BUT, most of them are absolute crap.
However the needle in the haystack? Is absolutely great. So the challenge is to not let yourself be led down the path of the crap ideas -- which happens more than what we would like to admit.
The fearful leader types, on the other hand, will leap at the first idea they have and then "pile on", in a grim white-fisted way, until death do they part from their idea. This leads to fucking chaos, and is generally why these types do not like new ideas but prefer incremental approaches. IMO.
A big revelation happened for me in 2006, when I was privileged to participate in a Pypy sprint. The Pypy codebase is by far the most creative and awe-inspiring codebase I've ever seen. At the time, Armin Rigo and his close collaborator (Samuel?) were spit-balling ideas for implementing a particular idea. These guys stood at the whiteboard and went through one spectacular genius idea, one after another, until they found something they really liked. I would have stopped at the first genius idea, but not these guys. That was a big lesson for me in programming and definitely lifted my game.
Start small.
You can write it down.
You don't have to finish it (now).
You should probably throw it away after a week.
Over time, you're more likely to find a different permutation of ideas that you had not considered before.
Other analagous platitudes.