Reversi (Othello)

Reversi is a two-player strategy board game played on an 8x8 grid with discs that are black on one side and white on the other. Players take turns placing discs with their color facing up. When a player places a disc that brackets one or more opponent discs in a straight line (horizontally, vertically, or diagonally), those bracketed discs are flipped to the current player's color. The game ends when neither player can make a legal move, and the player with the most discs of their color on the board wins.

Rules

1. The game starts with four discs placed in the center of the 8x8 board: two black and two white, arranged diagonally.
2. Black moves first.
3. A player must place a disc adjacent to an opponent's disc such that it brackets (sandwiches) one or more of the opponent's discs between the new disc and another disc of the player's own color, in a straight line.
4. All bracketed opponent discs are flipped to the current player's color. Multiple lines can be bracketed in a single move.
5. If a player has no valid moves, their turn is skipped.
6. The game ends when neither player can move (typically when the board is full).
7. The player with the most discs of their color wins. Equal counts result in a draw.

Strategies

• Corner Control: Corners are the most valuable cells because once captured, they can never be flipped. A corner gives permanent control of its adjacent edges and diagonals.
• Avoid X-Squares and C-Squares: The cells diagonally adjacent to corners (X-squares) and directly adjacent to corners along edges (C-squares) are dangerous. Playing there often gives your opponent access to the corner.
• Edge Strategy: Edge cells (excluding corners) are semi-stable — they can only be flipped along the edge line. Building a solid edge presence provides stability.
• Mobility: Having more legal moves than your opponent is a major advantage. Players often aim to restrict the opponent's options while keeping their own open.
• Disc Minimization (Early Game): Counter-intuitively, having fewer discs in the early game is often advantageous. Fewer discs means a smaller perimeter, which means fewer cells where the opponent can play, restricting their mobility.
• Stable Discs: A stable disc is one that can never be flipped for the rest of the game. Building clusters of stable discs, especially spreading from corners, is the foundation of a winning position.

The game of Reversi has a disputed origin story. It was most likely invented in 1883 by either Lewis Waterman or John W. Mollett, both of London. Each claimed to have created the game, and their competing claims led to a public dispute in the pages of The Saturday Review and other periodicals. Waterman published the game under the name "Reversi" and sold it through Jacques & Son, a well-known games manufacturer. Mollett claimed an earlier invention called "The Game of Annexation."

Regardless of who invented it first, the game was commercially produced in England in the 1880s and enjoyed modest popularity. It spread through Europe and was known in various forms throughout the early 20th century, but never achieved widespread fame.

The game was reinvented and rebranded in 1971 by Goro Hasegawa, a Japanese salesman, who registered it under the trademark "Othello." Hasegawa's version used the modern standard starting position and was marketed with the memorable slogan "A Minute to Learn, A Lifetime to Master." The game was first manufactured by Tsukuda Original in Japan and became enormously popular. The World Othello Championship has been held annually since 1977, making it one of the longest-running abstract game competitions.

Reversi/Othello became an important subject in artificial intelligence research. In 1997, the program Logistello, developed by Michael Buro, decisively defeated the reigning world champion Takeshi Murakami 6-0, marking one of the early instances of a computer program dominating a human champion in a strategy game. Today, computer programs play Reversi at a superhuman level, and the game is widely used in AI education for teaching evaluation functions, alpha-beta search, and machine learning approaches.