My recent correspondence with Vasik Rajlich

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

Moderators: hgm, Rebel, chrisw

Sven
Posts: 4052
Joined: Thu May 15, 2008 9:57 pm
Location: Berlin, Germany
Full name: Sven Schüle

My recent correspondence with Vasik Rajlich

Post by Sven »

Dear readers,

in case someone is interested, please find below the full contents of an email dialogue I had with Vasik Rajlich during the past two days. I did not remove or edit any part of the original emails except mail addresses, line break issues, and presentation of URLs. Both involved parties explicitly allow publishing of these emails in the TalkChess forum.

To the CCC mods: I suggest to leave this article here in the "General Topics" subforum as I think it might be of long-term general interest for everyone interested in computer chess, although I would also accept a different decision if that fits the current policy better.

As always, rational and civilized comments are welcome. I propose to concentrate on Vasik's statements mainly, not on the contents of my first email which does also contain my personal opinions and had the intention to introduce myself since I had to assume that Vasik did not know me until then.

Unfortunately my numbering of items was slightly inappropriate, in that I reused the numbers 1..5 in my second email instead of using 6..10. So please use something unique for that when replying.

My first statement after that dialogue is:
a) I can see a small step forward in both "state of Ipp* engines" and "R1 vs. Fruit" topics, and
b) both topics are not yet settled finally, more information will come in the future, and we should be patient.

To everyone again: please help to keep this thread respectful and clean. Thank you!

Sven
From: "Sven Schüle" <...>
To: "Vasik Rajlich" <...>
Subject: Request for your comments on computer chess topic
Date: Fri, 11. Jun 2010 12:26:29

Hi Vasik,

my name is Sven Schüle, I am a software developer from Germany and author of two chess engines "Surprise" and "KnockOut" (strength about 2000-2100 ELO). I guess you know "a little bit" about the current situation in the computer chess community since the day where "Ippolit" appeared, and especially the situation in TalkChess that has developed within the last ~8 months, although you are currently not an active CCC member.

I would like to read your comments on that situation. I strongly believe that the best thing to happen now would be to finally see some substantial steps towards the truth. There have been some partial efforts to compare "Ippolit" or its descendants against Rybka, some via the "black box testing" approach and some via disassembling, but none of these could really reveal the whole truth as far as I know. I think you are the only person, except the anonymous "authors" of Ippolit, who can really help to make progress in that area. The lack of a widely accepted "proof" has already split the whole CC community, which is a bad thing in my opinion.

In case you are interested in knowing more about me and my thoughts before replying to someone "unknown": my personal opinion in that "clones" topic is as follows.

1) "Ippolit" is obviously no original work. The published source code has many parts that are not manually coded by a programmer but generated by some tool.

2) As many other chess programmers, I believe that large parts of "Ippolit" (but not the whole engine) are the result of decompiling another very strong engine, most probably Rybka.

3) If 2) is really true and the "master" engine is really Rybka then for me it is clear that

a) your copyright has been violated by publishing source code as "public domain" that resulted from decompiling a copyrighted product,

b) apart from that legal aspect, "Ippolit" is also illegitimate software in the sense of not deserving any acceptance in the CC community.

4) The reality is different, however, since we don't know the truth yet. By "we" I mean that there is no consensus about it in the CC community. Even if you say you know the truth, and I say I believe that, and hundreds of others say they believe that, too, there are also those other hundreds saying "innocent until proven guilty". And they are right in principle, since currently we are lacking any widely accepted proof.

5) The thing that complicates the whole issue is the now long-standing allegation that Rybka (here I mean 1.0beta mainly but some people do not differentiate carefully enough) were partially derived from Fruit 2.1 source code which is under GPL. My clear opinion is that after looking at the information presented e.g. by Zach Wegner I can at most see some similarities in selected parts of evaluation and some similar implementation of UCI handling code. Bob Hyatt repeatedly claims that "code copying were proven undoubtedly" but personally I don't believe that.


So what could you do to help proceed in that messy situation we are in now?

- Could you help to identify some of these "anonymous" Ippolit authors somehow?

- Could you post messages that might have been exchanged between you and them, since you stated once that they kept you up to date about their activities at some time?

- Could you perhaps even post, or otherwise publish, some parts of Rybka source code ( :shock: sounds immoderate, I know, but is meant with good intentions only) that can help to prove undoubtedly what the origins of "Ippolit" are? Maybe you don't like this idea but would be willing to give relevant parts of "close to R3" source code to someone trustworthy and "neutral", like Dann Corbit for instance? (I assume you don't have the exact R3 source version anymore but "close" could be good enough.)

- Could you help to clarify the situation regarding "Rybka <-> Fruit" relationship? I know your statement that Rybka is an original work, and I really believe it. However, the CC community, here driven by few speakers like Bob, seems not to accept statements only, so what can be done? Can you perhaps name some parts of Fruit 2.1 ideas that have had the biggest influence on Rybka 1.0 beta, and describe in few words how you incorporated these ideas into Rybka, other than copying lines? (Which I assume did not happen, and which is also very unlikely to even work, from a software development point of view.)

- Could you do anything else on that issue?


I would highly appreciate your substantiated and valuable comments, and I am pretty sure large parts of the whole CC community would as well.

If you decide not to do that I will have no problem with it, just inform me, and add your reasons if you like.

If you decide to give comments then I propose to post on CCC "General topics" or "Engine origins (the latter requires logging in). Please drop me a note in case you post somewhere else since CCC is currently the only CC forum that I visit regularly.

You might also choose just to answer directly to myself, but in this case you would put the burden on me to decide which of your comments you would allow to publish and which not.


Finally I'd like to add that I have no personal, financial, or other connection to either side on that area. My only interest is to help finding the truth, in order to help proceeding in computer chess which currently seems deeply damaged by the "Ippolit" affair.


Sven Schüle
From: "Vasik Rajlich" <...>
To: "Sven Schüle" <...>
Subject: Re: Request for your comments on computer chess topic
Date: Sat, 12. Jun 2010 08:32:19

Hi Sven,

yes, these fun topics. :-)

Ippolit is disassembled Rybka 3 with changes. The changes are considerable but not even close to enough to leave any doubt. Robbolito is an evolved Ippolit, with more changes and more cleanup. I haven't checked the other new engines yet.

I'll definitely write up the Ippolit case at some point, for the historical record. Anonymous engines are not accepted by the CC community, so there is no hurry. I think it's best to wait one to two years before writing up an anonymous engine. Otherwise, cloners could use anonymous releases to get information, and then take more aggressive steps.

Re. tracking down the cloners: Not worth the energy, IMO.

Re. Fruit and Rybka: The Rybka source code is original. I did take a lot of things from Fruit, but legally. If there are some good concrete questions from credible people, please send them along.

Re. Rybka 3 source code: Unfortunately, I don't have it. (Yes, it was careless. I'm keeping the Rybka 4 source code.) It's not necessary for writing up the Ippolit case, but it would probably make a court case more difficult.

Best regards,
Vas
From: "Sven Schüle" <...>
To: "Vasik Rajlich" <...>
Re: Request for your comments on computer chess topic
Date: Sat, 12. Jun 2010 13:59:04

Hi Vasik,

many thanks for your immediate reply! The following points are still open for me, sorry for bothering you again:

1) Do you agree that I publish our email correspondence on TalkChess?

2) Since there are a lot of voices in the CC community requesting for more information about the Ippolit-R3 comparison: could you name those, say, three or four components of R3 at least where Ippolit source code is closest to R3, i.e. has the fewest changes? And maybe add a very brief summary of the biggest changes? Personally I accept your brief statements given so far but the level of acceptance for that in the CC community seems different.

3) Would you mind to publish one concrete example that shows how the sources match? Since that would be a piece of code that is already "public" I think that publishing also your version of it would not do any further harm to you and your business, except for perhaps making your coding style and some variable and function names public (which is of course more than "nothing", I know).

4) You wrote:
> Anonymous engines are not accepted by the CC community, so there
> is no hurry.
What makes you believe this is true? My impression is different, currently the CC community seems to be thinking like 50-50 about that, which may be surprising but is mainly driven by the "innocent until proven guilty" argument. Clearing the case once and for all very soon would be better than waiting, IMHO.

5) As to the R1-Fruit issue, you wrote:
> If there are some good concrete questions from credible people, please send them along.
Currently I can't send concrete questions but there is a website created by Zach Wegner some months ago that compares some code parts of disassembled R1 with Fruit 2.1 source, mainly concentrating on evaluation. The contents was compiled by Zach together with Bob Hyatt, and both are clearly "credible people" :-) Would you mind to comment on it a little bit? Here is the link:
https://webspace.utexas.edu/zzw57/rtc/eval/eval.html
being introduced with this CCC article (that started a huge thread):
http://www.talkchess.com/forum/viewtopic.php?p=328198
and later on continued to be discussed in various (sub-)threads, including these:
http://www.talkchess.com/forum/viewtopi ... 416#347416
http://www.talkchess.com/forum/viewtopi ... 258#355258

Just one last note: the R1-Fruit issue seems to be very, very crucial for the whole "clones" topic since the statement that R1 were *not* an original work serves as an important argument for claiming the application of "double standards" regarding the Ippo* engine family. Therefore also in this case I think that clearly removing all doubts as soon as possible would be very helpful.

Sven
From: "Vasik Rajlich" <...>
To: "Sven Schüle" <...>
Subject: Re: Request for your comments on computer chess topic
Date: Sun, 13. Jun 2010 11:36:19

Hi Sven,

(1) Sure, that's no problem.

(2,3) Let me address this all later in one statement, I think this will be the best way to handle that. It's on my to-do list.

(4) I meant rating lists, tournaments, etc. No doubt a lot of users won't really care about anything I have to say. :-)

The tie between author and engine is really important. If we allow anonymous engines, the field will become a mess.

(5) It seems to be guesswork. I'd estimate that 40% of the points are wrong, 40% are standard chess/computer chess concepts, and 20% are direct Fruit influence on Rybka.

Best regards,
Vas
gerold
Posts: 10121
Joined: Thu Mar 09, 2006 12:57 am
Location: van buren,missouri

Re: My recent correspondence with Vasik Rajlich

Post by gerold »

Thanks for the post Sven.
Nothing new in these statements.

Best,
Gerold.
User avatar
Dr.Wael Deeb
Posts: 9773
Joined: Wed Mar 08, 2006 8:44 pm
Location: Amman,Jordan

Re: My recent correspondence with Vasik Rajlich

Post by Dr.Wael Deeb »

gerold wrote:Thanks for the post Sven.
Nothing new in these statements.

Best,
Gerold.
Absolutely....he spoke a lot and yet didn't say anything worth hitting the submit button to create this thread....
_No one can hit as hard as life.But it ain’t about how hard you can hit.It’s about how hard you can get hit and keep moving forward.How much you can take and keep moving forward….
User avatar
Harvey Williamson
Posts: 2010
Joined: Sun May 25, 2008 11:12 pm
Location: Whitchurch. Shropshire, UK.
Full name: Harvey Williamson

Re: My recent correspondence with Vasik Rajlich

Post by Harvey Williamson »

Dr.Wael Deeb wrote:
Absolutely....he spoke a lot and yet didn't say anything worth hitting the submit button to create this thread....
If that was the criteria for hitting the submit button 90% of posts here are not worth the effort :P
Steve B
Posts: 3697
Joined: Tue Jul 31, 2007 4:26 pm

Re: My recent correspondence with Vasik Rajlich

Post by Steve B »

it does kinda sorta put things into a bit of perspective

ever since the IPPO engine hit the scene ..the Computer Chess world has been rocked to its core
depending on where you stand on the issue..
friends have become enemies and enemies have become friends
commercial sites have been threatening everyone they can with legal action
new forums are popping up all over the net on a daily basis
testing group members are at each others throats as to test or not
fan boys have risked life and limb to defend their favorite engines
moderators have been besieged from all sides
even entire families have been torn asunder

and during all of this time..
the one man that could have ended this debate.. at any time...with one post..
simply hasn't placed the issue up high enough on his "to do " list

Prioritizing Regards
Steve
User avatar
Dr.Wael Deeb
Posts: 9773
Joined: Wed Mar 08, 2006 8:44 pm
Location: Amman,Jordan

Re: My recent correspondence with Vasik Rajlich

Post by Dr.Wael Deeb »

Harvey Williamson wrote:
Dr.Wael Deeb wrote:
Absolutely....he spoke a lot and yet didn't say anything worth hitting the submit button to create this thread....
If that was the criteria for hitting the submit button 90% of posts here are not worth the effort :P
Post of the day....congratulations :lol:

:wink:
_No one can hit as hard as life.But it ain’t about how hard you can hit.It’s about how hard you can get hit and keep moving forward.How much you can take and keep moving forward….
User avatar
Dr.Wael Deeb
Posts: 9773
Joined: Wed Mar 08, 2006 8:44 pm
Location: Amman,Jordan

Re: My recent correspondence with Vasik Rajlich

Post by Dr.Wael Deeb »

Steve B wrote:it does kinda sorta put things into a bit of perspective

ever since the IPPO engine hit the scene ..the Computer Chess world has been rocked to its core
depending on where you stand on the issue..
friends have become enemies and enemies have become friends
commercial sites have been threatening everyone they can with legal action
new forums are popping up all over the net on a daily basis
testing group members are at each others throats as to test or not
fan boys have risked life and limb to defend their favorite engines
moderators have been besieged from all sides
even entire families have been torn asunder

and during all of this time..
the one man that could have ended this debate.. at any time...with one post..
simply hasn't placed the issue up high enough on his "to do " list

Prioritizing Regards
Steve
An excellent statement....couldn't say it better....thanks my friend :D
A big hug,
Dr.D
_No one can hit as hard as life.But it ain’t about how hard you can hit.It’s about how hard you can get hit and keep moving forward.How much you can take and keep moving forward….
benstoker
Posts: 342
Joined: Tue Jan 19, 2010 2:05 am

Re: My recent correspondence with Vasik Rajlich

Post by benstoker »

Sven Schüle wrote: Re. Rybka 3 source code: Unfortunately, I don't have it. (Yes, it was careless. I'm keeping the Rybka 4 source code.) It's not necessary for writing up the Ippolit case, but it would probably make a court case more difficult.

Best regards,
Vas
Vas says he doesn't have the source code for Rybka 3. Surely he doesn't mean he lost it. Did he just sell it to Chessbase or something? Is that what he means? Even so, why would he be so "careless"? What does that mean? Careless in losing the source code? Careless to sell it to someone?

He says he's keeping the source for Rybka 4, but doesn't have the source for Rybka 3. Therefore, Rybka 4 must have hardly any code from Rybka 3.
User avatar
Harvey Williamson
Posts: 2010
Joined: Sun May 25, 2008 11:12 pm
Location: Whitchurch. Shropshire, UK.
Full name: Harvey Williamson

Re: My recent correspondence with Vasik Rajlich

Post by Harvey Williamson »

benstoker wrote:
Sven Schüle wrote: Re. Rybka 3 source code: Unfortunately, I don't have it. (Yes, it was careless. I'm keeping the Rybka 4 source code.) It's not necessary for writing up the Ippolit case, but it would probably make a court case more difficult.

Best regards,
Vas
Vas says he doesn't have the source code for Rybka 3. Surely he doesn't mean he lost it. Did he just sell it to Chessbase or something? Is that what he means? Even so, why would he be so "careless"? What does that mean? Careless in losing the source code? Careless to sell it to someone?

He says he's keeping the source for Rybka 4, but doesn't have the source for Rybka 3. Therefore, Rybka 4 must have hardly any code from Rybka 3.
I think it is a clever use of words. He does not have the exact source from R3 that may be true. I bet he has source from a version very close to R3. It is not uncommon to produce several versions every week.
Albert Silver
Posts: 3019
Joined: Wed Mar 08, 2006 9:57 pm
Location: Rio de Janeiro, Brazil

Re: My recent correspondence with Vasik Rajlich

Post by Albert Silver »

Harvey Williamson wrote:
benstoker wrote:
Sven Schüle wrote: Re. Rybka 3 source code: Unfortunately, I don't have it. (Yes, it was careless. I'm keeping the Rybka 4 source code.) It's not necessary for writing up the Ippolit case, but it would probably make a court case more difficult.

Best regards,
Vas
Vas says he doesn't have the source code for Rybka 3. Surely he doesn't mean he lost it. Did he just sell it to Chessbase or something? Is that what he means? Even so, why would he be so "careless"? What does that mean? Careless in losing the source code? Careless to sell it to someone?

He says he's keeping the source for Rybka 4, but doesn't have the source for Rybka 3. Therefore, Rybka 4 must have hardly any code from Rybka 3.
I think it is a clever use of words. He does not have the exact source from R3 that may be true. I bet he has source from a version very close to R3. It is not uncommon to produce several versions every week.
If he doesn't have the exact source of the released version of Rybka 3, which presumably is what the Ippos stem from, then how is that a clever use of words?
"Tactics are the bricks and sticks that make up a game, but positional play is the architectural blueprint."