> we don't make hypotheses and don't do experiments. I don't formally write them down (unless it's a particularly complex case), but during debugging and perfo

> we don't make hypotheses and don't do experiments. I don't formally write them down (unless it's a particularly complex case), but during debugging and performance tweaking I find myself making hypotheses and experiments all the time (according to the lay definitions of those words, at least)