> In this particular example idling 99% of the time is cheaper, but even if idling 99% of the time is slightly expensive, it might still be a good idea not to o

> In this particular example idling 99% of the time is cheaper, but even if idling 99% of the time is slightly expensive, it might still be a good idea not to outsource due to resilience. Or in some cases it might make sense to become an outsourcing destination. Your otherwise idle equipment & skills become a revenue stream rather than a cost when your need it low, your local equipment and skills are tested & re-enforced which reduces the risk of finding them unexpectedly wanting when you turn to them in those 1% times, and you might end up with a small controlling advantage in your market (you get priority over any competitors that outsource to that part of you). It could become a significant arm of your business & profit. Think Amazon farming out its network resources.