Somebody got a good review for breaking it (but improving some other metric), somebody will get a good review for fixing it. Net result for users is zero, but a

Somebody got a good review for breaking it (but improving some other metric), somebody will get a good review for fixing it. Net result for users is zero, but all the churn looks like productivity in a "metrics based" GAMMA review process. I've seen worse. (Same problem with GDP as a measure of economic productivity BTW, but that's a topic for another day.)