The Third Path: Delegate the Operations, Not the Keys

The temp check gets the diagnosis right, so let us say that first. A DAO that votes on line items is over-governed and under-governed at the same time, ja. Token votes are slow, low-context, and binary. Running an org needs the opposite. On this, no argument. The remedy is where it goes wrong. Diagnosing "voting is bad at operations" and prescribing "move $100M in liquid assets and ~$250M in ENS to a five-seat board" are two different moves, and the proposal treats them as one. They are not. You can fix the operations problem without moving the keys. That the second does not follow from the first is the whole point. The custody transfer is the actual proposal, and it is not scoped to the actual problem. Delegate fatigue is solved by decision-tiering and an operating budget. Neither requires custody of the treasury or the Endowment. The proposal reaches for the keys while pointing at the calendar. So: the $350M base stays in onchain DAO control. The Foundation receives a capped operating envelope, streamed and renewable, enforced onchain, never custody of the base. It gets a budget, not the vault. "Trust us" is doing the load-bearing work. The paper's own framing is that credibly neutral protocols beat institutions you have to trust. Correct. Then it asks you to trust an institution: a board whose founding slate is assembled by the entities it oversees, with the founder holding a seat, ratified while that same founder has self-delegated roughly half the voting supply. A protocol spares you the favor of trusting character. This design asks for the favor. Fix: independent-majority board, no founder seat, and no self-delegated supermajority deciding its own empowerment. A change this size should pass the way changes this size pass. Moving the treasury off the protocol is not a normal proposal, so it should not ride a normal threshold. Constitutional-grade vote, explicit red lines written into the mandate, high supermajority. If the case is as strong as the paper says, it clears that bar. If it cannot clear it, that is information. Delegation should be reversible by default. Mozilla and Signal are cited as the model, and the same paper concedes their communities have no recourse if the board drifts. So do not copy the part that fails. Every delegated power carries a sunset and a clawback the DAO can pull without the board's cooperation. Power you cannot take back is not delegated, it is given away. Where this sits with the rest of the thread. This proposal is the custody spine, not the whole answer, and it is built to compose with the strongest work already here. The legal wrapper and institutional capacity: run the Foundation as an authorised agent, not a custodian, as Empowering the Foundation While Preserving Onchain Custody sets out. The release schedule: size a Foundation Pool to operational run-rate and tranche it quarterly rather than transfer the base whole, per Foundation Pool. The input side: a decision-premise register and a funded counter-search, so the DAO keeps the capacity to choose and not only to ratify, per Who Supplies the Premises. And the precondition under all of it: solvency, because ~$5.1M of revenue against ~$16M of spend is not fixed by any reshuffle of who holds the treasury, per The Revenue Mandate. Adopt the custody structure here, the agent wrapper, the tranching, the premise register, and the revenue mandate together, and you have a reform that survives a bad board, a bad year, and a bad actor at once. None of this blocks the Foundation. It scopes it. Take the operating mandate, earn the next slice with a track record, leave the base where no single hand can bend it. That is the version that survives everything the optimistic version assumes will not happen, because it never depended on all of it staying good. Na ja, this is not complicated.