Page d'accueil // SnT // Distinguishe... // Faultless Systems: Yes we Can - July 1, 2010

Faultless Systems: Yes we Can - July 1, 2010

It is our pleasure to host this distinguished lecture by  J-R Abrial, Professor at ETH Zurich in Switzerland. The lecture will be followed by a reception. Please feel free to forward this invitation.

Faultless Systems: Yes we Can

Abstract: The title of this text is certainly provocative. Everyone knows that this claim corresponds to something that is impossible. No! One can not construct faultless systems, just have a look around. Should this have been possible, it would have been already done for a long time. And to begin with: what is a ``fault''?

So, how can someone imagine the contrary? You might think: yet another guru trying to sell us his latest Universal Panacea. Be reassured, this talk does not contain any new bright solutions and moreover it is not technical: you'll have no complicated concepts to swallow. My intention is just to remind you of a few simple facts and ideas that you might use if you wish to do so.
My intention is to play the role of those who are faced with a terrible situation (yes, the situation of computerized system developments is not far from being terrible: as a measurement, just consider the money thrown out of the window with systems that fail). Faced with a terrible situation, one might decide to change things in a brutal way: it never works. Another approach is to gradually introduce some simple features which together will eventually result in a global improvement of the situation. The latter is my philosophy.
Prof. J.-R. Abrial is the father of the Z notation (typically used for formal specification of software), during his time at the Programming Research Group within the Oxford University Computing Laboratory , and later the B-Method (normally used for software development), two leading formal methods for software engineering. More recently he has developed the Event-B framework. He is the author of The B-Book: Assigning Programs to Meaning . Until recently he was Professor at ETH Zurich in Switzerland. For much of his career he has been an independent consultant, as much at home working with industry as academia.