On 4 Jun 2013, at 20:46, Tanner Swett <[email protected]> wrote: > On Tue, Jun 4, 2013 at 8:36 AM, Sean Hunt <[email protected]> wrote: >> No, this is awful because of the memory overhead. > > What memory overhead? Write a proto, put "submit the proposal" in your > to-do list for the day after tomorrow, forget about it, then it pops > up two days later and you submit it. That sounds like about thirty > seconds' worth of inconvenience.
The following not entirely relevant proto-proto comes to mind. Have you ever voted FOR a proposal while thinking "I agree in principle, but I would just tweak that..."? Obviously Agora has a well established tradition of proto-proposals and there is usually an opportunity to comment on proposals before they are distributed. But it makes sense to improve on this by obligating (and paying) a subset of players to review each proposal, while giving them the power to make amendments, because a) authors might not always make suggested changes and b) other players might not always be bothered to review each proposal. Broadly speaking if something is everyone's responsibility, then it's likely that no-one will do it. Imagine if each tracking the gamestate were left to whoever could be bothered at any particular time. Agora would die fairly quickly. Yet we leave proposal review to volunteers and spend more time on fix proposals as a result. So: a committee stage, which can be skipped for small, uncontroversial proposals. Each proposal is assigned to a committee like an appeals panel. The committee can edit the proposal and possibly increase its Distributability. After a certain time (5 days? A week?) the proposal is transferred from the committee and put in the Proposal Pool. To compensate for the time cost, make the following changes to the rest of the timetable: - Promotor distributes in the first 4 days of each week (currently a proposal submitted on a Monday could wait just under two weeks before distribution without the Promotor breaking a rule) - Voting period of 5 days - Assessor resolves within 4 days of the end of the voting period With a committee review period of 5 days, this reduces the maximum adoption time* for a proposal from 28 to 24 days. (*without someone breaking a SHALL, that is.) Or we could stick to 28 days (5, 7 and 5 time limits on the three above) if people think that's not enough time for officers and voters. This would also mean that errors in proposals could be fixed without allowing authors to cheat the Distributability system. I have some vague ideas about how to make this 'fun', with 'standing' (random) and 'select' committees vying for proposal assignment and committee members having some financial stake in the adoption/rejection of proposals they approve/disapprove of. This is obviously highly embryonic. I welcome comments and suggestions. -- Walker
