Dug up that old 1992 Dale Miller π-calculus paper that Frank alluded to in a conversation I had with him last April. So Dale shows how you can encode the π-calculus by adding a little predicate for channel messages and then axiomatizing how they communicate: you know, a read plus a write, leads to communication. The thing I arrived at by trying to reinvent this from scratch though doesn't involve any extra axioms --- the communication of reads and writes is just baked into what it means to be a read and write sitting in the same context. This is only a minor tweak, but it really depends on focusing to work. I wonder if it's just as old hat?