I agree that this seems to be true for some (many) of the problems, but it depends on the requirements. Some require only written proposals, some require deliverables etc. I think it's unfair to say the rewards are too low, especially when it is always up to the solvers discretion whether or not the terms are favorable.

Full disclosure: I'm not affiliated with InnoCentive, my brother just sent me the link this morning and I thought I'd share.

It does have a hint of 99designs about it - basically competing with others for a prize which is less than the minimum fee they'd expect to pay a competent contractor for time spent trying to achieve a solution.

I was actually wondering about that... there's a featured challenge on the front page: a biomarker for ALS.

The prize is 1m$... now, that is a lot of money, but something doesn't clash. I'm sure that'd come out to about the yearly salary for 10 researchers. No mention of lab costs, testing, publication, blah blah...

I'm not sure, but it smells like the typical Scriptlance (http://www.scriptlance.com/) fare, with a similar cost/reward point.

Meh. My lab will stay dusty :(

I agree about the reward / effort ratio, but it was unclear to me if that's purposeful (99 Designs) or if the intent is to spur innovation in certain areas (X Prize).

Looks very interesting. I'd suggest that you offer as much feedback to losing submissions as possible (see my other comment), to help losers improve and keep them interested in trying other challenges.