Randomization and trial supply

Arms, list hashes and seeds are shown only to people unblinded in the study. Kit numbers are random and say nothing about the arm. A participant is randomized once; repeating a request returns the first answer.