Someone - was it Kernighan? Ritchie? - wrote a nice description of this. In short, when you find the source a bug, don't just fix it - stop and think what cause

Someone - was it Kernighan? Ritchie? - wrote a nice description of this. In short, when you find the source a bug, don't just fix it - stop and think what caused you to make that error, and think about what else in the code might be a victim of the same line of reasoning (note: not necessarily the same code - just the same design decisions). Back in the day when you had to submit your code to the computers overnight and wait until the next day to find out if you got your output or just a one-line compiler error, you were much more careful to avoid errors, big or small. Having a longer iterative cycle can be good training.