Author: Hans Gerber
Date: 18:08:55 05/12/00
Go up one level in this thread
On May 12, 2000 at 17:41:17, Albert Silver wrote: >On May 12, 2000 at 14:06:55, Hans Gerber wrote: > >>On May 12, 2000 at 13:28:56, Robert Hyatt wrote: >> >>>On May 12, 2000 at 11:07:04, Albert Silver wrote: >>> >>>>On May 12, 2000 at 06:16:13, Hans Gerber wrote: >>>> >>>> >>>>>>I wasn't there personally, so I can't say who attended which meetings, and I >>>>>>don't plan on bugging Hsu to ask for recollections about what happened. >>>>>> >>>>>>Clearly by round 3 he suggested something was wrong. Little doubt about how >>>>>>to interpret the comment Albert gave you a link to. A direct accusation... >>>>> >>>>>Two aspects: 1) the exact moment of his question and of the denial of the >>>>>prints, 2) the difference between events in reality and the reality of articles >>>>>in newspapers about "events". Let us analyse before we make our conclusions. >>>>> >>>>>You hypostated a direct connection between his _public_ question (which in >>>>>itself is accusation and insult in your view) and the (therefore justified) >>>>>_reaction_ of the DB team and IBM. Up to this moment I don't have any proof that >>>>>Kasparov made any (public) accusations after game two. >>>>> >>>>>The article by B. Weber allows the ref went into the public with Kasparov's >>>>>request. Thus the request was stamped as impolite and based on a kind of >>>>>confusion. But _if_ it was made in discretion and in the belief of friendly >>>>>terms of talking? >>>> >>>>Kasparov isn't really know for his discretion, and if it had been printed >>>>against his will, he would not still be making this accusation _publicly_ until >>>>today. Here is a link to a speech given before the students of The Humanitarian >>>>University, St. Petersburg just now, April 10, 2000: >>>> >>>>http://www.clubkasparov.com/serve/templates/folders/show.asp?p_docID=4954&p_docLang=EN >>>> >>>>It speaks for itself. His argument still goes: if a PC program never plays the >>>>move, then DB couldn't have either. BTW, make sure you get the whole link as >>>>many browsers cut split it after the '?'. >>>> >>>> Albert Silver >>>> >>>> >>>> >>>> >>> >>>This is _still_ utter rubbish. Complete logs from all 6 games are public. >>>He says they have never been released. The logs for the key moves are easy >>>to understand, even though he says they are not (I believe part of this >>>confusion was caused by Amir's comments about the logs). >> >> >>Fact is that the DB team did do nothing to prevent such a confusion, if it's a >>confusion at all. This is not the way scienctists should behave... >> >>It's time to look through the confusion we have here in our debate. >> >>Kasparov is one of the best chessplayers but he is not a scientist nor a famous >>expert in computerchess. We should not doubt his honesty when he makes his >>claims. >> >>It doesn't look "good" to read "jerk" in your articles on and on. The question >>must be allowed if you are still an objective scientific observer and thinker >>for the whole question. A teacher, sure, can accuse his pupils of being dumb >>nuts, but does this analogy is allowed in case of Kasparov, the chess genius? It >>seems as if you wanted to continue the impolite, arrogant and aggressive style >>of Murray Campell in the NY Times' article... Why? >> >>Kasparov said that "from all what we know about computerchess" no computer would >>play such and such moves. >> >>Now -- if this is completely false, why not contact him or one of his >>confidentals to give him the opportunity to update his knowledge -- if you are >>sure that he is badly wrong? > >Do you really think this hasn't been done? First he doesn't even bother to try >to back up his claim of 'all we know' with anything so substantial as an article >or a quote from someone from the field of computerchess. At no moment since he >began his accusations, have I read or heard anything to substantiate his claims. >What I did read, thus constituting his 'all we know', is that even after days, >neither Fritz nor Hiarcs ever played some of the moves played by DB. He even >gave specifics of such moves, such as a particular h5 played by DB as black in >the opening (I'd have to look up the specifics). This was almost IMMEDIATELY >shown to be absolute rubbish as Junior 5 (at the time) played this move without >any qualms or encouragement. This was undoubtedly shown to him as he gave up on >this and went back to his ranting. IBM released the information requested on the >moves where cheating was suspected, but this wasn't enough to quiet him and he >has gone on and on. Now he is a little less direct about his accusations and >states that DB wasn't designed to play the best chess, but just to beat HIM. I >have no idea what this means. He plays the best chess in the world. No one >doubts this. IBM least of all, else they would have sought that other best >player to match DB. He wasn't trapped in any opening other than the 6th game, >and that was his doing alone. So what was the secret formula that IBM fed to DB >to beat him other than playing the best chess possible? He never says. Instead I >keep reading about how it was designed to beat him specifically. I'd LOVE to >understand how this could be done without playing the best possible chess. I think Kasparov means something else. From the whole history of developing DB he concluded (IMO that was justified) that the goal had been the construction of a machine that could play real good chess. He saw his part in testing that "strength" with his capabilities. Now can you understand his surprise when he realized that the DB team with their GM had tuned the machine just to play Kasparov, his openings and possible anti-computer play? More. He himself had never thought about a serious preperation against the machine. He was convinced that his actual strength was still far better than the machine. More. Now the most important point. If the machine had been made just to play good chess, why should they hide the logfiles? Game one had shown that Kasparov still dominated the machine. Game two. Kasparov chose a bad opening, one that left him with almost no counter-play. Still Kasparov managed to create a situation (that he must have seen in his mind long before) where he _thought_ that the machine would tumble, note, the machine in a "clear" better position if we take GM chess in comparison. But against a GM collegue Kasparov would never have chosen the opening anyway. IMO Kasparov gambled against the machine's limited depth. And he was surprised, that DB didn't fall into his trap, DB didn't go after the material but played a strategically strong move. Now he wanted to see the logfile. First they agreed then said no. His suspicion became stronger. His plans to play the machine were no longer realistic. What could he do if the DB team with their GM had prepared the machine exactly against such typical anti-computer tricks. Giving material in a distant area and then to create threats in more important regions... Then the aspect of the belief in the supernatural. It's well known that Kasparov is an easy victim for all sort of such things. The miracles with number 13 etc. The moment they had denied the prints Kasparov felt himself in a war that reminded him of the old times in his own country. Sort of constant psychological pressure. The complete contrary to an open scientifical situation, with him as the hero not to forget. The prints might not have helped him too much. But that's not the point. More important were the feelings in front of an invisible machine, that could not be controlled. That he could not understand. Cheating could not be prevented - in principle. Nor could it be proven when a cheating had happened. Is it now more understandable why Kasparov felt himself trapped in a kafka-like situation? Baseline. The formerly friendly relationship between DB team and Kasparov was destroyed. He understood that they were to get him. Absolutely not a situation he wanted to attend. Perhaps all his reactions were wrong, too exaggerated, fact is that Kasparov had no aid, for example as R. Hyatt could have been. With such an aid Kasparov surely wouldn't have "left" computerchess. So, I do hope that a new challenge machine vs man will be prepared when the smoke of the 1997 desaster will be gone. > I say >this without the least bit of sarcasm. > >Regarding his partnership with computerchess: what partnership? The minute they >became a serious threat he refused to confront them. Did you read his comments >on his refusal to play in Frankfurt if Fritz were a participant? He said that he >clearly remembered the difficulties he encountered when he played Genius in >London years before. He certainly uses computers enough and has said so on >innumerous occasions. Regarding his preparation against Anand last year, he said >that there had been a time when he would have been afraid to enter such >positions, but no more thanks to the exhaustive analytical capabilities of >programs that he used. He has also stated many times that his continual training >with programs has seriously improved his own analytical capacity, and even >mentioned that his classic book 'A Test of Time', which he had considered to be >immaculate for many many years was now shown to be so full of holes that it was >almost embarassing. These holes were revealed by his PC programs. > >I completely agree that he is very human, but his human failings have been >stomping all over people that he believes made him look bad. HE lost the match >against DB. No one lost it for him. His human failing is that he will NEVER EVER >assume responsibility for this. The only time was right after resigning the 6th >game, when he explained that he had been afraid of the computer, but after that >he changed his speech: It's IBM, it's the DB team, it's Hsu himself. They were >out to get him and poor Kasparov hadn't realized it. When it seemed clear that >he was not going to get a rematch, he went on and on about the injustice of it >all. Yet when Hsu picked up all the marbles in order to try to grant him just >that, and at the same time probably revolutionize computer chess and the chess >world by putting DB on a PC card, he suddenly wasn't interested. It was this, it >was that. Uh-huh. Right. > > Albert Silver Kasparov is surely not the optimal opponent against a machine. But one thing seems to be clear. He's a typical human being with big emotions. If you bring him into a situation of ice and unfriendlyness, better, a situation where you pay no tribute to his big self-esteem, you kill him. But that wasn't the intention of the DB team? What do you think? Some data point in that direction. No wonder that Kasparov has enough of that role he had to play in 1997. (Did he make (and still does) mistakes himself? Of course! But that doesn't change the main points at all. He was engaged as a chessplayer and not as the most diplomatic politician.) I want to thank you for your article. > > >> >>This afternoon, following A. Silver's hint, I visited Kasparov's site and tried >>to send him "feedback" about the crucial question. Unfortunately the >>transmission didn't function. >> >>This was my text: >> >>quote: >> >>Actually there is a debate going on in CCC about the match against DB in 1997. I >>want to invite either Gary Kasparov himself (perhaps with a pseudonym) or the >>editor of this site to take part and to argue against the statement by Robert >>Hyatt that "it is wrong what Kasparov always is repeating, that 'all what we >>know about computerchess is NOT made to explain why DB played the way it did in >>the second game'". >> >>It was mentioned that even micros find the moves in question. What are further >>arguments from the side of G. Kasparov? >> >> >>Sincerely Hans Gerber >> >>end of quote >> >> >> >>> >>> >>>> >>>> >>>>>At the time of this newspaper's article, the request was not >>>>>even known in public. So, the article documented a turning point, initiated by >>>>>the DB team and IBM, in person of M. Campell, the second man behind Hsu. Perhaps >>>>>you are right that the later official statement was given by Tan, but in this >>>>>article there is a clear presence of the DB team to be seen...
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.