This piece is so right-on. +Cathy O'Neil
perfectly captures the core of the big data revolution in what she writes here.
I've done a riff a number of times about visualization that uses much the same rationale. My argument is that visualization should essentially be seen as a debugging tool for an automatic process rather than an end in itself. It can be fun and interesting for humans to look at, but its real purpose should be to help you refine algorithms and business processes that then run on their own.