There was an article on HN, where the author stated that they didn't like to write own algorithms "like in '90". In this context, the Arseny's article should b

There was an article on HN, where the author stated that they didn't like to write own algorithms "like in '90". In this context, the Arseny's article should be titled: "How to be old-school and tailor algorithms & data structures to your needs". I have this haunting feeling that people nowadays are afraid of "reinventing wheel" even though it'd pay off. A recurring argument is that the standard library or boost or something available on github is better than a custom solution. But as we see, there are cases where saying "nope, I'll do better job myself" has sense.