Failure rates
"... the machine had never had a hardware failure before so software was more likely."
Is that actually a reasonable assumption? If a machine has never had a hardware failure, then doesn't the probablility of failure increase as time goes by?
Biting the hand that feeds IT
