Okay, the goal was simple to state while not actually being simple at all. The challenge at the heart of that challenge is that you can't create a game that requires you to buy it; you must create a game that will spread organically from player to player, while allowing for each new convert to spread the game easily as well. This means that you can't think in terms of manufacturing or licensing because those elements inhibit the ability of a game to travel easily to new players. You want the idea of the game to spread, so the specific physicality of the bits themselves must be secondary.
While not designed specifically for this competition, designer James Ernest has won it retroactively with Tak, a game reluctantly co-designed with novelist Patrick Rothfuss — and I say "reluctantly" because initially Rothfuss didn't want Tak to become a reality. He had introduced the game in his book The Wise Man's Fear, presenting it as an ancient game in his fictional world but not detailing how to play it. (Ernest also won TTYGDC challenge itself with Take-Back-Toe.)
Ernest, a fan of Rothfuss' work, worked with him to create three Pairs decks featuring characters and the world featured in The Name of the Wind, the first book in Rothfuss' The Kingkiller Chronicle series. While preparing for the Kickstarter for Pairs, Ernest asked Rothfuss about making a real game of Tak and Rothfuss refused many times, finally agreeing to talk about the game's background just to get Ernest to stop bothering him. Then as Rothfuss tells it:
Again, I said no.
"Why not?" he asked.
"Tak is supposed to be my world's version of Chess or Go or Mancala," I said. "I can't ask you to make a game like that. It's like saying, 'you know those games that have stood the test of time for hundreds or thousands of years? The best games ever? Do that, but in my world.' So first off, it's unreasonable for me to ask. Secondly, you can't do it. No one can. And thirdly, if you did somehow manage to pull if off, nobody would give a shit. We're living in the golden age of board games right now. Nobody cares about strategy games like chess anymore." ...
"Just let me try," James said. "Let me take a run at it. If you hate what I come up with, we'll never speak of it again."
So I told him, fine. Fine! Do it. Whatever. Jeez.
• Play on a square grid of any size from 3x3 to 8x8, with 5x5 and 6x6 being ideal. Use a number of pieces appropriate for the size of the board. Include a larger capstone piece for games on a board at least 5x5.
• To win, create a line of your pieces that connects opposite sides of the board.
• On a turn, either place one of your pieces on an empty space or move a stack of pieces that is topped with one of your pieces.
• When you place a regular piece, you can place it flat (allowing it to move or be covered later) or stand it on end to serve as a wall that cannot be covered and doesn't count as part of a winning line. Capstones cannot be placed on end.
• When you move, move the stack orthogonally, leaving behind zero or more pieces on the starting space and dropping one or more pieces on each space visited. You cannot enter a space with a wall — unless the top piece on the stack is a capstone; in this case, you can end movement by using the capstone (and only the capstone) to flatten the wall, turning it into a regular flat piece.
• If you create a line of pieces (flat ones and capstone) that connects opposite sides of the board, you win. If all the pieces have been placed, whoever controls more spaces with flat pieces wins.
Cheapass Games sent me a rough version of the game with beta rules (PDF), and I've played seven times over a single lunch: three times on the 4x4 board and four times on the 5x5 board.
Tak is one of those perfect strategy games in which it's easy to lose (or win) in your first games because you (and your opponent) have no idea what you're doing. You overlook the obvious things that an opponent can do to win out of the blue because you don't even know where to look or what to look for. The rules are simple, yes, but you need to internalize them in order to start defending against attacks and the only way to do that is to play the game and learn from your mistakes.
One final aspect to the rules: On the first two turns of the game, you place one of the opponent's pieces instead of your own. This functions as something akin to the pie rule in Hex that allows the second player to take control of the stone placed by the first player, but is more interesting since (as the second player) you're reacting to the first player while also influencing that player's next move, which will be with the color that you've just placed.
Once the pieces start piling up on the board, you have a lot to consider each turn. The threats are everywhere, and you start working through your head how the mutually assured destruction would unfold if one of you makes the move the other keeps worrying will be made. It's easy to forget about the walls because in some senses that piece is wasted, but at the same time, the wall can neutralize an opponent's attack and protect multiple spaces on the board from one or more threats.
Once you graduate to the 5x5 board and the inclusion of the capstone — the full game, as it were — you discover another level of play, with the capstone serving as the game's queen in how it can warp play around it on the board; the capstone is a black hole that influences everything around it; it's a boot on the opponent's neck that cannot be removed but only shifted so that they aren't choking nearly so badly as you might wish.
I have no idea how many more games can be created that might meet the thousand-year threshold, but it can be done — at least as I view the situation only one year after Tak was created — and I'm curious to see how long the game will survive once only the cockroaches remain behind to play.