All crowdfunding campaigns necessarily have uncertainty; it's just a matter
of where the uncertainty is.
I don't think you can raise money for a video game unless you've already got
it about halfway done. So with crowdfunding, there's a dilemma: either you
can require a certain goal to be reached, and either abandon it entirely
(wasting the effort you've already taken) or release the half-finished game
anyway; or you can not require a certain goal, at which point what stuff
promised in the crowdfunding campaign gets cut can be arbitrary. Either it's
a gamble for the developer of total failure, or it's a gamble for the funders
of not getting quite what they wanted.
My idea is to instead change the risk to what proprietary software developers
risk: the developer not getting as much money as hoped. This is better than
the risk of not getting anything at all, and at the same time avoids any risk
to funders.
