Only one third? That's pretty good! "How much time you spend fixing the open-source code" is the wrong metric. "How much time you didn't spend writing that ope

Only one third? That's pretty good! "How much time you spend fixing the open-source code" is the wrong metric. "How much time you didn't spend writing that open-source code from scratch in the first place" is the metric. (Open source sometimes fails that one, too -- there are times when you ask yourself whether you should have just started from scratch -- but it passes more often than it fails.)