Go (game)

Go is a very old Asian board game, perhaps most notable for having very simple rules produce incredible depth of gameplay — making it a prime example of “emergent gameplay”. It is viewed by many as being “more complex/interesting” than chess, and some evidence of this is how it took two decades longer for computers to beat the top human go players compared to top chess players.

Perhaps my favorite thing about go is how it has a surpringly good handicap system. As counterpoint, chess pretty much stops being chess if you play with handicap that one player has less pieces, but in go the equivalent of having the weaker player start with extra moves doesn’t change the nature of the game that much. It allows players with large skill gaps have interesting games against each other which very often results in some kind of teaching from the better player. Even very new players of go quickly get to a point where they can teach weaker players a lot.

Personally I played go for many years, mainly online, roughly reaching the rank of 1dan (on KGS). That’s not really impressive, some new very dedicated players can accomplish the same in about a year.

For players of go and those interested, I recommend the documentary The Surrounding GameThe Surrounding Game.

For those into anime, I recommend Hikaru no Go.

https://en.yna.co.kr/view/AEN20191127004800315

South Korean Go master Lee Se-dol, who retired from professional Go competition last week […], said his retirement was primarily motivated by the invincibility of AI Go programs. “With the debut of AI in Go games, I’ve realized that I’m not at the top even if I become the number one through frantic efforts,” said Lee. “Even if I become the number one, there is an entity that cannot be defeated,” he said in an interview with Yonhap News Agency in Seoul on Monday.

Lee didn’t deny that his retirement decision was also influenced by a conflict with the KBA over the use of membership fees

Ironically, Lee plans to face off against another AI Go player next month to commemorate his retirement.

Japanese joseki both players have good shape and are satisfied. Korean joseki both players have bad shape and are unhappy.

Would be funny if komi did not include the half point and instead in the even of a tie the winner would be determined by some kind of tie breaker mini game.

If we had an order of magnitude (or two) more computing power, it might be feasible to use the AlphaGo Zero approach to discover new board games as exciting as go. Just make the machine play itself millions of times and then look at the games after that if they look pleasantly emergent. No need to spend quite as much time as Zero did to reach pro level play in go. Of course the bigger problem is to convince anyone to actually play these new board games, but I think having the backing of the computer played games is way better than verbally describing you awesome new go variant. So I think it might work out if a new game even more interesting than go was found.

However I wonder a bit if go already somehow saturates the human capacity to appreciate emergence so that finding a meaningfully more interesting game might be outright impossible. People would continue to prefer go because of the existing player base (obviously chess still remains very appreciated). On the other hand, such a hobbyist would not initially need other humans as much because they have the AI to play with.

There’s already a lot of very interesting looking go variants people have devised, but with mere human power we won’t be able to explore their true depth.