A causal state machine encodes causal entities as types with the actual constraint represented as a causal function. One or more causal entities form a collection that is encapsulated in an abstraction called a causaloid. The causaloid represents the state, which then gets passed into a state machine for evaluation and subsequent action. Unlike a finite state machine, the causal state machine does not have to know all its causal entities upfront which allows for greater flexibility.
