[concurrency-interest] CSP, the pi-calculus and CPA-2005

Miles Sabin miles at milessabin.com
Tue Aug 30 16:09:21 EDT 2005


P.H.Welch wrote,
> This is a small chip in.  For a concurrency interest list, it is
> surprising to see no discussion based on analysing what you are doing
> and what you are wanting to do using the viewpoint and mathematics of
> process algebra - in particular, Hoare's CSP and Milner's
> pi-calculus.  These have lots to offer about practical issues (and,
> certainly, Java ones) and are starting to get together - a powerful
> combination.

I think you'll find that many of the people on this list have an 
interest in process algebras in one way or another. But I'm not quite 
so surprised that they don't get discussed here ... the process algebra 
approach to concurrency is very different from the threads, locks and 
shared state approach that's hard-wired into Java and reflected in JSR 
166.

FWIW, JSR 121, the Java Isolation API, has a CSP/pi-ish feel to it: 
communication is based on message passing rather than shared state, 
communication channels can be passed across communication channels for 
scope extrusion. Pete Soper's collected various things you might find 
interesting here,

  http://www.bitser.net/isolate-interest/

Cheers,


Miles


More information about the Concurrency-interest mailing list