At a technical level, the Bitcoin protocol is a clever solution to the consensus problem in computer science. The idea of consensus is very general — a number of participants together execute a computation to come to agreement about the state of the world, or a subset of it that they’re interested in.
Because of this generality, there are different methods for analyzing and proving things about such consensus protocols, coming from different areas of applied math and computer science. These methods use different languages and terminology and embody different assumptions and views. As a result, they’re not always consistent with each other. This is a recipe for confusion; often people disagree because they’ve implicitly assumed one world-view or another. In this post I’ll explain the two main sets of models that are used to analyze the security of consensus in Bitcoin.
