There's a human tendency to overemphasize things you can quantify. So we try to figure out how to test every code path rather than what we should do: try to fig

There's a human tendency to overemphasize things you can quantify. So we try to figure out how to test every code path rather than what we should do: try to figure out which inputs we should test against.