Rajlich/Osipov/Rybka/Strelka/Answers by Convekta

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

Moderators: hgm, Dann Corbit, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Edward German

Rajlich/Osipov/Rybka/Strelka/Answers by Convekta

Post by Edward German » Thu Jan 17, 2008 8:03 am

If you not know it, here you can read any answers by Convekta (by Victor Zakharov to Michael Diosi (Webmaster of Arena Forum).

Text (here in ARENA-Forum) is original and english only:

http://f22.parsimony.net/forum41668/messages/46023.htm

What is Your opinion about this Answers?

Tony Thomas

Re: Rajlich/Osipov/Rybka/Strelka/Answers by Convekta

Post by Tony Thomas » Thu Jan 17, 2008 8:46 am

He didnt provide new insight in to any of the questions. He kept repeating that only Vas can initiate any legal proceedings.

Edward German

Re: Rajlich/Osipov/Rybka/Strelka/Answers by Convekta

Post by Edward German » Thu Jan 17, 2008 8:55 am

Tony Thomas wrote:He didnt provide new insight in to any of the questions. He kept repeating that only Vas can initiate any legal proceedings.
In any Fora I have read that the name Yuri Osipov is only a pseudonyme name. In the Answer stand now that Yuri (Ivanovich) Osipov is (german: ..."vermutlich") his real name. Further stand that one of the persons there would released a engine by Convekta, was this Yuri Osipov. Convekta contacted then Vasik Rajlich, and Rajlichs answer was a no.

This thing is identical with them was Yuri Osipov wrote in the russian Forum by KasparovChess.

User avatar
Sylwy
Posts: 3572
Joined: Fri Apr 21, 2006 2:19 pm
Location: IASI (Romania) - the historical capital of MOLDOVA

Re: Rajlich/Osipov/Rybka/Strelka/Answers by Convekta

Post by Sylwy » Thu Jan 17, 2008 10:25 am

Edward German wrote:If you not know it, here you can read any answers by Convekta (by Victor Zakharov to Michael Diosi (Webmaster of Arena Forum).

Text (here in ARENA-Forum) is original and english only:

http://f22.parsimony.net/forum41668/messages/46023.htm

What is Your opinion about this Answers?

Hello !

I think all the future top release will be infected with ' Rybka's ideas" !
What to do ?
I DON'T KNOW !
What can us to do -for example-in real life , against some future suspect super releases from ChessBase,Lokasoft,UbiSoft........& so on ?

Regards,
Sylwy

Andrej Sidorov

Re: Rajlich/Osipov/Rybka/Strelka/Answers by Convekta

Post by Andrej Sidorov » Thu Jan 17, 2008 10:50 am

Sylwy wrote: I think all the future top release will be infected with ' Rybka's ideas" !
What to do ?
I DON'T KNOW !
Obviously from now any new program or version above 2800 Elo should be considered as illegal Rybka clone and Vas has a right to release them under his name. He can begin from Loop, or from Fritz 11, which is much closer to Rybka than to Fritz 10.

John Conway
Posts: 161
Joined: Wed Jul 11, 2007 11:44 am

Re: Rajlich/Osipov/Rybka/Strelka/Answers by Convekta

Post by John Conway » Thu Jan 17, 2008 11:30 am

Convekta rightly state that it is up to Vas to initiate legal proceedings, but I would have expected that they would have indicated that they will give him their full support (even if they don't mean it).

If Strelka had been a clone of Fritz, and Chessbase had been asked these questions, I think the response would have been a little different.

Andrej Sidorov

Re: Rajlich/Osipov/Rybka/Strelka/Answers by Convekta

Post by Andrej Sidorov » Thu Jan 17, 2008 12:04 pm

SzG wrote:Quote (1) - question by Diosi
"In the last month the engines Strelka and Belka were
AZ> released and advertised here: and
AZ> here . A
AZ> few days later those engines were discovered to contain
AZ> substantially code parts from Rybka 1.0.".

Good start. :) No code parts from Rybka could be discovered in these engines as no code parts of Rybka have been published.
You are right. But if they repeat this hundred times they start to to think that its proven fact :)

Andrej Sidorov

Re: Rajlich/Osipov/Rybka/Strelka/Answers by Convekta

Post by Andrej Sidorov » Thu Jan 17, 2008 12:08 pm

SzG wrote: So it is clear that it was not Osipov who wanted to go commercial. It was Convekta itself, the company who sells Rybka, who connected Rajlich with such an intention.
Oh, no! This cant be true. Saint Railich told us that scary Osipov blackmailed him to get permission to commercialize Strelka!

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

Re: Rajlich/Osipov/Rybka/Strelka/Answers by Convekta

Post by Rolf » Thu Jan 17, 2008 1:03 pm

John Conway wrote:Convekta rightly state that it is up to Vas to initiate legal proceedings, but I would have expected that they would have indicated that they will give him their full support (even if they don't mean it).

If Strelka had been a clone of Fritz, and Chessbase had been asked these questions, I think the response would have been a little different.
That is all clear after the text of that interview. I would ask myself if I were Vas if he shouldnt quit CONVEKTA for good. The point seems clear, I summarize:

CONVEKTA was seeking someone who allowed them to sell a mobile version with GUI which they had ready but they had no engine.

Then CONVEKTA contacted Russian programmers, again this is the story of Victor Zakharov from CONVEKTA. And whoopie a certain 'Yuri Osipov' offered them "his" engine and now for reasons a bit suspicious Osipov, not CONVEKTA contacted Vas if he would allow the publication.

If Osipov had created his own engine what Zakharov wants to insinuate (after all the information we have now!) then why should he have contacted Vas at all? So here the reason can only be that O had offered a program as his own but that was practically the engine of Vas. And normally CONVEKTA should have imediately have contacted Vas about this violation of his rights. But nothing the like Zakharov even admits that the company was eagerly waiting for a program that they could sell under the new GUI and just because "Vas wasnt ready with his RYBKA" they were willing to take a decompiled version of the RYBKA Beta. Note that Zakharov is the head of a programmer team, so he must have understood the illegality of the whole practice.

CONVEKTA was so eager to sell something that they themselves asked Vas the same for the Osipov clone (this is what it has been then and after Vas still now). So Vas is absolutely correct if he speaks of blackmail. Just count the known facts together. Suddenly a Osipov wanted to spring into the selling of Vas' creation and CONVEKTA was supporting that although Osipov didnt have a genuine original engine at all!

And to make the case even more suspicious Zakharov is now advising that a programmer should protect his code just what after Zakharov Vas has already done for his versions from version 2 on. This is all very cynical in the context that CONVEKTA would have used a stolen code - stolen by O? Or in order of CONVEKTA? - to sell a mobile program with a new GUI. Here already the copyright infringement is fulfilled.

But the crime case continues. After Vas had refused a) to O and b) to CONVEKTA to allow the publication of his code the stealer went public with this stolen code which is what we could all observe since May 2007.

O formerly was someone who decompiled in the belief that he could then become the "author" of the decompiled version of RYBKA if Vas had not refused. But after Vas had refused O then tried to harm Vas with a publication of first the program and then the whole code, under the name of Strelka and others even "cooperated" with Belka.

It's interesting how Zakharov is now reacting on the questions. I have only the English translation and that leaves open doubts in all really relevant topics.

The identity question "he dint try to hide" but Zakharov cannot say something with "100% certainty". So here every lawyer would have a problem.

Also in the question that interested mainly in this forum, if Vas potentially had told something exaggerated or false when he spoke of blackmail. Imaginme you reader got the contact that someone had decompiled your program, that you had formerly donated for free in the Beta version, and the contact person wanted to sell this decompiled version under Vasik's name apparently. But if Vas refused then the contact might publish his decompiled version later on. Wouldnt you feel blackmailed somehow by this contact and your own company? The whole case has similarities with another O in the field. When SMK had to leave Ossi Weiner because Ossi wanted to leave for good, SMK found a place in ChessBase where he basically worked then on Fritz mobile and could seperately well work on his Shredder which was sold by CB too. But SMK came to program Fritz mobile. In a way this Osipov figure is SMK only that he had no own Wch program. LOL.

I dont know anyone who could now continue to ask Vas should PROVE the identity of the Strelka/Belka stuff because that would further harm him because he then would publish his code which was NOT what he intended with the nice donation of his Beta.

I see a two variation problem.

Either Vas feels strong enough to continue on his own with the creation of a complete business but he must leave CONVEKTA for good because that company is IMO IMO IMO not trustable. Read the text of the interview yourself. Or Vas simply joins ChessBase as the new top programmer in a team with Feist and SMK. That would also solve the legal issues because ChessBase would at least have the manpower and the financial background to silence O figures of all sorts.

The link is given above but here is a short excerpt of the interview with Zakharov:


For a long time Convekta developed Rybka GUI for mobile platforms (PocketPC and Nokia Smartphones). As Rybka engine was not ready yet and Convekta needed to have some income from this development we agreed with Vasik Rajlich to sell GUI with another engine for a while. I sent proposals to a few engine authors for making their [sic!! Rolf] engines mobile and to release with Convekta GUI. One of them was Yuri Osipov. He agreed to publish his engine. Knowing about possible legal problems with Strelka we asked Vasik for his opinion about this. The message was sent by Osipov first and then I duplicated Osipov's message as Vasik didn't receive it. The answer was no. So this quesion was not considered any more.

AZ> 4) It seems that Mr. Osipov has disassemled the entire code
AZ> of the Rybka 1.0. Even if this is not the current engine you are
AZ> selling, the question we got by mail is what happens if he will do
AZ> this with the newest version of the engine also ? Any legal steps
AZ> to prevent this from happeneing ? How to protect Vas, your
AZ> customers and your own income ?

Convekta has no any rights for Rybka engine and source code and any steps can be started only by Vasik Rajlich.

From a few messages that I received from Yrij [sic! Rolf] Osipov I didn't see any intentions of him to continue decoding work.


------------------


So, the evil is, that Zakharov isnt interested in the least what the cloned stuff of Vas' Rybka is concerned as long as CONVEKTA can sell something... He doesnt even notice that he's talking about illegal practices from O right from the beginning because this isnt Osipov's engine at all, it's Vas' engine. This is all what Vas has told us before and where he told us the complete truth. It's beyond me where Vas should have lied or hidden something.
-Popper and Lakatos are good but I'm stuck on Leibowitz

ZeroOne

Storm in a teacup

Post by ZeroOne » Thu Jan 17, 2008 1:27 pm

Andrej Sidorov wrote:Obviously from now any new program or version above 2800 Elo should be considered as illegal Rybka clone and Vas has a right to release them under his name. He can begin from Loop, or from Fritz 11, which is much closer to Rybka than to Fritz 10.
I second that, that seems to be the only right way to go!

Seriously though, I think this whole episode is just a storm in a teacup... No, it isn't right to publish a new engine based on the sources of another, commercial engine and not tell about it. But I'd also expect the so called copied parts to be identified in the sources of both engines.

Also, in the long run this could benefit the computer chess community. This could boost the development of engines if some "secret" of Rybka's is revealed now. It is a little boring that Rybka leads all the rating lists, isn't it? And if Strelka's code was based on Rybka 1.0 beta, well, beta versions don't bring you any money anyway, and I don't think Strelka can be much better than Rybka 1.0 beta. Strong free engines have always existed and yet commercial engines are still sold. People might even buy more Rybkas now that they've apparently had a sample of it in the form of Strelka, and also just to show some support to Rajlich.

Post Reply