Fair points. Still, statistically, it is useful to recognize the empirical (but imperfect) correlation between Big-O and run time. One does not have to be confu

Fair points. Still, statistically, it is useful to recognize the empirical (but imperfect) correlation between Big-O and run time. One does not have to be confused to know this. Rather, one would be lacking perspective to not recognize that there is a connection. This connection is important if you want to be empirical validation of an algorithm on a particular computer. I just want to point out that your comment could easily be perceived as a ding against self-taught programmers. Many self-taught programmers (which I will define as ones that have not had a formal CS degree) read extensively. Many use their intrinsic motivation to really dive in. Also, many are successful. Bill Gates is one example. Yes, there are programmers of all kinds that have a tendency to write sloppy blog posts, to make overconfident and inaccurate statements, to forget things they've read, to hack their way around, and so on. There may even be a statistical correlation between self-taught programmers and such behavior. But I'd suggest we points out those behaviors when they are a problem rather than make assumptions about their educational backgrounds.