a problem with anyone at an event. The atmosphere is always very friendly and the discussions in the pub afterwards are always very friendly. The posts by the so called 'fanboys' I think would be a lot less aggressive if they had ever attended one of these events. Everybody gets on really well. people probably think I hate Vasik Rajlich after the events of the last 6 months. On a personal level I have never had an issue with him and our email exchanges have always been very friendly.

Best Wishes,
Harvey
  
Back to top
WWW  
IP Logged
 
SWJediknight
God Member
*****
Offline


Alert... opponent out
of book!

Posts: 900
Joined: 03/14/08
Re: Rybka disqualified and banned from WCCC
Reply #74 - 01/10/12 at 16:13:50
Post Tools
Having read the arguments and counter-arguments in full, I think the case against Rybka is very strong (maybe not 100% certainty, but at least 99%).  The main points from the Watkins report that led me to this stance are:
Quote:
At any rate, one concludes that the Rybka/Fruit data point was an outlier at 6-8 standard deviations in a pool of around 30 comparisons, or more than a 1 in a million chance of occurring at random.

Quote:
The Fruit code also includes some idiosyncratic redundancies, which curiously re-appear in Rybka.

The above looks like pretty damning evidence to me.  I could understand people independently coming up with the same ideas, and drawing inspiration from each others' ideas to help shape their own, but both having the same redundant pieces of code strikes me as highly unlikely to be coincidence.

Given the terms of the GPL a legitimate Fruit-derived engine would have to be free, plus credit Fruit for helping out with some of the code.
  
Back to top
 
IP Logged
 
tp2205
Full Member
***
Offline


I Love ChessPublishing!

Posts: 183
Joined: 09/11/11
Re: Rybka disqualified and banned from WCCC
Reply #73 - 01/10/12 at 14:28:33
Post Tools
Smyslov_Fan wrote on 01/10/12 at 08:35:21:
I'm not a computer specialist.


When David Levy claims that the code for the way most of the pieces (except the king) move is identical, how unusual is this? Is the code in all the programs identical for something so basic?  Is this code something special?

This is really an esoteric argument that requires some expertise even to begin to understand whether it's plagiarism.

I do not like the way Chessbase has handled this at all. They didn't report on the case until now, and their stirring defense appears to be very self-serving. If Rybka is the result of plagiarism, Rajlich deserves a ban, and probably a life-time ban. But he also deserves to be able to appeal the ruling.

Neither side has covered itself in glory on this issue. But an explanation of how code is written by other programmers, and how much of the "DNA" is unique to each program would go a long way to persuading me of the justice or injustice of the case. To use the analogy of DNA, the distinction between broccoli and cauliflower are  quite small at the genetic level. Various hybrids are patented. Yet the difference in the genetic code is largely identical to others.

Help us to understand!


I'll try. I don't believe that there are any fundamental ideas in the field which are kept secret. So these can certainly be used freely.

But what almost certainly has happened was cutting and pasting parts of fruit and crafty to simplify the task of writing Rybka (there seem to have been some later attempts to modify the cut and pasted code to hide this). This is
  • forbidden by the licenses under which these programs are published
  • forbidden by the rules of the WCCC
  • just plain nasty as any serious programmer will tell you. You might have spent years trying to figure out how to implement some ideas (and how to implement them well) and then somebody copies them claims them for him/herself.   


The whole idea behind these licenses is that you are willing to give other people a headstart into whatever your program is about because you are interested in progress in your field. The only requirement is that certain uses of your code (distribution, e.g. selling) require that modification or derivatives must be made public as well so that everyone can profit from new ideas. 

Now by showing to a select few programmers (which are sworn to secrecy etc.) the source code of the Rybka versions in question it would have been pretty easy to refute any claim that such cutting-and-pasting took place. In case of ridiculous claims that copying took place you might not feel like doing that, but here there is pretty condemning analysis as well as extremely well respected people on the committee (Ken Thompson). Now the claim that the old source code no longer exists is really hard to belief. Most programmers I know use things like version control (easy to restore any version) and have multiple copies of any code they wrote (especially a successful commercial product) stored all over the place.
  
Back to top
 
IP Logged
 
tp2205
Full Member
***
Offline


I Love ChessPublishing!

Posts: 183
Joined: 09/11/11
Re: Rybka disqualified and banned from WCCC
Reply #72 - 01/10/12 at 13:13:05
Post Tools
RdC wrote on 01/10/12 at 11:37:38:
There may be analogies to the book or article writing process. You are writing about a variation or position covered extensively and correctly by a previous author. To what extent is it legitimate to copy or paraphrase the previous analysis? Are you required to quote the previous source? Would it make a difference if you wee entering a competition for the "best" analysis.


The source code in question was covered by a license which essentially says:
1) if you use it just for yourself do whatever you want with it.
2) if you distribute it or programs based on it these programs *must* be covered by the same license and the modified source code *must* also be made available.

[edit: this is the license fruit used. Crafty's license is non-standard but prohibits taking the program in part or as a whole and entering tournaments (under a new name) without explicit permission from the authors]

Quote:
From a practical user's point of view, you want the strongest chess engine and programmers griping about whether the program is "original" is a theological debate of little practical interest.


This is irrelevant. If I publish a chess book entitled The "Lativian Gambit still lives" which is just a copy of the original with a few corrections (this is just hypothetically speaking, we know of course none are needed) and publish it then the users might like the added value but I suspect the original author would have a problem with that (especially if I would deny that I copied the book but make lots of money from it).

Quote:
The common ground appears to be that the Rybka team read the source code of Fruit and other programs and paraphrased the ideas.


The analysis of the disassembled code and the occurrence of identical bugs/identical illogical behaviour shows quite convincingly that source code must have been copied (as in cut and paste). If the ideas would have been reimplemented then identical bugs/illogical behaviour could not be observed.
  
Back to top
 
IP Logged
 
RdC
God Member
*****
Offline


I Love ChessPublishing!

Posts: 865
Joined: 05/17/08
Re: Rybka disqualified and banned from WCCC
Reply #71 - 01/10/12 at 11:37:38
Post Tools
There may be analogies to the book or article writing process. You are writing about a variation or position covered extensively and correctly by a previous author. To what extent is it legitimate to copy or paraphrase the previous analysis? Are you required to quote the previous source? Would it make a difference if you wee entering a competition for the "best" analysis.

From a practical user's point of view, you want the strongest chess engine and programmers griping about whether the program is "original" is a theological debate of little practical interest.

The common ground appears to be that the Rybka team read the source code of Fruit and other programs and paraphrased the ideas. One of the disputes is whether this should have been disclosed as part of the entry conditions for the computer world championship and whether such a hybrid should have been allowed to enter even if it was, as demonstrated by results, the then strongest chess engine.
  
Back to top
 
IP Logged
 
GMTonyKosten
YaBB Administrator
*****
Offline


Mr Dynamic?

Posts: 3093
Location: Clermont-Ferrand
Joined: 12/19/02
Gender: Male
Re: Rybka disqualified and banned from WCCC
Reply #70 - 01/10/12 at 10:13:22
Post Tools
I'm curious to know the human background to this whole affair. Do all the different programmers get on well with each other, or is there a lot of excessive rivalry? Are there different groups who work together, but avoid other groups? Did the fact that Rybka was the undisputed number 1 for so long (and presumably was the most successful commercially?!) create a lot of tension (jealousy?!) with its rivals? Undecided
Perhaps Harvey can shed some light on this as he has been involved with Hiarcs for so long, and must know the environment very well.
  
Back to top
IP Logged
 
JEH
God Member
*****
Offline


"Football is like Chess,
only without the dice."

Posts: 1456
Location: Reading
Joined: 09/22/05
Gender: Male
Re: Rybka disqualified and banned from WCCC
Reply #69 - 01/10/12 at 09:42:14
Post Tools
Smyslov_Fan wrote on 01/10/12 at 08:35:21:
Help us to understand!


If Bananas share 70% of their generic code with humans, then it is surely them that are behind the Fruit copying conspiracy  Shocked
  

Those who want to go by my perverse footsteps play such pawn structure with fuzzy atypical still strategic orientations

Clowns to the left of me, jokers to the right, stuck in the middlegame with you
Back to top
 
IP Logged
 
Smyslov_Fan
YaBB Moderator
Correspondence fan
*****
Offline


Progress depends on the
unreasonable man. ~GBS

Posts: 6902
Joined: 06/15/05
Re: Rybka disqualified and banned from WCCC
Reply #68 - 01/10/12 at 08:35:21
Post Tools
I'm not a computer specialist.


When David Levy claims that the code for the way most of the pieces (except the king) move is identical, how unusual is this? Is the code in all the programs identical for something so basic?  Is this code something special?

This is really an esoteric argument that requires some expertise even to begin to understand whether it's plagiarism.

I do not like the way Chessbase has handled this at all. They didn't report on the case until now, and their stirring defense appears to be very self-serving. If Rybka is the result of plagiarism, Rajlich deserves a ban, and probably a life-time ban. But he also deserves to be able to appeal the ruling.

Neither side has covered itself in glory on this issue. But an explanation of how code is written by other programmers, and how much of the "DNA" is unique to each program would go a long way to persuading me of the justice or injustice of the case. To use the analogy of DNA, the distinction between broccoli and cauliflower are  quite small at the genetic level. Various hybrids are patented. Yet the difference in the genetic code is largely identical to others.

Help us to understand!
  
Back to top
 
IP Logged
 
Markovich
God Member
*****
Offline



Posts: 6099
Location: Columbus, Ohio
Joined: 09/17/04
Re: Rybka disqualified and banned from WCCC
Reply #67 - 01/10/12 at 02:14:09
Post Tools
Analysis of disassembled code is a quite technical issue. What you wind up with is a series of machine instructions, but without the clarifying variable names, or storage area names, that you would have had the same thing been written in assembly language. And of course, you have no  comments.

But I believe you could tell fairly easily if one set of disassembled code was identical to another. Just like with DNA, there are very few building blocks at the machine level. A sufficiently long identical sequence would be sufficient to rule out an accidental match with a fairly high degree of certainty.

So I wouldn't be so quick to condemn the conclusions of the computer scientists.

The objective strengths of the programs in question isn't relevant. Someone can be proven to be your son on the basis of DNA, and yet not be as strong a player as you are.

I understand the impulse toward scepticism where condemations are concerned, but the evaluation of the evidence in this case really requires experts.

If the verdict is just, then I must say it's ironic that the punishment comes from the computer chess arbiters. The real crime here, supposing guilt, is that the GPL was raped so that some clever thief could make a buck. And no small buck, either. I have no problem with GPL violators being drawn and quartered; I only wish that the Open Source Institute, not some glorified chess club, had been the agent of this particular person's downfall.
  

The Great Oz has spoken!
Back to top
 
IP Logged
 
Page Index Toggle Pages: [1] 2 3 ... 6
Topic Tools
Bookmarks: del.icio.us Digg Facebook Google Google+ Linked in reddit StumbleUpon Twitter Yahoo