The Praxis of Software Engineering

Martyn Thomas

Thirty years ago, in October 1969, 66 of the world's leading software academics and practitioners met in Rome for a working conference on software engineering techniques, sponsored by the NATO Science Committee. The conference was intended as a direct sequel to the Garmich conference, one year before, where it was first widely acknowledged that there was a "software crisis", because the complexity and scale of software applications had outstripped the engineering disciplines that had been developed to build them.

On the last day of the Rome conference, Christopher Strachey spoke of the relationship between theory and practice: "one of the unspoken underlying themes of this meeting". He reported a view he had heard the previous night, that the people who were actually writing very large programs "were invited here like a lot of monkeys to be looked at by the theoreticians", and he "sought to make a little bridge" between the (theoretical) computer scientists and the (practical) software engineers.

Strachey's subsequent comments were generous, perceptive, and seem as relevant today as thirty years ago. Yet there has been progress, and not just that there are now professors of software engineering as well as professors of computer science!

This lecture provides a partial view of what progress has been made, and attempts to answer Strachey's question about the relationship between theory and practice.