C. A. R. Hoare. As with most of his work, he gets some things right but goes astray when he overstates the value of his favorite theories. Of the 5 reasons he cites: Management, Testing, Debugging, Over-engineering, and Programming Methodology, nothing he says about management has stood the test of time. What he calls "Over-engineering" is describing ways to avoid complexity, "avoid any kind of sophistication or optimisation in the design of algorithms or data structures". His section on Programming Methodology just lists out the advancements in language design.
To the surprise of no one, his conclusions still push for formal proofs and methods, but laments that a gap of twenty years between theory and practice is the reason why they haven't been adopted widely.
To the surprise of no one, his conclusions still push for formal proofs and methods, but laments that a gap of twenty years between theory and practice is the reason why they haven't been adopted widely.