Exactly, the link above starts well, but fails in some regards. If you use big O, you should give the mathematical definition and at least shortly explain it. T

Exactly, the link above starts well, but fails in some regards. If you use big O, you should give the mathematical definition and at least shortly explain it. The idea of an upper boundary isn't very hard, and it is actually important to understand that an algorithm running in O(n) is also running in O(n log(n)) is also running in O(n^2). It is at least necessary to understand the other greek letters, which really does come in handy in a deeper understanding of algorithms. The list of "common" O's is also kinda bad. Particularly, and I see this all the time, I think it is a mistake to go from O(n^2) to O(c^n), as this is the step that leaves polynomial time complexity, and glosses over the fact, that each level of exponent constitutes a different time complexity. Here the mathematical notation of O(n^2), O(n^3), ..., O(n^l) is indispensible. Nesting loops is probably one of the most commonly relevant applications of O-notations, so this actually has an influence on real-world implementations.