> If Google contract out the development of blue sky ideas to small companies or individuals and that yields products like gmail, that would be a sign. So far though, it's in house or acquisition (a form of contracting?) for really important things.

Blue-sky research is the classic example of the sort of stuff that is particularly hard to build a market around, hence most such research is publicly financed (shielded within universities and government labs) or privately financed by huge, private pseudo-governments like IBM, Microsoft, Bell, etc.

The fact that venture capital has allowed the market to compete with firms for certain kinds of "short term blue-sky research" (i.e., research that is very high risk, although still relatively short time-to-payoff) is certainly a great development. But that doesn't mean we should expect other blue-sky research to be the first area where the technological suppression of transaction costs allows markets to displace firms. Rather, it seems to me that the conventional wisdom is still mostly correct, and we should expect work that is more easily measurable than research to be freed from big firms first.

See also economist Robin Hanson for arguments and data that typical firm size will increase in the future, technology not withstanding:

http://www.overcomingbias.com/2013/09/expect-bigger-orgs.htm...

[2] I wrote a similar comment once, doubtlessly channeling Coase: http://slatestarcodex.com/2014/09/24/book-review-red-plenty/...