Theory and Practice of Concurrency

Rating:
Summary: A very important book
Review: The failure by computer science to properly take formal methods seriously is one of the tragedies of the late twentieth century - and it is a tragedy that continues.

Our failure to be able to demonstrate that software has useful properties - for example, that it terminates, or is free of deadlock, or that it meets its specification - has resulted in the loss of NASA spacecraft and worse, it encumbers us all with the unreliable and untrustworthy computing systems we all use day to day.

Professor Hoare's Communicating Sequential Processes - the subject of this book - is to computer systems behavior what Euclid's Elements is to Geometry.

One day software houses will realize the value of building software that has a mathematical foundation for its behavior. Until then the adhoc, quick to market, quick and dirty software we all suffer today will continue to threaten the lives of our heros and ourselves.

Bill Roscoe, though I haven't seen him in a decade, I consider to be an old friend. His commitment and diligence in the just pursuit of these goals is the merit of focused scholarship from which the world as a whole receives the ultimate benefit.

CSP needs to be in the hearts and minds of every computer systems engineer from silicon to application.

In more recent times I have come to believe that CSP has important applications in other disciplines. I would encourage therefore any scientist interested in mathematical systems able to describe or model dynamic behavior to take a close look at this book and CSP. CSP provides new and additional tools that reach beyond its initial application in computer science.

I trust that the volume of sales that this book deserves eventually meets the equivalent of the "New York Times Bestseller" in it's field. This book and Tony Hoare's original book on CSP, deserve to remain in print as long as Hilbert ... and Euclid.