from connection machines article → here
The game of Life is an example of a class of computations that interested Feynman called [cellular automata]. Like many physicists who had spent their lives going to successively lower and lower levels of atomic detail, Feynman often wondered what was at the bottom. One possible answer was a cellular automaton. The notion is that the "continuum" might, at its lowest levels, be discrete in both space and time, and that the laws of physics might simply be a macro-consequence of the average behavior o f tiny cells. Each cell could be a simple automaton that obeys a small set of rules and communicates only with its nearest neighbors, like the lattice calculation for Q CD. If the universe in fact worked this way, then it presumably would have testable consequences, such as an upper limit on the density of information per cubic meter o f space.
The notion of cellular automata goes back to von Neumann and Ulam, whom Feynman had known at Los Alamos. Richard's recent interest in the subject was motivated by his f riends Ed Fredkin and Stephen Wolfram, both of whom were fascinated by cellular automata models of physics. Feynman was always quick to point out to them that he consid ered their specific models "kooky," but like the Connection Machine, he considered the subject sufficiently crazy to put some energy into.
There are many potential problems with cellular automata as a model of physical space and time; for example, finding a set of rules that obeys special relativity. One o f the simplest problems is just making the physics so that things look the same in every direction. The most obvious pattern of cellular automata, such as a fixed three -dimensional grid, have preferred directions along the axes of the grid. Is it possible to implement even Newtonian physics on a fixed lattice of automata?
Feynman had a proposed solution to the anisotropy problem which he attempted (without success) to work out in detail. His notion was that the underlying automata, rathe r than being connected in a regular lattice like a grid or a pattern of hexagons, might be randomly connected. Waves propagating through this medium would, on the avera ge, propagate at the same rate in every direction.