Innovative projects frequently do not get off the ground because they cannot be aligned with existing policy, programs & projects.
We need a channel of funding and project governance which supports the initiation, funding, resourcing & measuring of what I'd call "Wild Hare" projects. These are projects which may fail, or may be wildly successful - we don't knwo when we start them. So they are inherently risky, and so I'd suggest that funding be strictly capped.
Wild Hares need to be funded on an ongoing (financial year to Fin Yr) basis - and probably best led out of a central agency.
Wild Hares are means by which we test if a Gov 2.0 innovative idea has legs.

Comments (1)
This is an extremely important suggestion. There does need to be some real fringe stuff funded.
If we look at the Venture and Angel capital model, there is always some ratios of success/failure. VCs and Angels tend to back a bunch of things, prepared to lose on most, in the hope the biggie
A look at people like Sequina Capital shows this approach in practice.
It is so hard to work out what will work and what will not. I know I have been wrong on many occassions.
So, yes, a nice little fund for some crazy ideas may well pay dividends