For some reason I'm constantly reminded of the story of the discovery of Karatsuba multiplication. At a seminar Kolmogorov suggests that the classic multiplica

For some reason I'm constantly reminded of the story of the discovery of Karatsuba multiplication. At a seminar Kolmogorov suggests that the classic multiplication algorithm was optimal, to multiply two numbers with n digits was O(n^2). Karatsuba, who is a student attending, come back a week later with an O(n^(log_2 3)) algorithm - which Kolmogorov was so excited about he cancelled the rest of the seminar and wrote up and published the algorithm without even telling Karatsuba (although he did list his as the author) There's a lesson in there somewhere