Advanced Chess Supercomputer?

In a 2006 paper, Professor Yochai Benkler wrote “a novel system of production … has produced … the fastest supercomputer.” The novel system of production to which Benkler was referring is commons-based peer production and the supercomputer that Benkler had in mind is the SETI@home project. Can the Open Encyclopedia of Chess Openings effectively harness the power of commons-based peer production to create the equivalent of an “advanced chess” supercomputer?

Commons-based peer production (CBPP) is “a socio-economic system of production that is emerging in the digitally networked environment. Facilitated by the technical infrastructure of the Internet, the hallmark of this socio-technical system is collaboration among large groups of individuals, sometimes in the order of tens or even hundreds of thousands, who cooperate effectively to provide information, knowledge or cultural goods without relying on either market pricing or managerial hierarchies to coordinate their common enterprise [emphasis mine]” (Benkler 2006).

The SETI@home project is an example of both commons-based peer production and distributed computing. This project “is a scientific experiment that uses Internet-connected computers in a Search for Extraterrestrial Intelligence (SETI). The data sets collected from large radio telescope observations are immense. The project was organized to harness the computer processing cycles of millions of volunteers with computers connected to the Internet to process these vast data sets. Participants download a small free program that functions as a screen saver when they are not using their computers. At that point, it downloads and analyzes radio telescope data. According to statistics maintained on the SETI@home website, as of August, 2003, the project had absorbed over 4.5 million users from 226 countries, and provided an average computation speed almost twice that of the fastest “supercomputer” then in operation in the world” (Benkler 2006).

While the SETI@home project requires relatively little effort from the end-user besides downloading and installing the BOINC program, other CBPP projects involve a greater amount of active participation by project participants. Take, for example, the NASA Clickworkers experiment. “In this project, tens of thousands of individual volunteers collaborated in five-minute increments to map and classify Mars’s craters, performing tasks that would normally require full-time PhDs working for months on end, freeing those scientists for more analytic tasks” (Benkler 2006).

The Open Encyclopedia of Chess Openings provides yet another means by which individuals can direct their otherwise underutilized energies and the unused computing power of their computers towards productive ends that produce value for other members of society. As Benkler wrote in a different paper, “peer production draws effort that in many cases would otherwise have been used in purely non-productive consumption—say, watching television instead of marking craters on Mars, ranking websites for the Open Directory Project, or authoring entries for Wikipedia. On a macro level of social productivity, then, an economic system that incorporates peer production as one component in its production system will add a vehicle for tapping effort pools that would otherwise not be used productively at all” (Benkler 2002).

Everyone is welcome to join the Open Encyclopedia of Chess Openings and to participate alongside other human+computer teams as we analyze and map the topical variations of opening variations. Come be a part of the world’s first “advanced chess” supercomputer. 

Benkler, Yochai. (2002), Coase's Penguin, or, Linux and the Nature of the Firm. The Yale Law Journal 112(3)

Benkler, Y. and Nissenbaum, H. (2006), Commons-based Peer Production and Virtue. Journal of Political Philosophy, 14: 394–419. doi: 10.1111/j.1467-9760.2006.00235.x

Towards Perfecting Advanced Chess

A relatively new mode of social production (i.e. commons-based peer production) is now being used to generate chess opening theory and to extend opening theory deep into the middle game. The result may be a more advanced form of advanced chess.

Advanced chess is a form of competitive chess, popularized by Garry Kasparov, in which at least one human pairs together with at least one computer chess program to form a team. Teams of human + computer then compete against other human + computer teams. Ideally, advanced chess combines the strongest attributes of human chess players with the strongest attributes of computer chess programs. The resulting quality of chess produced by a human/computer advanced chess team is often considered to be superior to that which could have been produced by either a lone human grandmaster or a lone computer program.

While computer chess programs are generally recognized to excel in positions requiring tactical precision, computer programs nevertheless suffer the handicap of having blindness beyond their move horizon. Humans, on the other hand, frequently display superiority in judging positions where strategic insight is crucial. However, humans are demonstrably inferior to computer programs in most tactically complex situations.

The combination of human + computer program produces a chess playing entity which is able to expertly sail through the sea of tactical complications while also anticipating the ocean beyond the computer program’s move horizon.

So, what could possibly produce a higher quality of chess than that produced by a team of humans paired together with the strongest chess programs? The answer may lie simply in the creation of an advanced chess peer-review process.

This advanced chess peer-review process is the foundation upon which the Open Encyclopedia of Chess Openings (OECO) is built. Advanced chess teams (human + computer) analyze chess positions and add the resulting analysis to the OECO wiki. This analysis is public and subject to review by other teams of humans plus computers.

In this way, the OECO brings together a community of human minds, each with its own unique chess perspective, and a variety of computer programs, each with their own strengths and weaknesses, in order to continually and relentlessly iron out all of the wrinkles (i.e. mistakes) of the collective body of existing analysis. Computer analysis guided by human judgment is followed by counter-analysis and counter-judgment in a dialectical process which may ultimately converge upon chess perfection.

Wiki Accounts

The Open Encyclopedia of Chess Openings (OECO) wiki now provides chess players the ability to create their own OECO wiki account. (OECO accounts enable the account holder to edit the wiki).

Only two pieces of information are required in order to create a OECO
wiki account:
1) Username
2) Password
Other fields are optional.

The previous method required prospective members to first email
gary@finaltheoryofchess.com. Accounts were then manually created based
upon these email requests.

We hope that the new system streamlines the process of new account
creation and encourages greater numbers of chess players to become
active contributors to the wiki.

Everyone is welcome to bring their own unique contribution to the Open Encyclopedia of Chess Openings.

Campaign for Chess Opening Diversity

Does the Open Encyclopedia of Chess Openings not have coverage of the chess opening that you are looking for?

Join today and help us with our new Campaign for Chess Opening Diversity!

Please help us as we launch a new campaign to expand the number of openings which are covered within the Open Encyclopedia of Chess Openings (OECO). Greater opening diversity, we hope, will increase the value of the OECO for the chess community at large.

Currently the OECO only addresses a relatively narrow set of chess openings albeit in considerable detail. In the coming months, we hope to expand that number to include most commonly played openings.

We encourage all current members of the OECO community as well as any chess player who is thinking about joining to consider adding a few wiki pages about an opening which has yet to be given any coverage.

Membership is free. 

The triumph of the analytical movement, which formed in the '30's and '40's, was precisely what earned the Soviet masters the acclaim of chessplayers the world over. Unfortunately, it must also be noted that, for today's chessmasters, the watchword is practicality. - Mikhail Moiseyevich Botvinnik

History of Humans vs. Computers

Deep Blue, Chinook, Quackle, Polaris, Huygens, and BKG 9.8 are the names of computer programs which have emerged victorious against top human competitors in games of strategy. Each of these programs were also recently referred to in a CNN article about the upcoming match between human Jeopardy champions and the computer Jeopardy program called Watson.

Deep Blue is IBM's computer chess program which famously defeated the world chess champion, Garry Kasparov, in a 1997 match. This defeat of the reigning world chess champion is not without controversy, however.

Chinook, written by Jonathan Schaeffer and his team at the University of Alberta, defeated long-time world checkers champion, Marion Tinsley. Visit the Chinook project website and play the program which made history. I also recommend Schaeffer's book, One Jump Ahead - Computer Perfection at Checkers, for those interested in the story surrounding the creation of the unbeatable program which always plays perfect checkers.

BKG 9.8 was a Backgammon program which, in 1979, defeated the reigning world backgammon champion, Luigi Villa. BKG 9.8's defeat over Luigi Villa is considered to be the first time that a sitting human world champion of a board game was defeated by a software program. Villa, however, was only recently crowned world backgammon champion when he was defeated by BKG 9.8 and Villa was handicapped with inferior dice rolls. Dr. Hans Berliner, the author of BKG 9.8, is also known for his popularization of the Option Principle which has been a topic of discussion at this blog and within the Open Encyclopedia of Chess Openings wiki.

Option Principle

A worthwhile yet neglected contribution to chess literature is the Option Principle. The Option Principle was first enunciated by Weaver Adams. In his book, White to Play and Win (1939), Adams wrote:

A third characteristic of a good move has to do with the conservation of valuable options, as follows: If a piece appear to have two or more equally attractive moves in a given position, this usually may be taken as a sign that a move of this particular piece is incorrect at the moment, or, in other words, that the best square for this piece is not as yet determinable. Necessarily, however, the best square for the correct piece to move to is determinable (otherwise the move would be incorrect). And because of this fact, the correct piece to move can usually be seen to have but one attractive square to move to. … "In practice it is easy to overlook an optional square for a piece, especially if this is one which at the moment is impossible due to obstruction, or undesirable because attacked by enemy pieces. Any one of these squares represents an important and valuable option in proportion as it can be foreseen that the game might take a course which would render these squares available and/or desirable at some future time (p.15).  

Sixty years later the Option Principle was reintroduced into chess literature by Dr. Hans Berliner. In 1999, Dr. Berliner published his book, The System, which incorporated the Option Principle as a central pillar of his system. According to Berliner:

The Option Principle states: make the move (develop the piece) which does the least to reduce your options to make other important moves. When there are several pieces that can be developed, move the one for which the Optimal Placement is most clear. This is a generalization of Lasker’s rule ‘knights before bishops’. Usually a bishop has more good locations to choose from than a knight, so develop the knight first. However, the System options principle is much more general. It frequently encourages the non-movement of a piece that is already well placed on the back rank. It may also discourage castling, if the rook is well placed for an attack. … The Option Principle also prohibits making a move that blocks a friendly piece from reaching its optimal location. … there is always at least one move that is crying out to be played before other moves. (p.36-37).

In 2008, the Option Principle again found its way into print with the publication of The Final Theory of Chess and in 2009 it found a home on the World Wide Web with the Final Theory of Chess Open Encyclopedia of Chess Openings.

In addition to the three aforementioned books, the Option Principle is hinted at with varying degrees of vagueness by several other chess writers.

• Philidor warned against obstructing one’s own pawns. He advocated 2…d6 (after 1.e4 e5 2.Nf3) in order to avoid losing the option of advancing Black’s c-pawn.

• The second World Champion, Emanual Lasker, applied the Option Principle to the range of plans in which multiple pieces could cooperate. He wrote: “or, to use another term, say flexibility, or adaptability or elasticity. The main idea of this co-operation is to increase the range of possible plans to follow, without specifying too early which road you would prefer to travel. By co-operation you aim to keep the position plastic, alive; by lack of co-operation you take the life out of your position, and to infuse it with new life you will need outside aid” (Lasker, Emanual. Lasker's Manual of Chess, p.231).

• Nimzowitsch emphasized the reduction in options entailed when the central pawns are advanced early. In his notes to the game Nimzowitsch-Rubinstein (Berliner Tageblatt Tournament, Berlin 1928) he wrote of 1.Nf3: “Certainly the most solid move, whereas moves such as 1.e4 and 1.d4 are both ‘committal’ and ‘compromising’.”

• A final example comes from the classic book, The Art of Attack, by Vladimir Vuković. Parsimoniously, Vukovic states the Option Principle with a single phrase:  “moves entailing fewer obligations should be carried out before those which are more strongly binding” (Vuković, Vladimir, The Art of Attack, p.12).

Please feel free to comment on this post to share further examples (either implicit or explicit) of the Option Principle appearing in chess literature.

Epistemology of the Chess Advantage

What does it mean when we speak of an "advantage" in the game of chess? Furthermore, what does it mean when a computer program evaluates a position at, for example, -0.38?

The chess analyst must always look upon computer analysis with suspicion because the computer evaluation of a position is most likely incorrect. The measure of "advantage" that you and I deal with regularly (i.e. centipawns) sits upon a questionable epistemological foundation. In any given position, there are only three “true” evaluations. These are:  1) 0.00 2) infinity 3) - infinity. Stated differently, assuming perfect play, a position must be a draw, a win for White, or a win for Black. Therefore, any evaluation which is not 0.00 or an announced mate in # is by definition wrong.

To illustrate this, take an endgame position with only a Black king, a White king and a White bishop. White cannot be said to have a 300 centipawn advantage. In such a situation, being “up a bishop” is meaningless. Since the computer recognizes such a position to be a draw, it will give us the correct evaluation of 0.00. In other less recognizable but perfectly drawn positions, the computer will insist that one side or another has an advantage of (insert number) of centipawns. The same is often true in positions where a forced win can, after more exhaustive analysis, be demonstrated.

If there was no move horizon beyond which the computer can not see then the computer would be able to see all possible forced results in any position (win, loss, or draw). If there was no move horizon then the only evaluations of any given position would be positive infinity, 0.00, negative infinity. Any evaluation which is not one of these three values is by definition incorrect. It is a product of the chess engine programming which is designed to compensate for imperfect information (i.e. move horizon).

Arithmetic evaluations of chess positions are useful but ultimately fictitious. 

This brings up the notion of ‘arithmeticism’ in chess. Especially with the appearance of chess-playing computers which update a numerical assessment of the position on every half-move, there are players who tend to think in terms of arithmetic advantages, e.g., ‘White is better by 0.33 pawns’. This has its uses, but can lead to a rather artificial view of the game. What happens when both sides make a few moves which are the best ones, and suddenly the 0.33 pawns is down to 0.00, or full equality? The defender of this point of view will say: ‘Well, I didn’t see far enough ahead. If I had, I would have accurately assessed the original position as 0.00. The only problem with this point of view is that chess is a draw, and all kinds of clear advantages (in the sense of having a good probability of winning a position in a practical game) are insufficient to force a win against perfect defense. So most positions would be assessed as 0.00, which is not very helpful. In the extreme, we have the same problem when we claim, for example, that 1.Nf3 is ‘better’ than 1.e4, or 1.d4 is better than 1.c4. These are rather meaningless statements, unless we put them in the context of ‘better against opponent X’ or ‘better from the standpoint of achieving good results with the least study’ or some such. As for the objective claim of superiority, what would be our criterion? I would suggest that only if a given first move consistently performs better than others against all levels of competition might we designate it as ‘better’ in a practical sense. Since all reasonable first moves lead to a draw with perfect play, a claim of ultimate theoretical superiority for one of them cannot be justified.

Watson, John. Secrets of modern chess strategy: advances since Nimzowitsch. Gambit Publications, 1999. 232-33.