I forget all the details, but once upon a time we were brainstorming comms systems, and things were in the process of stalling.
I asked "how would you do that with yoghurt", and things got moving again - and reached a useful conclusion.
Trivial? Yes, of course.
Feynman told the story of being sent out to assess a production facility for making the bomb in WW2. He knew he wouldn't be able to provide any detailed technical assessment. He did randomly select a valve and ask what would happen if it jammed. That kicked off a discussion amongst the local staff, and they did discover a significant vulnerability.
Yes, but the ability of someone "with a different toolkit" or who isn't "in the middle of the trees" can be significantly helpful.
I'm sure you can think of such cases from your own experience.
Unfortunately, often they do turn out to be useful in ways not imagined.
That's my excuse, and I'm sticking to it.
Yes, and that's a key point.
That's why brainstorming /must/ have the /second/ phase: select possibly helpful avenues and discard the rest.
Yes indeed; that's the other key point.
If anyone wants examples of that, there are whole website devoted to spotting them in software.