Nottingham's commentary on RFC 8890, an IAB document arguing that IETF protocol decisions should favor end users — actual people — over other stakeholders when interests conflict. Key points for practitioners: it's persuasive, not binding (IAB document, no IETF consensus). It builds on RFC 7258 (pervasive monitoring is an attack) and RFC 6973 (privacy considerations), and explicitly rejects the "layer 8 is somebody else's problem" stance — protocol design decisions like DoH and Encrypted Client Hello have political outcomes whether participants admit it or not. The interesting operational claims: the IETF is a quality filter whose refusal to adopt proposals carries real weight; there are no standards police, so it can't stop vendors from deploying things it dislikes (DoH was deployed by browser vendors before standardization); consultation matters but doesn't grant external parties a veto; and the "pessimal environment" principle — a win for users in one jurisdiction that harms users elsewhere is a bad tradeoff. If you participate in the IETF, the actionable delta is to evaluate proposals on user impact with the same seriousness as technical merit, and consult affected parties before standardizing things that will surprise them. I remember when "the Internet is for end users" was implied by the fact that it was users at both ends of the 300-baud line. Now it takes an RFC, an IAB vote, and a 2,000-word essay to state the obvious. Progress reportedly ships next quarter.