[Google, Facebook, Microsoft] Research are probably similar in many ways -- well funded by rich companies that are self-interested in moving the field forward.

[Google, Facebook, Microsoft] Research are probably similar in many ways -- well funded by rich companies that are self-interested in moving the field forward. We'll only know their true impact decades down the line, of course. MapReduce has certainly been very influential. Bell Labs wasn't just a playground with infinite cash, they were very much supposed to figure out telecoms stuff for the benefit of Bell -- they just had a very wide berth for determining what that something was. I suspect the same is true for the aforementioned research departments. (That said, I don't think either, nor Bell Labs, is a meaningful model for academia in general)