About

Game of Life is a cellular automaton devised by the British Mathematician John Horton Conway in 1970. [Wikipedia link]

Here the world consists of a grid of cells, with some of them being alive. A dead cell comes alive, or is Born, if it has exactly 3 immediate live neighbours. Similarly, a live cell survives to the next generation, or Stay alive, if it has 2 to 3 neighbours; It will die if the neighbour count in any lesser or greater. This is abbreviated as the B3/S23 rule.

The game starts with an initial configuration of live cells, and evolves into different patterns in subsequent generations according to these simple rules. Despite the short rules, the evolution of an initial pattern can be surprising.
There are variations of Game of Life, created by changes in the B/S rule. Seemingly small changes in the rule can give rise to an evolution pattern that is drastically different. There is the charmingly structured Maze and the explosive Seeds. For examples, check out other game rules in the above interface.

Life forms

There are several naturally occuring and cleverly engineered patterns disovered in different cellular automata rules. Static life forms which do not move, yet do not die out, life forms which are static yet pulse in their current position with various intervals, and Spaceships which are patterns which keep moving in one direction, not unlike what is imagined of spaceships.
Check some examples in the above interface. Note that, a pattern shows its guaranteed behaviour only in its native set of rules.

Implementation Details

In the original definition, a cellular automaton consists of an infinite grid of cells, growing in all directions. Above implementation uses a toroidal grid, meaning the top and bottom are virtually stitched together, and similarly the left and right sides. So a cell which moves past the right end, continues from the beginning of left end. This enables an infinite world of sorts, except that moving cells can collide with existing cells unlike an infinite universe.

The implementation also uses a naive implementation of rules. More optimized algorithms for Game of Life include Listlife and Hashlife.
PS: Turn off Trace (colored trail) in the interface if a crowded grid is making the simulation slower.