I missed the original "Facebook Lost a Great Engineer" post until now. I agree with a lot of what was said in both posts, I think puzzles are generally a bad w

I missed the original "Facebook Lost a Great Engineer" post until now. I agree with a lot of what was said in both posts, I think puzzles are generally a bad way to hire and they get especially dangerous the more senior an engineer is (the longer they are out of school or out of the initial learning phase if they are self-taught, the more such questions are detached from day to day development and thus you're only testing if the developer has been practicing to interview lately, not whether he or she will be an effective developer). However, this statement from the "Great Engineer" post struck me the wrong way: "I would fire any developer who chose to re-implement standard library functions." That's a terribly absolute stance to take -- consider, for example the widely-discussed fast inverse square root: http://en.wikipedia.org/wiki/Fast_inverse_square_root http://en.wikipedia.org/wiki/Fast_inverse_square_root That was a re-implementation of a standard function (rsqrt), but for a very good reason (speed was more important than accuracy for the use case). While that specific trick is mostly obsolete in development today, if you're pushing the envelope even a little bit with your code, it isn't that rare to run into a situation in which it makes sense to side step the standard library implementation of something. tl; dr - Developers (at least non-Sith ones) should never deal in absolutes.