Thoughts...

[Rolf]

My interests are a little more tightly constrained than that. I am interested in making my program stronger. And in doing so, I am interested in helping others achieve the same results. Ken Thompson, Dave Slate, Tom Truscott (Belle, chess 4.x, duchess) among others spent hours on the phone with me answering questions and helping me make my program better. And occasionally I had ideas that made theirs stronger. That's all I care about today. The commercial interests are off my radar. And always will be. If you look at 40 years of computer chess, almost all of the dishonesty, bending the rules, etc, have been done by commercial chess developers. Some of the early WMCCC events were classics in deception, dishonesty and conspiracies. Our non-commercial events went off without a hitch. Until the two were combined.

Hi Bob,

this is actually the last part.

First of all a question that interested me since long. I just ask even if it's stupid in the eyes of real experts.

In the times you created the CRAY chess computer or better the software for the CRAY, how was your code protected? Was it linked with the famous factory? So that a violation could have cost the intruder much money? I just want to ask this in our actual context because you seem so cool while debating a possible damage for the commercials like Rybka while long ago you were a commercial too, well without cutomers who could have bought it but for the company it was certainly a relevant piece of advertisement having a winning chess machine, no?

Did you have any protection at the time against cheating by opposing competition? Was that your job or was that done in the CRAY company?

Years ago you told me that those days had also a lot of stress compared to these days now. What was the reason? Could you explain it a bit? Money wise or what stress?

Another question: how was the control of new competitors done at those times? How could you experiment on their machines? Or was it more based on casual observations during the tournaments?

Back then, did you ever observe a case where something to copy had indeed been copied or stolen? How was that done, since you didnt distribute the software I assume.

Bob, another critical question:

would you totally exclude the possibility that a smart MIT absolvent could analyse the FRUIT details and then find a way to profit from that code even when he had no reason to copy the code in parts because the strength of his new engine was NOT to copy by definition since Fruit was weaker than Rybka.

Another technical lay question:

is there somewhere a collection (just as a summary of subjects) of known computerchess programming tools open to all interested? How many big topics?

Then:

A usual chessprogram, how much individually different % it contains compared with its open to all code parts? 5% or more?

Now the killer question without irony:

Why is it so difficult for making Crafty as strong as Rybka if you a) know Fruit like everybody else and b) might have got the secret tricks of
Rybka itself. Why do you chose to be below the commercial machines and now aslo these Fruit profiters?

Another important question from recent debates:

GPL topic. You argued but then didnt continue to explain that if Fruit were GPL and someone copied code from Fruit then also this copy is GPL. The question came if then Fritz and Shredder or other commercials who profited from Fruit would also be GPL in future? Or is there a method to copy that you couldnt discover? - That chapter was the reason why I asked you to help that the interested commercial programs for many million people should NOT be destroyed. And that you should help to protect them, because you as a scientist, are you not interested in the many users who use something that you helped to grow in the past decades? Your Crafty UCI is even officially on the page of ChessBase. So, excuse me, you cant be a total enemy of commercial companies in our hobby?

That's it for now. I only hope that the many readers understand the legitimity of the topics.

All the best to you for the weekend. -Rolf

[Dr.Wael Deeb]

There is a difference between speculation and having factual information to support a claim. Nobody has looked carefully at Fritz. Or at Shredder. Why? Don't know, don't care. I've not started an investigation into any programs other than the ones that have been suspected (and later proven) to be clones of my program. I just don't care about this stuff that much. I enjoy competition. Fair competition. I know there are cheaters out there. Always have been. Always will be. You can either worry and lose sleep over the issue, or simply ignore it. I enjoy winning when _I_ win. I'm not inclined to cheat, because that is not much of a "moral victory" when you know you didn't deserve the win. Many don't feel that same way. More power to 'em. I'm not going to let their lack of morals ruin a hobby I have enjoyed for 40+ years. If their moral compass is so far off that they can enjoy winning with things they didn't write, so what?

Thanks so much for your many detailed answers. I get one after the other because there are so many different topics.

Could we forget for a minute about business and commercial products?

Please imagine a boy who did not grow up with the famous antique UNIX times, colorless mathematics for people with that inborn talent. He plays chess, which is art, science and sport all together. Without any mathematical cells in his brain he wont follow your open source, Bob. But he will pray to God and thank him for the chance to live later than his forefathers who never had a machine computer system to collect the whole data and now even to train with such a strong tool like Rybka. Your hobby, Bob, is different. But please for a second minute just concentrate on the millions of chess players. Who actually have the chance to get to a veritable artificial GM in chess for 50 bucks. Wouldnt you have liked it in your youth?

Isnt it a bit religious if you are so determined to ignore the tools which satisfy also other senses than just our abstract mind process? Will you always use the more autistic language on ICC instead of normal English like here in your messages - if this is possible with the space you have in modern computers?

You dont need to change your habits but please try to understand the satisfaction that colors bring to your senses. Or do you insist that for you the symbols of your code are enough for you to explain the problem a novelist like Roth is expressing?

Just for now and the first exerpt please explain me and to all here the following problem:

In Rybka 1 beta for free, were there new details chesswise so that Rybka was much stronger than Fruit and all the others anyway? If yes, was that copied? From what original? If not, if it was new from Vas, how do you understand it? Was he creative and had created something new? If yes, the why whining about copied Fruit? How can a copy be 50+ Elo stronger than its model?

Bob another aspect, thanks for your last responses in general, they read really peacefully and I can learn without feeling pressure of any sort. Thanks so much.

It is quite easy to copy a program, and spend a year working on it to improve it. That has happened more than once. But if you copy a GPL program, your source is also required to be GPL if you distribute copies of the binary you must make the source available as well. That's point 1. Point 2 is that once you have a working program, and that program is in part a copy of another program, it is _very_ unlikely that those copied parts disappear. And that's a problem.

But for me, the problem is the original copying. It is simply not reasonable to copy someone's code, even if you modify it, and then pawn it off as your own original work. _That_ is a real problem of morals.

So we are left with a legal issue (GPL) and a moral issue (using someone else's code). Not exactly a never-before-seen occurrence, unfortunately. Nor the last.

After this answer, Rolf, will you have anything to say?

Instead of wasting your time in supporting a lost case, why don´t you gather to us to test RobboLioto new version?

Slobo,all the Rybka sectarians,fanatics,call them whatever you want will vanish into thin air with their fanatism called Rybka sooner than you think....

Somehting big is happening in the computer chess world and Rybka won't be a major event in the new world as in the rating lists as well....and these people defending blindly a piece of software will think twice before they post anything related to this topic when the time comes....I will personaly show them what rubbish they wrote back then....
Mind over heart regards,
Dr.D

[Rolf]

In Rybka 1 beta for free, were there new details chesswise so that Rybka was much stronger than Fruit and all the others anyway? If yes, was that copied? From what original? If not, if it was new from Vas, how do you understand it? Was he creative and had created something new? If yes, the why whining about copied Fruit? How can a copy be 50+ Elo stronger than its model?

It is quite easy to copy a program, and spend a year working on it to improve it. That has happened more than once. But if you copy a GPL program, your source is also required to be GPL if you distribute copies of the binary you must make the source available as well. That's point 1. Point 2 is that once you have a working program, and that program is in part a copy of another program, it is _very_ unlikely that those copied parts disappear. And that's a problem.

But for me, the problem is the original copying. It is simply not reasonable to copy someone's code, even if you modify it, and then pawn it off as your own original work. _That_ is a real problem of morals.

So we are left with a legal issue (GPL) and a moral issue (using someone else's code). Not exactly a never-before-seen occurrence, unfortunately. Nor the last.

Bob, thanks for the clear answer. But you didnt mention Rybka in particular, right? Or do you say what others imply, that this is exactly what Vas has done and what you discovered with others? Again, the other commercial and closed programs, will they be examined too? If no, why? - If yes, wouldnt that lead to GPL open programs all over the place? Who then will make progress??

[Matthias Gemuh]

Why is it so difficult for making Crafty as strong as Rybka if you a) know Fruit like everybody else and b) might have got the secret tricks of
Rybka itself. Why do you chose to be below the commercial machines and now aslo these Fruit profiters?

Bob answered that in the past: Crafty is kept simple for pedagogic reasons.
Young programmers appreciate Crafty source code because is has no complex (strength-bringing) tricks.

Matthias.

[Alexander Schmidt]

Sorry for pulling up the old thread, but when I open a new one about Ippo, it will be deleted because of "spamming". The moderation asked me to post about Ippo in existing threads.

Since we can't get a useful discussion here, I asked in the Rybka forum. I should have done this before.

http://rybkaforum.net/cgi-bin/rybkaforu ... 204064;hl=

VR didn't answer in the forum but we changed some PM's.

First of all, unlike some people here, he understand the questions and thinks it is necessary to give more details. He just can't do it at the moment. He will probably tell more details later.

What he told is: Ippo is a clone of Rybka 3, Rybka code ist used "extensively verbatim" and it is not used partly, it is "some of everything".

He did not give an answer on my question "How can you know it is Rybka code used verbatim?"

This is the main question to me. The code of Ippolit is generated automatically. I can't imagine how VR can see "verbatim" code of Rybka inside it. Beside that, verbatim code would result in identical behaviour in one or the other case.

He also told me why he can't give answers at the moment. I don't quite understand the reason yet, but OK. I can wait for further details, but they are necessary. And before some of you again start to offend me: VR has the same opinion.

Kind regards
Alex

[Steve B]

Sorry for pulling up the old thread, but when I open a new one about Ippo, it will be deleted because of "spamming". The moderation asked me to post about Ippo in existing threads.

not quite
moderation asked you to not start up new threads merely to ask a question that you asked in another thread(s) which went unanswered.
and we suggested you try asking that same question in existing threads otherwise the board can be filled with new threads merely asking the same unanswered questions from the same poster ad nauseum

Since we can't get a useful discussion here, I asked in the Rybka forum. I should have done this before..

thats right and it was myself that asked you if you tried asking in the Rybka forum since you did not get an answer to your question here even though you asked it 3 times
an idea that had not occurred to you until i suggested it
your welcome for the advise

Steve

[Rolf]

Sorry for pulling up the old thread, but when I open a new one about Ippo, it will be deleted because of "spamming". The moderation asked me to post about Ippo in existing threads.

Since we can't get a useful discussion here, I asked in the Rybka forum. I should have done this before.

http://rybkaforum.net/cgi-bin/rybkaforu ... 204064;hl=

VR didn't answer in the forum but we changed some PM's.

First of all, unlike some people here, he understand the questions and thinks it is necessary to give more details. He just can't do it at the moment. He will probably tell more details later.

What he told is: Ippo is a clone of Rybka 3, Rybka code ist used "extensively verbatim" and it is not used partly, it is "some of everything".

He did not give an answer on my question "How can you know it is Rybka code used verbatim?"

This is the main question to me. The code of Ippolit is generated automatically. I can't imagine how VR can see "verbatim" code of Rybka inside it. Beside that, verbatim code would result in identical behaviour in one or the other case.

He also told me why he can't give answers at the moment. I don't quite understand the reason yet, but OK. I can wait for further details, but they are necessary. And before some of you again start to offend me: VR has the same opinion.

Kind regards
Alex

So now you pretend that you are on level with Vas. How crazy!

The whole message is crap for the following reason:

1) Vas himself has NEVER claimed that his verdict that was in very shortness "Clone!" would be the end because he were God.

2) Stupid and evil critics of Vas have insinuated this sort of crap. That is what happebned here and you give just another version of that crap.

3) The hype is that after Vas told you what he told everybody before that for now he wouldnt want to make explanations, you appear here and insinuate that because of your oh so correct question Vas suddenly saw the necessity of having to deliver further explanations. Also this is total crap.

4) So, Vas repeated to you that the clones ARE clones. You enter here again and report it what all honest people have always trusted Vas for but YOU never. And now because Vas told you that nothing for now you appear and pretend that you have convinced him that it's necessary to explain.

5) You are with some others interested into these clones. For what it's worth you support chess cyber terrorism. Vas told you now that yes it was clear, clones! And now you do as if you could have it both ways. You campaign for the blackmailers and cloners and are against Vas but because he told you "Not now but later, wait a little bit" you are healed?

Let me summarize.

Nobody half way sane in his mind did ever pretend that Vas saying "Clone!" were a proof of almost scientifical value. But as I could rxplain, in the logic of the business, in special Vas partner ChessBase, it was sufficient, what Vas told the public. Possible ChessBase got more information. Anyway on their server no clone tatatatu.

So what are you reporting here. Vas said "Later" and because he told you so you might still think that the clones will have Freedom of Thieves and are allowed to play and to be tested. But exactly this wont happen.

And now??

Keep us informed about the human rights of your anonymous heroes and please on a regular basis, I would expect every hour all day long. That would also shorten the time we all must wait until Vas will finally resign to A. Schmidt and deliver his proof. BTW he cant have proof because it was all lies when Vas said "Clone!". At least in your dreams but not in reality.

[bob]

In Rybka 1 beta for free, were there new details chesswise so that Rybka was much stronger than Fruit and all the others anyway? If yes, was that copied? From what original? If not, if it was new from Vas, how do you understand it? Was he creative and had created something new? If yes, the why whining about copied Fruit? How can a copy be 50+ Elo stronger than its model?

It is quite easy to copy a program, and spend a year working on it to improve it. That has happened more than once. But if you copy a GPL program, your source is also required to be GPL if you distribute copies of the binary you must make the source available as well. That's point 1. Point 2 is that once you have a working program, and that program is in part a copy of another program, it is _very_ unlikely that those copied parts disappear. And that's a problem.

But for me, the problem is the original copying. It is simply not reasonable to copy someone's code, even if you modify it, and then pawn it off as your own original work. _That_ is a real problem of morals.

So we are left with a legal issue (GPL) and a moral issue (using someone else's code). Not exactly a never-before-seen occurrence, unfortunately. Nor the last.

Bob, thanks for the clear answer. But you didnt mention Rybka in particular, right? Or do you say what others imply, that this is exactly what Vas has done and what you discovered with others? Again, the other commercial and closed programs, will they be examined too? If no, why? - If yes, wouldnt that lead to GPL open programs all over the place? Who then will make progress??

How clear have I made this already? I looked at the "threesome" of Fruit, Strelka and Rybka 1. They have common pieces of code. Significant common pieces of code. It would be incredibly time-consuming to verify that strelka and rybka are 99% identical. But they have far more than enough in common to convince any computer scientists that they were not independent developments. So yes, I firmly believe that Rybka 1 was derived from Fruit. I won't venture to guess what percentage of fruit remains today, but since that is not the issue being discussed, that is irrelevant.

Just compare my source to (say) gnuchess to see how much in common you can find. It would be interesting to take the source of each, sort them into lexical order, and then use "diff" to report how many duplicate _lines_ of code are present. This won't identify duplicate blocks, but just duplicate lines will be quite revealing, since it is probable there are none, except for the lines that end loops and procedures and have a single "}" character on them.

I'll do this later and post the results, just for fun.

This is not just a "random chance" event.

You keep asking about the other commercials. And I keep giving you the same answer. Should someone choose to reverse-engineer one as was done with strelka, then they would quite likely be compared. Until that happens, I doubt anyone is willing. I'd rather spend my time working on my program, not looking at someone elses. And every time you ask this same question, I am going to give you the _same_ answer. It is _not_ going to change.

[bob]

You forget a key point. Kasparov got up on "the world stage" at the post-game press conference, and carefully suggested that somehow the DB/DB-team were "cheating". At that point, who in their right mind _would_ cooperate with him? He instantly turned the match from a tremendous scientific experiment to a soap opera complete with conspiracies, cheating, subversion, etc.

Sorry, I have to disagree. If it was a serious experiment i the eyes of the US side then the team always had the occasion and duty to redirect Kasparov on track. Yes, perhaps he gambled and asked for too much but the team had no ther chance than to agree because without his cooperation a real result couldnt be obtained. Yes, the event had a result anyway but you know like me that a disturbed opponent (here also client for the event) wouldnt be what you wanted to beat. Question was if the machine could beat the best human player, not the best player being psyched out. That was exactly the argument of the late German GM Unzicker, who was a judge BTW in real life. But also for me it was piece of cake to find out the absolute no go in the affair. Yes, Bob, you are probably right, Kasparov wass factually a sore loser, but still in the decades to come the event will be linked to the unsensible treatment of Kasparov by the IBM team. Look, if they had done what he wanted and he really had been the ognorant in CC just like you assumed, Kasparov wouldnt have suddenly got the clue for the future games. I fear that they couldnt show the data because Kasparov would have seen that there was something fishy. Again, the team consisted out of experienced scientists who knew that the experiment would be without valid result after they made upset their client, so it was in their hands. If you say, what you did, that it was a IBM decision not to cooperate, then they went for the money but betrayed their science basics. They should have told IBM that this way they would lose all the incentive why they had done the whole effort. But apparently they went for the money. No bad thing in the USA. But it ruined their fame. Kasparov however also lost something. Because also he continued the match just for the money. He felt cheated but continued just for the money. And erroneously he hoped to win another game and then the match. In that respect your critic of him is valid.

what a convoluted bit of reasoning. I would not (a) accuse someone of cheating and then (b) expect them to act courteously to my accusation and help me investigate that claim. That line of reasoning would never occur to me. Nor to most others. Once you accuse someone of cheating, the relationship immediately becomes adversarial. And it is unlikely to change for a good while after that.

[K I Hyams]

DO you have any proof that the program is a clone or any proof
that it is not a clone.

I have the proof:
1. Ippolit is a clone of Rybka.
2. It is not a clone of Rybka 3, it is a clone of next version of Rybka.
3. It can not be the result of reverse engeneering. It was made from original sources of Rybka.

"Vasik Rajlich: "There was an open-"source" (using the term loosely) clone of Rybka 3 released in the spring. Unlike the last time, there was no real attempt to hide the cloning - the hackers were even kind enough to keep me updated via email".
You can't both be right, or can you?