Hacker News Reader

Those criticisms are definitely common these days: lots of Sipser-style material is not very relevant or pedagogically compelling. I think push-down automata are a great example that many people would be happy to skip.

On the other hand, that choice of topics is pretty idiosyncratic. There are a lot of big concepts, some pretty tangential, that are covered very briefly and could easily be skipped. So I don't think that exact course is a good outline for a new canon. But I agree with the premise for sure.

I took ToC about a year ago, as the final required course in my PhD studies, and we used Sipser’s text although the professor also wandered off into the arithmetic hierarchy which he doesn’t cover.

I will absolutely credit the course with igniting an intense curiosity in the foundations of computing (and of mathematics in general) but I agree that it’s time for a more modern approach.

Interesting, thank you for the links. This version does not seem to have taken off, though. (The links are from 09.)