Proposed Committee Structure Changes

I wanted to get working on some modifications/amendments to RPIPs 10, 11, and 15 vis-a-vis the committee structure. Consider this post a trial balloon to float some ideas and get some feedback. I’d like to do the actual work of writing this stuff in January so that we can get it voted on in February (for a specific reason below).

One-Year Terms for All Elected Positions
As of right now, all elected positions (IMC, GMC, pDAO Treasurer) are basically for life or until someone gets tired of it or is removed. It would be good to have fixed terms. This would allow for regular approval from pDAO voters, would allow for more people to potentially get involved, and frankly is more fair to the people in those positions who therefore have a regular opportunity to say "thank you, I'd like to step down" without having to feel like they are disappointing anyone.

I’d like to push for all elected positions to be one-year terms, with no term limits. That means people could put themselves up for renomination if they would like to continue to serve, and in fact my expectation is that many people would choose to run for multiple terms. But at least this way they are subject to annual voting from the pDAO.

Election Calendar
In tandem with the above proposal that all elected positions go up for re-election annually, I would propose the following calendar. The term of each position is January 1 to December 31. On or around November 15 of each year the nomination process would begin, following the protocol outlined in RPIP-10. That would allow for the election to be completed by December 31. Each management committee would then select a Treasurer early in January, with the understanding that it could be the same person who did it previously.
Vacancies
We do not currently have a specific set of rules for vacancies. As per the discussion in the above-linked thread, I would propose that should any vacancy occur, the position be offered to the next-highest-vote-recipient in the most recent election. The pDAO would then go down through the list in order of votes received until the position is accepted. That person would serve the rest of the one-year term. Should the list be exhausted, snap elections would be called to fill the seat(s), with the results of that snap election now being considered "most recent election" for any future vacancies in that year.
RPIP 15 Community Member Status
Credit to @Pieter for [pushing this forward](https://dao.rocketpool.net/t/gmc-clarifications-membership-vacancies-and-community-status/1251/11?u=calurduran). RPIP-15's current specification of community/team labels doesn't really work with the recent elevation of Rocket Scientists to oDAO positions. The sentiment in the poll in that post, and my own inclination, is to drop the community majority requirement and leave it to the will of the voters who they want on the committee.
RPIP 15 Membership Numbers
RPIP-15 says that the GMC should have at least 7 members. I regret leaving it open-ended. A committee should have a fixed membership if it's going to be elected. I would prefer to just change that to exactly 7, for a 5 of 7 multisig (which is what it currently is).
RPIP 15 Grant Frequency
This one is more speculative. During the voting for RPIP-15, someone pointed out that doing rounds of grants every two months was too frequent, and was significantly more frequent than other grant cycles in crypto. It is potentially a recipe for GMC burnout. I'd like to open a discussion about moving it to quarterly, with award rounds opening on January 1, April 1, July 1, and September 1.
8 Likes

I think the one-year terms, vacancies and Community Member Status points form a great summary of community sentiment expressed in the previous thead: GMC clarifications: membership vacancies and 'community' status - #11 by Pieter. :+1:
Fixed membership numbers and reduced grant frequency make sense to me personally as well. I also like combining many ‘household’ improvements into a single vote for efficiency and reduced voter fatigue - as long as we have rough consensus here.

The proposed fixed annual election calendar has the advantage of clarity, but I also see some possible disadvantages. As it is the same schedule for all committees, it would concentrate workload into a small span of time. The vote might overlap with the busy holiday season, so we could see reduced engagement unless we make sure to conclude the vote before mid December or so.
Alternatively, we could have varying calendars, e.g. 1 year after a committee establishment or at arbitraty dates spread out across the year (decided on 1 time per comittee.)

I also have a question process-wise. RPIP-10 is living, so it can be updated with changes voted in without a problem. However, RPIP-11 and RPIP-15 are both marked final, and as such only allow errata and minor clarifications. The proposed changes to RPIP-15 are likely beyond that scope. Concretely, does this mean introducing an entirely new RPIP that supersedes RPIP-15?

Yes on the last process thing. A new rpip that either supersedes or modifies, eg, RPIP-15. I think it would be OK to add a post-facto note about the existence of the new RPIP-15.

If we do annual, I agree the year change is a bad time. Could do November and make RP anniversary based? Or any other time really, but that came to mind.


I’m not sure about the autofilling vacancies thing instead of oversizing committees. I like oversizing tbh - simply presents a quorum of membership alongside the quorum of votes.

Autofilling works well when there’s a lot of good candidates that stepped forward (gmc), but not if it was tough to fill slots (pDAO treasurer had only 2 candidates, eg).


Agree on community member status.


For grant frequency, I’m OK with that if we provide something nimble too.

My suggestion is a 4th category of “GMC discretionary” rewards where each GMC member gets a little RPL to give out as they wish each period (unused RPL just goes to general GMC treasury at the end of the period). Track in a little public Google sheet so the community can provide signaling feedback.

That makes sense about structuring the parts that modify 11 and 15 as a new RPIP. I also agree that the holiday period was poorly thought out on my part. The idea behind having all of the elections at the same time is that there are time costs involved for the people organizing them (and the voters) and so it’s nice to just do them all at once. And having the committees all turn over with the new year made some sense to me. But I’m also okay with having them staggered. Or keeping them all at the same time of year but pushing the start of the election process up to Nov. 1, with elections wrapping up by Dec. 15 and new people taking their seats on Jan. 1.

What does overfilling mean in these cases? If the current GMC multisig is 5 of 7, then would the idea be to set it to 9 so that you would likely never drop below 5 during a year? What if it did? We’d still need a process on the books for replacement. Oversizing would just push the autofilling off until we’d hit the minimum needed for committee functioning. Neither autofilling nor oversizing really work for something like the pDAO treasurer, since you can’t oversize a one-person position. For the committees, I don’t think they are mutually exclusive. We could set both IMC and GMC to 9-person committees (since I think they are both 5 of 7 multisigs and would become 5 of 9), with autofilling for vacancies. I think I’d prefer to leave them at 7 and do autofilling but it’s not a strong preference.

I agree that some sort of way to have smaller awards for time-sensitive things would be good. Not sure what that would look like. Could be that there’s an open call for small grants (not bounties, not RAs) with a maximum RPL cap and a three-person subcommittee of the GMC rotates every month taking turns reviewing any small grant submissions? Sort of like the Supreme Court’s shadow docket. Maybe limit it to a person can only receive one such small grant per calendar year.

I’d say no more than 9 because at some point it becomes unwieldy (even with 7 we have failed to find a single time slot in a given 10 day period where all 7 of us are free).

Fixed committee size seems tighter to coordinate to me (esp. if you have lots of candidates and no upper bound either.)

If it’s tough to fill spots then you’re going to have trouble replacing people, regardless of autofill (you either don’t have candidates for autofilling or you’re likely sitting at a minimum committee size and require immediate reelections if anyone leaves.)

I don’t know if autofill would apply to specific single positions like a treasurer. But if it would, autofill would actually be advantageous as in not requiring immediate election if the one person filling that specific role steps down midway.

Agree on fixed terms, however a year is a very long time in crypto, I would expect a high amount of turnover in that length of time. A quarterly (or 4 month) term would not be overly onerous voting wise once it becomes routine and would reduce fatigue/stagnation in committee membership.

Autofilling vacancies from next highest vote makes sense if votes are recent and not too many people are being replaced but some minimum vote share requirements would be needed to prevent election by default.

On membership numbers, while members of the IMC have been fairly active, I have had some concerns about bus factor rendering funds locked, I would suggest adding an additional signer (6/10) , perhaps another team signer.

1 Like

Maybe it depends on the committee. For something like GMC, anything less than a year term would be pretty painful for continuing members as you’d be constantly losing institutional memory. It’s true that time works differently in cryptoland, but I can’t think of many (or any really) voluntary positions in organizations that are less than a year commitment, in crypto or out of it. I’d rather set the expectation at 12 months and if a person or two feels the need to step down mid-year, then you replace them. Rather than proactively setting the committee system up to potentially replace whole committees every three or four months.

A year term seems reasonable to me. Less than that we risk vote fatigue, loss of institutional memory, whiplash strategic direction, and broader surface for malicious interference. If the community is unhappy and wish to change direction mid-cycle, the normal governance channels can always be utilized. I think as long as a majority of the committee is elected in the normal way, the ‘next highest’ can fill vacancies; however, I would encourage you to consider a clause to call immediate elections if [a majority minus one] (i.e. 3 or 4) vacancies have occurred- both for appearance of democracy and because it may signal a maladaptive committee if people are resigning en masse.
For unique role vacancies within the committee, can the committee choose the replacement either amongst themselves or outside of their choice?

What’s our plan if someone wants to/needs to step down mid-term? We should have a structure for calling special elections and how said elections would run.

That’s the “vacancies” section in OP. My counterproposal is to simply oversize committees. Ie, elect 8 if it must have 7 - this allows time for elections easily (and in most cases when people step down they’ll probably be able to serve during the election tbh).

We originally had some language around expedited elections in RPIP-10, but that didn’t make the final cut.

@epineph mentioned unique role vacancies. For IMC and GMC treasurer, those are selected by the committee, so no real issue. For pDAO treasurer that would be an annoyance.

There are three possibilities for vacancies that I can see:

  1. Fill from previous election list (vacancy section in the OP)
  2. Oversize the committee and leave any vacancies open until the next election
  3. Snap elections

I really don’t like (3) for us. It would be fine if we were an in-person group of people or something, but elections require significant work for the pDAO, and we could have multiple vacancies throughout the year.

I’m fine with (2), though even in that case we will still need to have a written plan for what happens if a number of vacancies accrue during the year that you get near the minimum for a multisig. Perhaps a combination of adding a couple extra people to GMC/IMC (and any future committees) and then having a “break in case of emergency” plan for filling vacancies past that. That could either be going to the previous election’s list or having snap elections.

For idiosyncratic positions like pDAO treasurer, I could see the argument for a snap election or for filling from previous election lists.

Agree with this. Oversizing or filling from the previous list can both work. For oversizing, I still think we should have a sane maximum committee size as well as minimum. The larger the committee gets, the more overhead and effort is required in coordination. Another downside of oversizing is that candidates who barely receive any votes could still easily make the cut - and then why vote at all?

Perhaps add (min multisig votes / 2, round up) required as ‘oversize vacancies’?
So for example a 5-of-7 multisig committee, you’d get a minimum committee size of 7 members and a maximum size of 10 members.
(We should be careful to distinghuish the committee committee size of 7 from the multisig minimum votes of 5 here, referring to both as simply ‘the minimum’ can be confusing.)

Such a committee can still function optimally at 7 and technically even at 5 or 6 members. But you don’t want to be at 5 or 6 for an extended time. I think that when a committee falls to the committee minimum it should be on alert (perhaps notify the community) and when it drops below, a snap election should be called.