It's often a good source of creativity to take a question and flip it around.
About when I pulled out of science, (and before I honed my political skills, at least a bit, in the private sector) I was at a meeting of paleoclimate modelers, when I kept saying "flip the question". They were trying to pull together ways to apply modeling to study paleoclimate, but I thought and think that the best approach would be to use paleoclimate data to study how to improve climate models. (I don't think anyone ever even got a clue about what I was so agitated about.)
I am now writing an article about "computer programming for everybody". Here, a clear statement of the flipped perspective is offered by Mark Guzdial: "I'd never before thought about computing for learning as opposed to learning about computing".
This site originates in a question flip. People are thinking hard enough about how to communicate science to the public, but as with the other two questions, they aren't thinking about it very well because they haven't looked at the dual, the flipped question. The flipped question is how to communicate how the public thinks to scientists and scientifically inclined people. Without some care about the flipped question, the communication of science to the public will tend to fall on deaf ears.
The way ideas prevail in science and the way they prevail in society are distinct. To some extent the way ideas prevail in society is broken and needs fixing (in that it is too emotional, and in that networks of trust are failing, and in that public opinion is too vulnerable to cynical manipulation), and to some extent the way science works is broken (mostly in that it is too clubbish and inaccessible). Anyway, given that we have matters of great importance at stake, we have to cope with the situation we have.