Blueprint for Real Progress in Software Engineering

While there's general agreement that the software engineering field is in great need of major advances, efforts are dispersed and not always directed to the really critical needs. I will present a view of what is fundamentally needed to produce real progress. The view is influenced by my own concerns and the talk is also an opportunity to present the research of my groups at ETH and Eiffel Software, with a special emphasis on: providing a basis for Trusted Components; certification of commercial components; realistic proofs of classes; advanced object-oriented techniques, making concurrent programming as simple as it should be; providing better tools; and management issues. Rather than emphasizing the fundable and fashionable, I will attempt to highlight topics that are truly important for our constituencies.