Why is a Go board 19 lines by 19, when it used to be 17?

Nobody wrote down a reason, but we do know the board grew into that size. Archaeological evidence shows Go was played in earlier centuries on a smaller 17×17 grid. By the time the game reached Korea in the 5th century CE and Japan in the 7th, the 19×19 board had become standard, and it was certainly standard by the Tang dynasty (618–907). Two extra lines each way take the board from 289 points to 361.

That extra room is a big part of what makes Go so deep. Through most of a game, a player has around 150 to 250 legal moves per turn, compared with an average of 37 in chess, and the number of legal board positions has been calculated at about 2.1 × 10^170, far more than the number of atoms in the observable universe. That is why Go held out against computers for so long: a program that tries to compare every legal move loses much of its power when there are that many. A computer first beat a professional in 2008, and only with a nine-stone handicap. Then, in October 2015, Google DeepMind's AlphaGo, which learned mostly by playing itself hundreds of millions of times, beat European champion Fan Hui five games to none on a full 19×19 board, and in March 2016 it beat Lee Sedol, considered the world's top player, four games to one.

The game itself is older than its board. Go was invented in China more than 2,500 years ago and is believed to be the oldest board game continuously played to the present day. A legend says the mythical emperor Yao had his counselor Shun design it to be a good influence on his unruly son, Danzhu. Go later became one of the four arts of the Chinese scholar-gentleman, alongside calligraphy, painting and the guqin, and a 2016 survey counted over 46 million people worldwide who know how to play.