In the buzz over the release of the draft Gov 2.0 report the SMH has reported this as recommending "the public service turn to new media for 'crowd sourcing' ideas to elicit feedback as part of an effort to put bureaucrats more in touch with the people they are employed to serve."
Technically crowdsourcing is "distributed participatory design". Or as a reviewer of Crowdsourcing: Why the Power of the Crowd Is Driving the Future of Business noted "In his prescient 2006 article in Wired, Jeff Howe coined the term "crowdsourcing" to describe how the Internet has enabled large, distributed teams of amateurs to do work that was previously the domain of isolated experts or corporations."
An example of how Goverment might run crowdsourcing is the call for input to the realising our Broadband Future conference. People have been asked to contribute ideas before hand, and to participate from outside the event through twitter and all such other forms of new media.
When I was in secondary school I wrote an essay on Robespierre and the Terror followng the French revolution. In that essay I described Robespierre as the great democrat because he responded to "the will of the people" as best he could interpret it. Bobbitt who I referred to earlier suggests that market states will use the techniques of citizen referenda and recall to "by-pass" the role of elected officials (P.89).
The difficulty is how to distinguish between a crowd and a mob. In the Origin of Wealth (I think - it might be in Nudge) there are references to studies about the popularity of music and studies that show the "network or bandwagon" effect of how others rate a song can be more important in popularity tha the song itself.
In a crowdsourced environment it is very hard to distinguish a genuinely good and widely supported idea from a merely popular one. The authors of Gov 2.0 need to remember that, in the end, the guillotine that Robespierre deployed at the whim of the mob terminated Robespierre's life.