I’m at CHI right now and just saw this talk. Haven’t read the paper yet, but it seems like it might have some interesting takeaways about a population of scientific programmers with interesting, idiosyncratic needs https://dl.acm.org/doi/10.1145/3706598.3713621