Intel building fabs (fabrication plants) in China was strategic error that Stu predicted and was proven right on, but was 'right but broke', illustrating cost of being correct without institutional power.