Describing changes caused by the execution of actions is a classical topic of research in logic-based artificial intelligence. A large body of work has been devoted to the frame problem--the problem of describing what does not change when actions are performed. This work has led to the emergence of several formal theories of nonmonotonic reasoning and to the invention of action description languages, such as STRIPS and ADL. We are interested in action languages based on nonmonotonic causal logic, in their relation to logic programming, and in their applications to the automation of reasoning about actions.