What is a "good" encoding of guarded choice?

The pi-calculus with synchronous output and mixed-guarded choices is strict ly more expressive than the pi-calculus with asynchronous output and no cho ice. This result was recently proved by C. Palamidessi and, as a corollary, she showed that there is no fully compositional encoding from the former i nto the latter that preserves divergence-freedom and symmetries. This paper argues that there are nevertheless "good" encodings between these calculi. In detail, we present a series of encodings for languages with (1) input-g uarded choice, (2) both input and output-guarded choice, and (3) mixed-guar ded choice, and investigate them with respect to compositionality and diver gence-freedom. The first and second encoding satisfy all of the above crite ria, but various "good" candidates for the third encoding-inspired by an ex isting distributed implementation-invalidate one or the other criterion, Wh ile essentially confirming Palamidessi's result, our study suggests that th e combination of strong compositionality and divergence-freedom is too stro ng for more practical purposes. (C) 2000 Academic Press.