Can Programming Be Liberated from the von Neumann Style? Alive
The FORTRAN creator's Turing Award lecture attacking imperative programming — 'the von Neumann bottleneck' — and proposing functional programs as algebra. FP's foundational manifesto.
The man who led FORTRAN used his 1977 Turing Award lecture to argue conventional languages were 'becoming ever more enormous, but not stronger' — bound to the von Neumann model of word-at-a-time memory traffic. His FP language proposed programs built by combining functions, manipulated as algebra.
FP itself failed to take hold; the argument did not. The lecture legitimized functional programming as a research direction rather than a Lisp curiosity, and 'von Neumann bottleneck' entered the vocabulary. Backus spent the rest of his career on the FL project pursuing it.