I once had the pleasure of listening to Knuth speak at my university's computer science department. I joined the very large queue of other people, and expected

I once had the pleasure of listening to Knuth speak at my university's computer science department. I joined the very large queue of other people, and expected the great man to totally revolutionise my understanding of – well, I didn't know what, but something. Computer programming to TeX, I don't mind. I was expecting enlightening solutions to famously difficult problems, a great study of algorithms, that kind of thing. Well, what happened was that he spoke for nearly two hours on self-referential aptitude tests. The kind of "20 questions" game where the 20th question is "The maximum score obtainable on this test is (a) 18 (b) 19 (c) 20 (d) indeterminate (e) achievable only by getting this question wrong" and all of the others are inherently recursively linked. I had no idea these things existed, even less of an idea why anyone would want to study them, and came away later with both mind dripping down the side of my skull and a greater appreciation of SAT solvers and compiler design in functional programming languages. Highly recommended. (And if anyone wants to do the quiz, it's here: [1] https://www-cs-faculty.stanford.edu/~knuth/paradox.pdf https://www-cs-faculty.stanford.edu/~knuth/paradox.pdf)