Feature request: split proposals into a 2-stage process

JC Jaimie Cosmia Public Seen by 79

There are several reasons I think this is a good idea, but basically it boils down to that it allows ballot-setting to be organized and for legitimate decisions to be clear and conclusive as such, while still keeping things decentralized and equitable. It also makes proper score or ranked voting possible. The two stages could be framed as ‘Problem’ —> ‘Solution’ or perhaps better as ‘General Proposal’ —> ‘Specific Proposal’, if we don’t wish to frame all decisions in the form of a problem. General Proposals must receive a particular threshold of approval votes before progressing to the Specific Proposal stage.

Here’s why I think this makes sense:

1) Requiring the General Proposal to provide a deadline and a category* allows voters themselves to address approval or disapproval of these modifiers as a first priority without implying judgment on the actual content of the proposal. For example, Bob proposes we make a food delivery to the family at 123 Downing, that the deadline for the decision is 7 days and that the category of the decision is ‘communications’. The General Proposal begins getting a few upvotes, but Alice says, “hang on a moment” and posts her own General Proposal with the same exact content but a deadline of 3 days and under the category ‘mutual aid’. She explains that the family is in dire need now and that ‘communications’ wasn’t really the best category for the decision. Members begin to upvote her proposal instead, and it passes the threshold for Specific Proposal viability, leaving no ambiguity about the below-threshold votes for Bob’s and whether his proposal approaches a conclusive decision, despite that all the votes that did take place for his were approving. The General Proposal could be required to provide (or defaulted to) a deadline both for the determination of it’s own viability as well as for decision on the Specific Proposal.

* Loomio doesn’t appear to allow categorization of proposals at this time. I think that should be on the roadmap not only because it would help organize delegation of votes, if that feature ever comes, but because it would encourage clarity and would allow users to sort both ongoing and completed proposals by category. Categories could refer not only to content-type (environmental, economic, etc.) but also to decision-type (one-time action, statement of intent, permanent rule, etc.)

2) Building on the above, GP —> SP prevents ambiguity, focuses attention, and alleviates unnecessary stress, even where alternative deadlines/categories are not proposed for a decision. Another example: in a group of 500, Richard proposes we go to the family at 123 Downing and deliver them a jobs fair pamphlet. 4 users vote yay, 1 votes nay; the vast majority of the users never even saw the proposal, because it got lost in a deluge of other proposals, though they would have voted against it, viewing it as insulting or ineffective. But under the current platform design the measure . . . passes? It clearly shouldn’t be considered a legitimate expression of the group’s will, but there’s no procedural rule that makes that clear, leading to ambiguity. One might respond, “ok, we’ll simply require that all proposals engage a certain percentage of participants to be marked as conclusive”, but this allows unlikely votes to sneak up on people. A user might just ignore a proposal that seems fringe to them, only realizing too late, after any chance for deliberation, that the proposal is going to receive enough votes in the last hour to cross the participation threshold. A two-stage proposal focuses voter attention; when a General Proposal becomes viable, they will receive a notification, and they’ll know that a decision IS going to be made on this issue*, demanding their urgent attention. Conversely, if General Proposals are coded in such a way that only upvotes are possible, users can take comfort in the fact that their lack of awareness of a GP isn’t aiding the passage of an undesirable decision (at least, not to the same degree it would under current design), freeing them from the need to read and respond to every single proposal in a large group, no matter how ill-conceived. In a Reddit-like way, this system can draw popular GPs to the top of a sorting list, helping them to gain attention and reach viability—or increasing pushback in discussion threads/competitive proposals with different deadlines or categories. Overall, I think the progressive structure provides a lot of clarity regarding how seriously to take proposals at each level of development and that this is crucial in large groups.

*I do, however, think “Take No Action” should always be presented as an option, just in case voters end up dissatisfied with all Specific Proposals and decide to back out of the decision. Viability of a GP still focuses attention, though, as this is likely to be an infrequent event.

3) GP —> SP further improves organization and reduces ambiguity by gathering similar concerns in the same place, leading to more decisive and carefully selected action. Imagine again a large group with lots of members and lots of proposals, and consider the following scenario on the platform as it currently exists: Alice proposes we bring 5 loaves of bread to the family on 123 Downing in a surprise gesture; Bob proposes we bring 4 loaves; Richard proposes we bring a cabbage; Marianne proposes we first confirm whether the family would appreciate aid and that if so, she will personally volunteer to take them 6 cans of beans, a cabbage, 3 loaves of bread, 3 lbs. of chicken breast, and 4 lbs. of tomatoes, using no more than $36 from the group’s mutual aid fund. Jackie would have voted for Marianne’s detailed and generous proposal, but she didn’t see it as it was scattered among a lot of other unrelated proposals, so she votes for Alice’s—an inferior outcome. Lots of other people did see Marianne’s proposal and voted for it, while many others may or may not have seen it and voted for Alice’s. Sooo . . . what now? Both measures passed—are we doing everything Marianne suggested and also taking an additional 5 loaves of bread? Are we surprising the family with the 5 loaves and asking them if they would like to receive the other order? Our scattered approach to proposals has kept some from being seen, split votes, duplicated votes, and generally caused a lot of messy ambiguity. It also makes it harder for related proposals to take inspiration from and build on one another over the course of the decision period, rendering the whole process less deliberative. It is easier for people to agree on problems and general ideas than it is on more specific proposals. GP —> SP first rallies us around the problem or basic suggestion and allows us to decide whether it’s an issue there’s even sufficient interest in taking up; it then allows anyone to put forward a more specific and detailed proposal for actually carrying the decision out. GP: let’s get some food to the family at 123 Downing. SP: here are some detailed options for how to do that. Another advantage of this approach is that it allows us to then use score or ranked voting to choose between the Specific Proposals that are put forward, enhancing our capacity to identify the most ideal/preferred solution.

Apologies that this is such a long post, I’m eager to hear your thoughts!


Deleted account Sun 24 May 2020 1:00AM

To take complex decision, maybe an amendment system would be required, where everybody can propose amendments, and then both the proposition/s and the amendments are voted, during several phases.


Robert Guthrie Sun 24 May 2020 5:28AM

I have heard many different but similar requests for this UX. I would encourage you to keep iterating on the specification and, particularly, to draw sketches or wireframes, because I get totally lost reading this stuff.

It might take a few weeks to mock up, but it would be worth it if you can describe it really clearly.


Jaimie Cosmia Mon 25 May 2020 2:28AM

Learning to wireframe right now—I’ll probably come back to this at a later point!