Jun 01 2009
One of the many buzzwords that you hear in the discipline of software engineering is metrics. They're supposed to be a measure of how effectively your coders are functioning based upon how many lines of code they write a day, how many bugs they make (for some value of 'bug'), how reusable their code is, how much money per line of code your project is burning through, or some other arcane measurement. The numbers are generated through techniques that appear to have more in common with gematria than with engineering and make managers salivate with glee (or rabies). The theory is that if you can measure it you can quantify it, and if you can quantify it you can control it toward the end of improving it.
Generally speaking, the word 'metrics' makes people like me break out in hives.
But now there is a truely helpful metric that all of us can get behind: how much fail is in your code.