dwink i owned a company and all i got was this stupid blog

7Sep/090

Intuitive Logic

I've been playing Go on and off for several years now, never really making it past the 'novice' phase. The more I study and practice, however, the more fascinated I'm getting; there's a reason it's one of the oldest games still played in its original form. Or, forms, really, as it was an ancient games before it became standardized.

The thing about Go is that the simple pieces ( one type per player, a stone ) and play structure ( a turn consists of placing a stone or passing ) can express a vast array of ideas and strategies. When I say vast, I mean it --the Go board allows for so many possible board positions that despite a thousand-year history it is believed that no two Go games have been alike. And on top of that, a good portion of the skill in Go involves internalizing abstract shapes, letting the subconscious mind help you to analyze the board quickly through shape & pattern interpretation.

Add to this the fact that because Go is a territory-capture game, and stones are only removed from the board when completely surrounded, and you have a subtle conversation between players: which part of the board will I take, which will you? How much of the board can I really defend? The answer is guaranteed to be less than "all of it", so unlike the winner-take-all approach of games like Chess or Checkers, the players must compromise, and a highly aggressive strategy is extremely unlikely to succeed.

The upshot of all this is that Go is among a class of games that traditional computer science has had a terrible time capturing. The sheer number of possible moves, the intuitive logic, plus the subtle conversation between players, give a nuance to the standard game tree and make figuring out the 'best' move among thousands difficult. Early moves in a Go game are more symbolic gestures than true commitments to a position -- at any time, either player may abandon one piece to focus elsewhere; there are likely to be two, three, even ten 'battles' for influence going on at once, and focusing on only one will almost surely fail you.

I think that the time I spend on Go is literally teaching my brain how to think intuitively about concrete problems. I'm sure someone will or has studied this, but I think that with every step I take toward mastering another Go situation, I am equipping my brain to subconsciously attack strategic decisions, both as a leader and as a coder. It's like a trip to the gym for my brain.

Now I just need to take a trip to the gym for my body. If only there wasn't so much working out involved...