“I write notes to myself, explicit ones, with sentences, not just “try power series expansion” or “see Dunford-Schwartz for $p>1$.” I think by writing… When I started to prove theorems (the acceptably low-key phrase for the more stuffy-sounding “doing research”), I would keep writing, as if I was conducting a conversation between me and myself.” Paul R. Halmos, I Want to Be a Mathematician: An Automathography, 1985.