Slashdot Mirror
Kasparov Beats the World
Ender_Stonebender writes "I noticed on a link to a story about the World Chess Team resigning the game against Kasparov last night. "
← Back to Stories (view on slashdot.org)
← Back to Stories (view on slashdot.org)
It's clear microsoft was involved heavily in this game. It shouldn't be surprising that microsoft would influence this at all, it's a marketing scheme. Maybe microsoft bought Kasaprov... .. well, take 1m USGS photos of people's homes and businesses and .. sell them. And we all know microsoft for it's technical genius...(BS) unfortunately, many many other people think that microsoft is responsible for all the technology it markets. This is very sad when a company can use people's naiveity of technology to take their money.
try http://terraserver.microsoft.com
Here is an example of how microsoft uses it's wealth to
Microsoft put itself right by the side of the worlds best chess player, and made sure he won. also making sure kasparov got a fat check and microsoft got the rights.
Yes, he had a "team" and not only that, they used computers! to help them! This was not a secret. What you have "discovered" here is just the fact that you were unaware of something about the match. Congrats.
Otherwise you wouldn't make this statement.
If it was actually the grand masters vs Kasparov, then it had a better chance of success than a democratic "world" team, simply because any very good player is going to be better than the "average" player that would result. (a worse than average player is more likely- you have to factor in the risk avoidance that seems to grow with large groups)
you can't run a successful organization by committee. He won because he was a single person making the decisions rather than a voting block. That's why dictatorial countries are better than the US.
It's kind of silly to stick with theory when experiment has proved the theory wrong. It was indeed a real match, to the surprise of many (including me).
The endgame FAQ
The world's last chance at a draw was lost at move 51.
Therefore, Microsoft's problems at move 58 sped up the inevitable but did not change the outcome.
There was a version of this on the International Go Server (IGS), and may be another soon, depending on the how current this line from the message of the day is:
* NHK introduces Yoda Norimoto 9p vs ALL of IGS !!
That was a win for Yoda Norimoto 9p last time, but that was fast, almost certainly no chance for anything but voting.
IGS is good for playing GO on, but you want to be strong already, these people are strong, even at the lowest level.
The world's moves were better than any single person on the BBS b/c people would point out problems with different moves until there was a consensus "best move"
Also Kasparov was not alone. He worked with his small team of analysts and computers.
I just think this should help his crushed ego
... if it was taken from the pool of the 2nd - 10th players it would be more accurate.
... could we not come up with something even more powerful than deep blue?
... um most of us are doing other stuff ... (see point 1 again) ... um lets see Kasparov getting jiggy w/ gcc ... guess ms would prob. pref. "v(c++|j++|b)" ... urgh shudder to think
... um aren't there better things we human have to do than solve pspace complete problems? (yeah ... chess is not np)
... ms pr ideas are as bright as a 10 watt light bulb
but is it really the best the world has to offer consider the following:
1. playing the world is like averaging so it more like the 1st ranked player plays the say the 1000th ranked player
2. if all the talent of the world were mustered
3. how many people in this world actually make a living just playing chess?
and goodness
this is pathetic
As was previously reported here on /. this whole thing can't be considered as a real chess match. I think M$ can't seem to do anything right, as they are the cause of all the problems in this 'match'.
It's not Kasparov against the collective intelligence of the world. It's Kasparov against the average intelligence of the participating players. No wonder he won!
It was actually Kasparov against Irena Krush backed up by the Russian Grandmasters school and a lot of people on the BBS.
Different people would analyze different lines and point out pitfalls, and then everyone voted for the consensus best move.
Irena Krush admitted that she could not have played on the level that was played without the backup and support of the BBS
That's right: Deep Thought was Deep Blue's predecessor. It couldn't beat Kasparov, but it did take down some great names: Bent Larsen, for example, in a tournament at Long Beach in 1988, where it tied for first place with Tony Miles, ahead of Tal. It lost only one game in this event, to Walter Browne.
I don't have a number for total moves, but the the number of possible chess games is something like 321**6300,or roughly 10**15790. But this brute force approach to chess programming is not interesting. In practice, chess programs play better, the more able they are to leverage their `brute force' calculating ability with algorithms that mimic a human approach to evaluating positions.
Wrong. Deep Blue was tuned to play against Kasparov, true. But even its predecessor could regularly beat Grandmasters ten years ago, without being `coached' to play specifically against them.
it is not difficult to show that Deep Blue is smarter than the entire world put together.
that would imply that the "entire world" participated in the the match against him. i am a member of the set of all people in the world, and i did not participate. hence, your reasoning is false.
--Siva
Keyboard not found.
Keyboard not found.
Press F1 to continue.
I see a lot of you saying that people are 'forced' to pick one of the _suggested_ moves by the analysts. This is simply not true. It just goes to show you that some people here comment on things which they have no idea what they are talking about.
You can choose ANY move on the board you want and submit that move. That analysts are there only to supply sugegstions, not give us a list to pick from.
Also, what is this nonsense about "Microsoft trying to defeat Kasparov"? That post was moderated up as insightful?! Microsoft and Kasparov set the whole thing up as a PR campaign to get more interest in chess. Microsoft benefits because more people will visit their "Zone", and Kasparov benefits because more intrested in chess which is perfect for him as chess is his living and means of money.
People will bitch about anything that has the words "Microsoft" tied to it and I'm so sick of it. The bitterness here is overwhelming and if some of you people keep it up, more and more people will slip through the fingers of the Linux movement, disgusted with the childish zealots that taint it.
The "World vs. Kasparov" match is a good thing for people who love the game of Chess. Stop trying to turn it into your little OS holy war!
It would be possible, but if Slashdot ran the game, all the good moves would be moderated down to -1, and we'd end up losing...
(score: -1, mocking the moderators)
0 1 - just my two bits
It doesn't matter whether she's strong or not... I think her trainer provided strong back up force for her...
Btw I really wanna ask, is the 10th move her discovery? Or anybody has played this move before?
I read an interview w/ Garry Kasparav where HE stated that he fears Irina more than any other chess player in the world when it comes to taking his title away. I was just regurgitating that...
<SIG>
I think I lost my work ethic while surfing the web. If you find it, please email it to crispy@crotch.caltech.edu.
</SIG>
My sig has a broken link in it.
As was pointed out in the article. Garry's analysis led him to believe that even the move that Irina suggested would have been losing. Just not as fast as the route we ended up taking. I'm sure someone will do a thourough analysis to confirm it... but in all likelihood we had already lost at that point in the game.
<SIG>
I think I lost my work ethic while surfing the web. If you find it, please email it to crispy@crotch.caltech.edu.
</SIG>
My sig has a broken link in it.
um dunno you have to consider this is done by consensus ... consider how many people in this world are @ his skill level and how many people played ... don't quite think you get a synergy effect as you have to spend effort filtering out bozo's and as a result you also lose some bright ideas too ...
I like the sound of this. One of the problems with Chess is that it has not yet (AFAIK) become a 'solved' game. Checkers can easily (as can tic-tac-toe) be solved, where every possible outcome can be determined before each move. Even Kasparov has no chance of solving a game like Chess. I think he has the ability to see 7 moves in advance.
Solving Chess is quite a task... If, on every move there were 10 (at least) possible moves, this games would have 10^62 possible outcomes. Holy crap!
A client for this would actually be fairly straight-forward, as almost no strategy is required, just hard number crunching and results examination.
GRH
Ya, this would maybe let me be the grand master I'll never be. :)
:) I've never heard a definitive count, since it depends on how many moves the game runs for before there is a victor.
As to your questions:
a) the total number of move possible is, as you say, astronomical. Hey you don't need seti@home power to figure out tic-tac-toe
b) the hope, when 'solving' a game is to predict every possible outcome of the game from the very beginning. So that when Kasparov makes his first move, our chess@home goes "got you now Gary!" I don't think any computer chess game can see more than 10 moves into the future, but I could be off here. The point is, the more the better. The problem in tournament play is that there is only so long you can crunch on the problem before having to make a move. THis is what has limited all previous attempts.
c) This is an interesting question. Remember the time factor. Each participant would have to retur n their results before a move could be made. Some like a coordinating computer figures out all possible "first next moves". Each of these is then passed to a participant to determine the outcome.
d) AGain interesting. Obviously we're looking for a path to a checkmate, but we don't want to start throwing all pieces to do it. I think each available path to checkmate (there'd probably be thousands), would have to be evaluated to determine which path ensures our position is always the strongest.
e) ?, I have the partial skills and the partial time. Maybe I'll fool around to see how feasible this would be.
GRH
There were only three times in the Kasparov vs the World game where Irina Krush's advice wasn't followed. These were moves 51, 52, and 58.
58 was primarily Microsoft's fault. Irina's advice was not posted largely due to their NT server's error. At least that seems to be what they're implying at Irina's SmartFAQ. The game was definitely lost here if not earlier.
However moves 51 and 52 may have been due to many suddenly interested Slashdotters voting without having previously followed the game.
On Thursday, September 30th, 8:45 am, the news-story "Kasparov vs. The World: It's all different" by Hemos was posted on Slashdot. At 2 pm that same day, Irina's advice was not followed for the first time. This was move 51. Irina recommended 51..Ka1, but the world instead played 51..b5
Irina replied to this deviance:
The World chose a sharp and very interesting continuation with 51..b5 and must now deal with the discovered check on its King (after 52.Kf6+). I strongly recommend the move 52..Kc1.... In my estimation, 52..Kc1 is essential.
However, on the next move that Saturday, her essential advice was not followed. The world instead chose 52..Kb2. Irina claims that this was likely the losing point of the game (as stated in her SmartFAQ). To see the history of Irina Krush's (and the other analysts) comments, go to http://fdl.msn.com/zone/kasparov/gameanalysis.txt
Perhaps in similar future events, a stipulation giving more voting strength to longer term players could prevent such occurrences.
This whole time I had thought that the World was playing just Kasparov. But, in fact, Kasparov had a whole team working for him, as is evidenced by this quote from Danny King:
He [Kasparov] showed me just a few of the critical variations he and his team had worked on.
This quote came from http://www.zone.com/Kasparov/gameanalysis.asp
I would like to see a distributed chess match. You'd have teams A and B. Each team would get 1 month between moves. When it was b's turn team b would split all the possible moves up amoung their computers-and have each computer test all moves and all possible counter moves within their move range. Then chose the move that puts the least amount of your men at risk(rate them according to points) and gives you the best counter move(killing the highest person on their team according to points). Then let the computers go at it! I think this would be a MUCH more exciting match-because we've alread proven that a computer can beat the best human. Now we can watch the computers play each other.
.sig
matisse:~$ cat
kasparov wasn't playing the ENTIRE world by a long shot. I'll bet about 250,000 people at the most. Also deep blue makes logical decisions. Humans don't-we make emotional decisions. Deep blue chooses a move because it has the least possibity of error. Also you should remember that a group of 20 really good chess players could have been there. If they all voted for the most logical move it could get drowned out by the other 249,980 people. Many people would vote for a move purely because everyone else voted for it-it may have nothing to do with which move is best.
.sig
matisse:~$ cat
You describe Correspondence Chess. They did it in the old days by sending moves on a postcard. Tournaments ran for several years. Today, they do it by email. There would still have to be some time limit. Otherwise I just refuse to send a move if I realise I'm losing. Wait till my opponent dies of old age(or boredom) and claim the win.
An experimental match was conducted in Europe this year between Karpov and Anand. They played over-the-board games but were allowed to use computers for analysis. Viewers could see the computer screens to get an idea of what the players were doing. Anand won handily, primarily because he is much more computer-savvy than Karpov.
Bah. He lost to one computer, so what good would linking up a bunch of them do?
I actually think it'd be better to have a distributed chess project doing things like determining the best openings, and analyzing a bunch of different moves, with no human player involved.
--
Win dain a lotica, en vai tu ri silota
The game would actually have been won by the world team.
I don't have the exact position in front of me right now, but this is just before Kxg6:
... Nh8 (instead of queening)
Now g6 Nxg6 wins
Kf6 loses by a tempo.(White cannot queen because blacks queen reaches f8 first).
I can throw myself at the ground, and miss.
It was an interesting concept, unfortunately it was tainted by Microsoft (Note: Not a generic Microsoft dig, they just did a lot of bad moves, e.g. letting only MS Windows users decide at one point...)
:)
Would there be a possibility of a rematch if it wasn't run by Microsoft? Slashdot could run it
i dont display scores, and my threshhold is -1. post accordingly.
Discuss
IIRC they removed the Chess Programming and just used the box as a web server after that.
i dont display scores, and my threshhold is -1. post accordingly.
Discuss
Actualy, checkers is not solved.
void post { post_random_comment("slashdot.org"); karma--; }
Oops! Actually.
void post { post_random_comment("slashdot.org"); karma--; }
The world was in trouble point wise by move 11 when they ignored the treatened fork from the knight to king and rook and pressed on with the queen move. They recovered well enough to be 3 pieces ahead but GK whittled it down and advanced/protected his pawns on his right/blacks left where he could get another queen and do attrition on the black major pieces. The resignation was inevitable.
NPR reported this morning that while the advising team had voted to resign, the rest of the players voted to continue to the bitter end.
No...like every other move, resignation was chosen by majority vote.
Despite the problems with a few things, I think it went pretty well overall. The technology used wasn't the best, but then what do you expect?
-- CP
First off, wasnt it Deeper Blue that beat Kasparov? He clobbered the original Deep Blue.
Sure deeper blue won, IBM put millions into making the computer and programming it, for one match. Then if i remember right, they dismantled it. They may have realized that their machine got lucky and didnt want to risk a second chance.
icq:=22921393;
A computer-vs-computer distributed match could be interesting, but only if the computers are running different software (an AI programming contest?). But what would really be cool would be to run a distributed chess game vs. Gary Kasperov. Show the world that lots of little computers can do alot more than one big Deep Blue.
-------------
Deep Thought also tried to beat the world champ, and failed. Deep Blue was the next-gen chess playing system, and the name is a twist on Deep Thought plus the blue of 'team IBM'.
You know what to do with the HELLO. ...
Help create an open-source world
To give an example, I have never seen a team use an effective sacrificial combination as it takes a lot of courage to do this and having a team only makes it harder.
I am amazed as to how long the world lasted as team long term planning is so dificult.
Is it me, or are there few actual chess players on this discussion group.
I don't suffer from insanity- I enjoy it immensly!
Whats seems useful and meaningful to you might seem completely worthless and a waste of time to other people.
This sentence is false.
This does sound really intresting. A Few practical questions though:
a) How many moves total are possible in a game of chess? (I tried looking this up but couldn't find it, its some astrominical number)
b) how far into the future would these computers have to think?
c) how many moves would each "packet" consist of?
d) how would the moves be rated? (shortet possible moves to checkmate? most defensive?
e) who would code it?
Well, if we could answer these questions, I'd be more than happy to help set something up.....
One equal temper of heroic hearts,
Made weak by time and fate, but strong in will
Perhaps it could be one HUGE recursive thing.. Each compy searching 3 levels down (for EVERY possible move) then outputing those results, (with score) and bumping those results to other computers?
::chuckle::
The traditional pyramid scheme?
e-mail me, since your mail isn't posted.
One equal temper of heroic hearts,
Made weak by time and fate, but strong in will
Deep Blue was programmed to beat Kasparov. It was loaded with a helluva lot of Kasparov's games to give it a database from which to draw 'what if' statistics. Guarantee that any other GrandMaster would have whipped it.
> 1. playing the world is like averaging so it more like the 1st ranked player plays the say the 1000th ranked player ... if it was taken from the pool of the 2nd - 10th players it would be more accurate.
If each person had voted alone without any communication, then maybe you could have said this.
But actually, because of the BBS and the discussions on the BBS, it was like combining all their strengths. Even Krush (one of the four coaches) admitted that she could not have played at this level without all the contributions from the BBS.
Ok. Maybe I was still asleep when I heard the report and they said that the World was -expected- to vote against resignation (but hadn't yet). Or maybe (it pains me to say this :) ) NPR was wrong.
Daniel
Hurry up and jump on the individualist bandwagon!
We usually are.
;-)
As everyone knows, the intelligence of a group is equal to the minimum intelligence of its members divided by the number of members. This point has been proven often enough in software engineering.
Of course, teams are a little different -- that's the only reason we lasted even this long, we had a good team helping us.
If we want to ever do any good, we need a surgical chess team (c.f. Mythical Man Month). And when I say team I don't mean a bunch of people who just met, or met over the web!
Or perhaps this is just the beginning of the defeat of chess -- perhaps it'll be replaced by the One True Strategy Game, Go. Now THAT is a good game.
-Billy
For several reasons.
First, most of "The World" is not sophisticated enough to be able to see enough moves ahead to be able to make the right choice, but rather, the most obvious one.
Second, Irina Krush, who was selected for most of the moves, is not a *great* player. If she was a man, she would probably only be a strong master, or maybe an international master. I believe Kasparov can beat five grandmasters simultaniously (walking from one board to the next each move).
My brother who is a master (Yaacov Norowitz, look him up at http://www.uschess.org if you're interested), can beat Irina about %50 of the time in speed chess. Of course speed chess is different from regular chess, but it gives you an idea.
Women are not worse in chess than men, it's just that there are much more serious chess players who are men than women.
It has been widely known that vote stuffig has started at least at
move #4, maybe earlier. Kasparov has acknoledged that the
correct move #51 proposed by the core of world team would be
a straightforward tie. Furthermore, it has been known for a while that
some crucial moves in the middle of the game have been stuffed,
that time in accordance with the core team recommendations. It is
therefore hard to say that this was a fair game although the world
loss is due to its decentralized and uncontrolled nature as much as
Microsoft glitches and outright failures. If anything this game has shown
that a different structure is needed for this to be successful. Anyone
who claims this game to be a success is lying - it was exactly as the
pessimists have foreseen. Anyone who wants to repeat this game needs
to think about democratic process through the internet. This game shows
that just trusting the user is unacceptable.
I just think this should help his crushed ego
... if it was taken from the pool of the 2nd - 10th players it would be more accurate.
... could we not come up with something even more powerful than deep blue?
... um most of us are doing other stuff ... (see point 1 again) ... um lets see Kasparov getting jiggy w/ gcc ... guess ms would prob. pref. v(c++|j++|b) ... urgh shudder to think
... um aren't there better things we human have to do than solve pspace complete problems? (yeah ... chess is not np)
... ms pr ideas are as bright as a 10 watt light bulb
but is it really the best the world has to offer consider the following:
1. playing the world is like averaging so it more like the 1st ranked player plays the say the 1000th ranked player
2. if all the talent of the world were mustered
3. how many people in this world actually make a living just playing chess?
and goodness
this is pathetic
>Thanks to Irina Krush. She will take away Garry's title someday.
Not likely. Krush is vastly stronger than any of us, but that doesn't mean she's stronger than Kasparov. Like him or hate him - and I'm not exactly a fan myself - the objective evidence indicates that Kasparov is the strongest player in the world today and probably the strongest ever. Even the loss to Deep(er) Blue doesn't change that, because DB was tuned specifically to play against Kasparov and would fare worse than Kasparov if both were to play in a round-robin with the other 14 top players. It's not uncommon for a less-skilled player to maintain a winning record against one of their betters because of style issues, but it doesn't change their relative strengths.
It's possible that IK could someday be better than Kasparov. In fact, that's highly likely since she's younger and chess ability under tournament/championship conditions does seem to fade a bit with advanced age. What's more interesting is the question of whether IK might someday be the best player in the world[1]. Yeah, it's possible. However, it's also possible that any of a thousand other players - from Anand to Morozevich to Khalifman to Xie Jun - could raise their own abilities and forever stay above her. Remember, they're also highly disciplined professionals capable of analysis just as deep as IK's. Statistically, the chances that IK will ever be the best player in the world are slim.
[1] One of the English chess columnists - Miles, I think - has written a far better piece than I ever could about the difference between being the best player and being world champion. It's somewhere on www.chesscafe.com IIRC.
>We need to commend her on her efforts she put a ton of work into the game! She contributed months and months of painstaking, in-depth analysis. Thanks, Irina!
Absolutely. The above comments about her ability aside, her devotion and passion and work ethic are highly commendable. She did a great job as "team captain" for the world.
Slashdot - News for Herds. Stuff that Splatters.
http://dailynews.yahoo.com/h/nm/19991020/wr/chess_ microsoft_5.html
has an accusation that a critical move recommended for ``The World'' was not posted
by Microsoft, thereby effectively losing the contest, or at least eliminating the possibility
for a draw. GK denies this, claiming the different move wouldn't have mattered.
JMR
Try e-gold - (contact me). I'm NOT e-
Longtime opponent Deep Blue, when asked about Gary replied in an email that, "He keeps ignoring my phone calls."
Seems to me, that where there REALLY shouldn't have been any involvement from ANY outside parties, Microsoft decided to stick its nose in at least once, and quite possibly twice, depending on how trustworthy one considers them to be.
Don't get me wrong, I'm not alleging that Microsoft somehow cheated, or allowed Kasparov to cheat. What I am saying however, that in a scenario in which NO participation was needed, ANY involvement from a third party is going to be misinterpreted. For better or worse, it's going to hinder or help at least one side. In a battle of wills between two people (counting "the world" as one entity), any involvement from any third party is damaging to at least one other party.
My question, is why did Microsoft bother to stick their nose in? I mean, I can almost understand Microsoft not allowing them to stuff the vote, but still, it was 66% FOR stuffing. Now, in my opinion, not allowing resignation is interference.
Also, I haven't followed the article (don't play chess, don't like chess. sorry), but does anyone know how many members/teamleaders there were? I'd be interested in the ratio.
Another poster suggested that it wasn't Kasparov against the most intelligent competition, but rather, K against the AVERAGE intelligence of the competition, which I don't agree with, but would skew heavily depending on how many members/team leaders, the ratio between them, and how much conferencing was done OFFline to discuss strategy. Otherwise, Kasparov is able to see everyone else's decision making process, which, tends to skew data as well...
"I've said it before, and I'll say it again - democracy just doesn't work."
-Kent Brockman
You know what to do with the HELLO. ...
Help create an open-source world
I totally agree. The amount of information she provided gave me a great insight on how a grandmaster thinks. Irina Krush has shown (to me at least) that she understands how to take her ability to another level (by translating it to someone who ISN'T a chess professional).
Garry Kasparov better be wary, she is only getting stronger..
-Dextius Alphaeus
-- Java is not a Jedi trait... "do, or do not, there is no try" --
It was originally called Deep Thought where it was originally created (CMU). After it moved to IBM they renamed it Deep Blue.
NPR reported this morning that while the advising team had voted to resign, the rest of the players voted to continue to the bitter end.
Daniel
Hurry up and jump on the individualist bandwagon!
Let's see... I'm thinking back to the media coverage of the Deep Blue fiasco. By the kind of thinking so much in evidence at the time, Deep Blue beat Kasparov at chess, ergo Deep Blue is smarter than Kasparov (witness the outbreak of "will computers replace humans?" panic stories).
By similar reasoning, and based on the fact that Kasparov beat the entire world at chess, it is not difficult to show that Deep Blue is smarter than the entire world put together. Perhaps Deep Blue ought to be renamed Deep Thought.
"Never let your sense of morals prevent you from doing what is right" -Salvor Hardin
...when it was clear that we had lost. This was only a formality. There were many that wanted to play Gary to checkmate, but in a true professional match, one resigns immediately when ones defeat is assured (or at least very strongly suspected.)
I think that with all that we have learned, and with a few minor rule/organizational changes, we could possibly beat him, or at least draw.
In fact, this game should have been a draw - it was only due to technical problems that we lost.
This was much more instructional than his match against deep blue, because the slow speed of play helped to give us time to study chess theory in depth.
I for one learned a thing or two...
Thanks to Irina Krush. She will take away Garry's title someday. We need to commend her on her efforts she put a ton of work into the game! She contributed months and months of painstaking, in-depth analysis. Thanks, Irina!
<SIG>
I think I lost my work ethic while surfing the web. If you find it, please email it to crispy@crotch.caltech.edu.
</SIG>
My sig has a broken link in it.
how about kasparov vs. every computer in the world? We can do a seti@home type of thing where computers user their idle time to come up w/ checkmate. That would have been more interesting
The game can be seen here.
I find it strangely telling that when IBM set out to defeat Kasparov, they did so by building the fastest computer that ever played chess, and succeeded, while Micrsoft set out to do so with a lot of hype, media attention, and flashy web pages, and lost...
-
After The World resigned, Kasparov announced a 25-move forced mate after move 62 (read: the world would've lost on or before move 87, no matter what it did). You can see the page here, and it has a bit of commentary with Kasparov's analysis.