> think about a real time system that can know when it has spent too much time planning, and needs to use whatever its current best plan is That's why iterativ

> think about a real time system that can know when it has spent too much time planning, and needs to use whatever its current best plan is That's why iteratively deepened depth first search kicks ass in board game AIs (specifically chess).