David Levy asked for opinions

Discussion of anything and everything relating to chess playing software and machines.

Moderators: bob, hgm, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Frank Quisinsky
Posts: 4852
Joined: Wed Nov 18, 2009 6:16 pm
Location: Trier, Germany
Contact:

Re: David Levy asked for opinions

Post by Frank Quisinsky » Thu Feb 24, 2011 4:11 am

Hi Bob,

I think it need only time for the organisation of such a site. Because we have more as 500 programmers. What can be do is to wrote a mail to the group of programmers.

Example:
Christian Barreteau, Rudolf Huber are MTD(f) experts. So we asked this programmers for basic sources to MTD(f). Miguel is the creator / expert for GaviotaBases, so here we can asked Miguel.

Or ...
Mark Uniacke like "techinique x". So he can write to technique x. Each programmer can write a bit with basic sources to a topic he like.

With a good concept (should a professional programmer create) I think such a page is possible and great for each programming beginners. Noboday must make copy and paste or cloned programs.

Sorry Bob for my idea. I know you have around 100.000x or more ... knowlege to it and it's overconfident that I as non programmer discuss with you about it ... but we all must different things do after the situation we have with clones. We can't await that ICGA will be active because ... honest ... we don't have an official organisation for computer chess which are able to organice a bit more as a simply WCCC.

Best
Frank
I like computer chess!

Tom Barrister
Posts: 227
Joined: Tue Oct 05, 2010 3:29 pm

Re: David Levy asked for opinions

Post by Tom Barrister » Thu Feb 24, 2011 5:10 am

geots wrote: if Rybka has to stand up to certain expectations, then why would that not mean Shredder, Hiarcs, Fritz and all the rest would have to do the same. There may be nothing there they have done wrong- but no one has even dared question them. If its good for the goose, it's good for the gander. Reading I guess between the lines, I see you saying that this whole thing could easily snowball and turn into one heel of a mess that no one is even expecting.
George
All I'm doing is facing reality, but that's the gist of it: it's a mess.

There are so many hurdles here that I don't see a clean solution:

First, you need to get the private and/or commercial engine authors to even agree to reveal their source.

Where are we going to find a tribunal of people who the closed-source engine authors trust to reveal their source to, who are qualified to make a judgement, and who are to be willing to put in the effort?

Then standards have to be established as to what constitutes a violation and what is acceptable. I doubt it's completely clear. What constitutes unacceptable use?

Complicating that is the fact that the engines change constantly. That means that future engine entries will also need to be analyzed to ensure that nobody slipped anything in that shouldn't be there.

Even asking hints that there's widespread cheating going on. That may well be (or not be) true, but it gives the community a bad name.

Then what happens if some authors refuse? As an example, I wouldn't blame the people at Shredder for refusing to reveal their source code, even if they were completely innocent. By refusing, it could be interpreted that they have something to hide.

All I see this tribunal idea accomplishing is to divide the community, alienate some, possibly cause the commercial/private engine people to create their own tournaments and/or organizations, cause more collusion therein, etc.

Then there's the practicality of even bothering. Commercial chess engines, for the sole purpose of strength, is a thing of the past, in my opinion. With Critter, Houdini, Stockfish, Ivanhoe, Fire, etc. all available for free, it's going to be hard to convince anybody to buy Rybka, Naum, Zappa, Hiarcs, etc., simply for their playing strength, no matter how many world championships they've won. Successful authors/companies are going to either bundle their engines with other software, or will need to add other features, such as blunder-checking, variable strength, coaching, etc. The horse left the barn long ago. It's more a matter of the programming community competing with one another. The public may watch with interest, but when it comes down to shelling out their hard-earned cash, it will be hard to persuade them to buy Rybka when Houdini is a download away.

In other words, the tribunal is going to be there mainly to satisfy the programming community. Very little commercial impact will occur, in my opinion. The end result will probably be that the public has less trust in the community as a whole. That also won't be good for "business."

Note that this is independent of the situation with the Ippolit/Robbolito series, or with Rybka and Fruit. That's another matter. If Mr. Letouzey and/or the Free Software Foundation wish to pursue the matter in court, and if they do prove that Rybka 1.0 started out as essentially a clone/derivative of Fruit and in violation of the GPL, then the tournaments will have proof and can take whatever steps they wish. The same holds true for the Ippolit series and Robbolito. But it's for each tournament group and/or organization to decide individually. Trying to standardize it and force it down the throats of every tournament and/or organization is just going to lead to more alienation, distrust, arguing, fighting, name-calling, finger-pointing, etc.

That's my view for what it's worth.
This production is being brought to you by Rybka: "The engine made from scratch.™"

Steve B
Posts: 3697
Joined: Tue Jul 31, 2007 2:26 pm
Contact:

Re: David Levy asked for opinions

Post by Steve B » Thu Feb 24, 2011 11:46 am

geots wrote:
I have wondered before that if Rybka has to stand up to certain expectations, then why would that not mean Shredder, Hiarcs, Fritz and all the rest would have to do the same.
Sorry George but i disagree
if someone is accused of something due to a plethora of evidence and then taken to court ..it does not mean that everybody else in the room also needs to stand trial where no such evidence exists

none of the engines you mentioned have the cloud over their head that RYBKA did from day one
from a sixth place finisher in one tournament to a
huge increase in elo literally overnight ..going on to fame and glory becoming a 4 time WCCC champion
which engines that you mentioned have another highly respected engine author questioning their authenticity and similarity to their engine?
any other engines with the "Zach" analysis?

Lets remember .. Fabien Loutezy surfaced 5 years later asking questions about Rybka that showed he was concerned about the similarities to his engine..establishing that he was not OK with Rybka..
the false assumption that Fabien was OK with Rybka all of these years kept it above any real scrutiny
lets remember..this discussion would have taken place 5 years ago had Fabien surfaced then and not now and not decided to give up on computer chess
the situation has become muddled in the intervening years allowing the engine author to slowly make changes to the engine and gain a foothold in the community with a high level of acceptance by virtue of winning titles in major events specifically the ICGA world championships..

but now the Arc of justice ..slowly has come to shore

the problem with public forums like this one is there is so much chatter day after day ..talking in circles ..back and forth ..day after day ..that whatever happened just 3-4 weeks ago is forgotten because its 5 pages down on the view list
Fabien Surfaced ..late.. its true..but surface he did
and now a close examination of Rybka will be made by the governing body of the ICGA World Title.. comprised of an expert panel of its peers

For the integrity of our hobby.. i for one ..wish them
God Speed Regards

Steve

benstoker
Posts: 342
Joined: Tue Jan 19, 2010 1:05 am

Re: David Levy asked for opinions

Post by benstoker » Thu Feb 24, 2011 3:31 pm

Steve B wrote:
geots wrote:
I have wondered before that if Rybka has to stand up to certain expectations, then why would that not mean Shredder, Hiarcs, Fritz and all the rest would have to do the same.
Sorry George but i disagree
if someone is accused of something due to a plethora of evidence and then taken to court ..it does not mean that everybody else in the room also needs to stand trial where no such evidence exists

none of the engines you mentioned have the cloud over their head that RYBKA did from day one
from a sixth place finisher in one tournament to a
huge increase in elo literally overnight ..going on to fame and glory becoming a 4 time WCCC champion
which engines that you mentioned have another highly respected engine author questioning their authenticity and similarity to their engine?
any other engines with the "Zach" analysis?

Lets remember .. Fabien Loutezy surfaced 5 years later asking questions about Rybka that showed he was concerned about the similarities to his engine..establishing that he was not OK with Rybka..
the false assumption that Fabien was OK with Rybka all of these years kept it above any real scrutiny
lets remember..this discussion would have taken place 5 years ago had Fabien surfaced then and not now and not decided to give up on computer chess
the situation has become muddled in the intervening years allowing the engine author to slowly make changes to the engine and gain a foothold in the community with a high level of acceptance by virtue of winning titles in major events specifically the ICGA world championships..

but now the Arc of justice ..slowly has come to shore

the problem with public forums like this one is there is so much chatter day after day ..talking in circles ..back and forth ..day after day ..that whatever happened just 3-4 weeks ago is forgotten because its 5 pages down on the view list
Fabien Surfaced ..late.. its true..but surface he did
and now a close examination of Rybka will be made by the governing body of the ICGA World Title.. comprised of an expert panel of its peers

For the integrity of our hobby.. i for one ..wish them
God Speed Regards

Steve
The more I think about it, the more I believe Hyatt and Levy's ICGA tribunal can be a success. It's the "Build it and They Will Come" principle. If a respected team of experts conducts a serious, scientific examination of a program with a well-defined set of criteria, there is really no way to hide from that.

Assume that the ICGA panel completes an examination of prestigious Commercial-Closed-Source-Program-X and determines it to be an "impermissible derivative." Assume that determination withstands peer review. Would the sponsor of this forum (ChessUSA) continue to sell an "impermissible derivative"? Chessbase? Would GMs or FIDE associate or co-sponsor such a program?

One way or the other, it would be a serious problem for the producer of that "impermissible derivative"; that producer (not "author") would be a pariah and shoved off to the computer chess netherworld such as sites like Immortal Chess. Exposed as a fraud, the days of wine and roses would be over. No more parties, fancy crackers and Brie. No more walking into tournaments and experiencing the adoration of chess enthusiasts goo-goo eyeing your very presence.

Moreover, the copyright holder of the original program would have in his possession an invaluable expert report to make use of should he wish to take the matter of copyright infringement to other tribunals. The cost of experts is a fantastic impediment to seeking redress in the courts of law.

If chess itself and computer chess are in the last throes of waning cultural importance and popularity, then, better to go out with a bang and not a whimper.

Steve B
Posts: 3697
Joined: Tue Jul 31, 2007 2:26 pm
Contact:

Re: David Levy asked for opinions

Post by Steve B » Thu Feb 24, 2011 3:50 pm

benstoker wrote:
The more I think about it, the more I believe Hyatt and Levy's ICGA tribunal can be a success. It's the "Build it and They Will Come" principle. If a respected team of experts conducts a serious, scientific examination of a program with a well-defined set of criteria, there is really no way to hide from that.

.
Agreed
the panel now seems to have been enriched even further with the addition of the well known "BB" who is clearly held in high regard as an expert in the field

The Arc of Justice Moves Slowly Foward Regards
Steve

User avatar
Rolf
Posts: 6081
Joined: Fri Mar 10, 2006 10:14 pm
Location: Munster, Nuremberg, Princeton

Re: David Levy asked for opinions

Post by Rolf » Thu Feb 24, 2011 4:35 pm

Steve B wrote: Fabien Surfaced ..late.. its true..but surface he did
and now a close examination of Rybka will be made by the governing body of the ICGA World Title.. comprised of an expert panel of its peers

For the integrity of our hobby.. i for one ..wish them
God Speed Regards
Steve
Sorry to disagree. Logical were if Fabien went to court now and won. Then it were solved. But if not a panel cant have the same legal authority than a court of Law. So, everything Fabien did with his late appearance is causing vane suspicions.

Steve, as a chessplayer you could also take a deep look into the games of the Rybka versions that were playing already in 2004. I would prefer that chessplayers could make conclusions from the games instead of believing in suspicions.

Let's do some research.Regards
-Popper and Lakatos are good but I'm stuck on Leibowitz

User avatar
Harvey Williamson
Posts: 1801
Joined: Sun May 25, 2008 9:12 pm
Location: Media City, UK
Contact:

Re: David Levy asked for opinions

Post by Harvey Williamson » Thu Feb 24, 2011 4:41 pm

Rolf wrote:you could also take a deep look into the games of the Rybka versions that were playing already in 2004.
Here are some of those games pre Rybka 1.0 beta http://www.open-chess.org/viewtopic.php?f=3&t=798

User avatar
Rolf
Posts: 6081
Joined: Fri Mar 10, 2006 10:14 pm
Location: Munster, Nuremberg, Princeton

Re: David Levy asked for opinions

Post by Rolf » Thu Feb 24, 2011 5:54 pm

Thanks, Harvey, in the meantime I have them all. You cant download from the links in Openchess. But if you enter the Chesswar page you get it.

The chess is the important factor. But I had also some ideas while watching the standings.

The often quoted solitairy low result of Rybka is although correct, but it says nothing for itself IMO.

Look for yourself into these data. Rybka appears under different versions. Interesting is further that the 1.6 or so is less successful than later the 1.0 etc As I said everything should be carefully examined WITH the gamescores.

So, from that closer look until now I see no reason to argue as if out of these results alone one could conclude that something suspicious there must have happened.

For me it was interesting that at that special time Rybka had changed from USA to CZE. There the big jump later happened.

Another observation I saw one result where Rybka ended third or so with Elo of >2900, much higher than the following competitors, Fritz, Naum etc. So, Elo alone didnt guarantee to win.

This brings me to the idea that perhaps already then the cooperation with book author Jiri Dufek made Rybka so strong.

Finally it's also normal that a special hardware assistance made the real superiority of Rybka. And that isnt visible in such tournaments like Chesswar because there the hardware was weaker.

For me it's strange that nobody came with the typical chess proving of the similarity of engines. Why it wasnt possible to prove the ancestorship between R1beta and Fruit? Isnt it a proof for seperate existances?
-Popper and Lakatos are good but I'm stuck on Leibowitz

bob
Posts: 20573
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: David Levy asked for opinions

Post by bob » Thu Feb 24, 2011 6:04 pm

Rolf wrote:
bob wrote:
Mike S. wrote:If reverse engineering is illegal, then the so called peers have no basis whatsoever to judge upon, significantly.
Reverse-engineering is not illegal, however. But what you do with the reverse-engineered product could end up being so.
I see a serious problem with the known "Norman" who is kind of stalking Vas. So if this is brought into the tribunal by you, Bob, with all the other friends, I see no chance for a tribunal. Because it's normal that all only want to get the source of Rybka. So, Vas must be an idiot if he would agree to such a tribunal. There is no other commercial product that is competing with Rybka so that it would be incentive for a check up.
So, one of the two following statements is true:

(1) Vas lied when he said no part of Fruit is in Rybka, but that Strelka was an exact clone of Rybka 1, since there are similarities between strelka and fruit. SO if strelka is a clone of Rybka, and it has "fruity code" something is wrong.

(2) Vas lied when he said Strelka is an exact clone of Rybka. Which for us is irrelevant since we have not been comparing strelka to Rybka in quite a while since we have been comparing the R1 beta binary to Fruit.

Either way...


In my eyes the practice of these anonymous guys who somehow took Rybka and made their own thing, only to publish the source, what then destroyed the chances for Rybka again, is a real problem. If the community does not want to outsort these chaots, I see no chance for a tribunal either. Because then we had no Law.

I dont get why Fabien doesnt make the expected move into a legal case. If he cant trust that he could win, then you cannot expect from Vas that he gives into such a strange tribunal.
You keep saying "why doesn't Fabien move..." Give him time. I believe he is planning to "move" but it takes time to produce the necessary evidence. Trouble probably looms here, but it isn't going to happen at your pace, it will happen at Fabien's...

bob
Posts: 20573
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: David Levy asked for opinions

Post by bob » Thu Feb 24, 2011 6:07 pm

Milos wrote:
bob wrote:I have no real idea of the worth of the "world computer champion" title in advertising circles. But it is worth more than zero, and losing that might make a difference. And in addition to not being able to participate with a new clone or derivative, one also gets the stigma that will be associated when the plagiarism is made public...
Anyone can advertise what they wish.
However, ICGA calling its events "World championship" is just ridiculous. A private organization practically owned by David Levy decides about who is champion of the world is just preposterous.
Computer chess is so unimportant in circles outside of couple of these geek forums that it's not even worth to make an independent organization that would claim its own "World title championship" what is the case with real chess.
CC is also totally unimportant in academic circles. You can see that by number of theses and number of publication compared to other AI research areas. If the field was more interesting the progress would be much faster. Frankly speaking the field in which the total number of authorities is less than 10 and most of them are in fifth or sixth decade of life is far from being promising. In reality it's not scientific but engineering field without real commercial prospect, meaning it's doomed to be the back hole of real science and commercial engineering.

And having some idea about self-importance thanks to IBM or Cray self-(mostly hardware) promoting events from the past is nothing but delusional.
It's nice to be CC enthusiast, but obviously some ppl are lacking a thorough reality check...
Just for the record, ICCA (later to become ICGA) was created by chess programmers in 1977 at the 2nd world computer chess tournament in Toronto, Canada. Who else other than the people involved should do this? FIDE does the world championship. Except for the period of time when the PCA had their own...

Post Reply