Random 'choices' and the locality loophole

Abstract: It has been claimed that to close the locality loophole in a Bell experiment, random numbers of quantum origin should be used for selecting the measurement settings. This is how it has been implemented in all recent Bell experiment addressing this loophole. I point out in this note that quantum random number generators are unnecessary for such experiments and that a Bell experiment with a pseudo-random (but otherwise completely deterministic) mechanism for selecting the measurement settings, such as taking a hash function of the latest million tweets with the hashtag #quantum, would be as convincing, or even more, than one using quantum random number generators.