In the late '80s, Bob Griffiths and Roland Omnes, independently, worked on decoherent histories and they published the idea before Jim and I did. In fact I'd never written anything about quantum mechanics until the very end of the 1980s. But I'd thought a lot about it, and the thoughts, as I say, were sort of continuous. Jim Hartle and I have worked on decoherent histories with a particular point of view. Some people have tried to get the minimum conditions for consistency allowing probabilities to be assigned to course-grained histories, so they get a weaker and weaker form of decoherence. We have been interested in finding a stronger and stronger form of decoherence, trying to appreciate what actually happens in the real world where decoherence is far from minimal. The actual decoherent histories with which one conventionally deals are decoherent in a very strong manner, and it's that strong decoherence that we've tried to describe, approach and so on. But now if you consider a set of decoherent, course-grained, alternative histories of the universe, given the two fundamental principles; the unified theory of all the particles and forces, and the initial condition, initial density matrix or a wave function, then… the… let me see, what was I going to say… if… if you're given all that, there's a question of… of which set of decoherent histories you use. And we like to talk about a realm, which is a set of alternative decoherent histories subject to some condition. Usually it's a kind of maximality condition so that you… if you were to fine-grain further you would start to lose the… the decoherence, something like that; or if you fine-grained further you would start to lose either the decoherence or some other desirable property, and such a system we call a realm.