This is really good information! Thanks for writing it up. Honestly, I never actually "trust" the complexity analysis. Whenever I find a paper I immediately lo

This is really good information! Thanks for writing it up. Honestly, I never actually "trust" the complexity analysis. Whenever I find a paper I immediately look for their specific results on an experiment, and if I can't find that I will assume the paper is only of theoretical interest. There's of course still a lot to learn from a purely theoretical paper, but I've always ended up being disappointed when I've implemented something which only had a "good" asymptotic bounds.