For my mathematician colleagues, what is the best way to find the flashes of insight necessary to complete the proof?

Similar questions and discussions