GMC clarifications: membership vacancies and 'community' status

[Pieter]

Membership vacancies

Soon after the initial GMC vote concluded, Wander, one of the elected members, withdrew from the committee.
This left us with 6 members, where RPIP-15 stipulates a minimum membership of 7. Informal community opinion in the Discord Governance channel indicates approval of selecting the next-in-line candidate since the vote concluded so recently. However, this process is not formalized in neither RPIP-10 nor RPIP-15.

Since this is relevant not just to the GMC, but committee selection in general, I would propose a vote to amend RPIP-10 with text to the following effect.
Under ‘membership selection’:

• If any selected nominees withdraw within 14 days of a succesful vote, resulting in insufficient members to fill all spots, the nominees with the next highest vote weight SHALL be selected as their replacement, without requiring a new vote.

On this topic I also noticed there’s no stipulation for MC members to voluntarily withdraw their membership. Has this been discussed? We also need to consider what happens if members become incommunicative for whatever reason. Perhaps their positions should be considered vacant after X period of time?

Community status

RPIP-15 (GMC) includes the following additional requirements regarding committee composition:

• The GMC SHOULD be composed of a mix of core team and community members, and SHALL at any point in time have a majority of the membership be community members.
• Individuals that receive a salary from Rocket Pool Pty to work full-time on Rocket Pool or who are members of the oDAO SHALL NOT be considered community members for the purpose of this RPIP.

The objective of these requirements is to avoid potential for financial conflicts of interest. With the recent addition of the Rocket Scientist role to a shared oDAO seat, the latter point has come under renewed attention. It raises the question of whether RS members can still be seen as ‘community’ for the purposes of GMC membership. This issue has been discussed in Governance, see here and again here.

Does a split oDAO seat (an 1/13 share in the case of RS members) constitute enough of a conflict w.r.t. this objective? I don’t think this is the case. After all, part-time financial compensation is allowed by RPIP-15 as well. And as calurduran said: “People become Rocket Scientists in recognition of their work for the protocol. It is therefore weird to also say that a majority of our grants committee must be people who have not been so recognized.”

At the same time, we want to avoid changing rules just to benefit a small set of influential community members, as it could raise doubts on the legitimacy of governance. So I’d prefer not to have Rocket Scientist-specific language in any rule clarifications.

Three suggestions have been discussed so far:

• ‘Non-boolean’ team / oDAO membership, so up to x% of full time salaray or y% of an oDAO seat allowable to qualify as community. (Disadvantage: more complex to use, exact percentage is still an arbitrary discussion.)
• Subjective self-identification as team or community in a nominee’s conflict statement. (Disadvantage: how to handle disagreements here?)
• Removal of the oDAO clause entirely (so it’s either ‘team’ or ‘community.’ Not a great fan personally, as I think it’s worthwhile to try to avoid potential financial conflicts of interest through oDAO membership too.)

If we can find a potential consensus here regarding community status, perhaps we could bundle it in a single vote with the RPIP-10 vacancy clarification, for efficiency.

[calurduran]

Re: Membership vacancies - that makes sense. I can’t imagine this being a common thing moving forward (someone leaving with X days of election, where X is small) but it doesn’t hurt to have it in there. Not sure if 14 is the right amount of time, or if something like 30 or even 60 days would work better.

Re: Vacancies - personally, I think I’d actually prefer a broader discussion about whether these committees should have specific numbers of people instead of minimums, and that each “seat” should have a set, staggered term, with re-election possible. We could even then specify that if the seat becomes vacant that you could refer to the next-highest-vote recipient from the most recent election (up to 364 days earlier) and have that person fill out the rest of the term. Maybe these topics need their own thread though?

Re: Community status - this remains tricky. I don’t think self-identification works, since that doesn’t really screen for conflicts of interest. oDAO is so hard to handle here. If someone works at one of the organizations that has an oDAO seat but they are also an active member of the RP community, how do we count them?

[drdoofus]

Maybe we consider adding alternates. If the committee needs n members, then the participants who received the (n+1)th and (n+2)th most votes are designated alternates. They step in in the event someone on the committee drops out or becomes inactive.

I am still a fan of removing the oDAO designation and having the committees only distinguish between the “team” and the “community”. Self-designation opens a potential can of worms. Oh, the non-boolean approach is also fine with me, though.

[Valdorff]

Recent vote thing: I’d go more like 30-90 days. Remember the pDAO can always proactively vote to change the membership should they wish to (sidenote: I think that text should probably get expanded, as it’s not clear what the process is… Offhand I’d do a kick vote followed by a selection as written).

I don’t think voluntary withdrawal is needed as a separate item. A two thirds vote kicks and someone wanting out should result in a unanimous vote, really. I think this would also be used for incommunicative etc.

Community: I like self-identifcation. If folks feel strongly enough that someone misidentified, they can point it out loudly and/or vote against them. It’s not perfect, but it’s simple, and that may be more important.

@calurduran

whether these committees should have specific numbers of people instead of minimums

I think this requires an autoreplace built in. The IMC cannot be nonfunctional for even a day if it’s the wrong day. That’s why I quite like a minimum and keeping above it.

[Pieter]

Yeah, 14 days is probably too small a window. I chose it arbitrarly to open the discussion
With a larger window it can still function for the purpose of alternate / back-up nominees. I don’t think it should be indefinite or too large a window either though - at some point the vote results are stale as existing nominees will have moved on, newly interested ones become available, or other external conditions have changed. 60-90 days seems like a good balance.

@Valdorff

I don’t think voluntary withdrawal is needed as a separate item. A two thirds vote kicks and someone wanting out should result in a unanimous vote, really. I think this would also be used for incommunicative etc.

That’s a fair point. Perhaps the ‘incommunicative member’ and ‘voluntary withdrawal’ can still be added as illustrative examples, as these applications of the vote kick didn’t occur to me immediately.

Re: self-identification. The idea is that people report their team / oDAO compensations in their conflict statement and argue whether or not it’s substantive enough for independence. If they omit anything they can be called out for it - the problem I see here is more in the grey area where a nominee has several side gigs going and people disagree on their status.

[epineph]

A) My recs re: vacancies:

1. Replace vacancies with next highest vote; there should be no time limit from election to replacing vacancy; we need a succession even a year out, because a snapshot takes several weeks and the work must go on.
2. I think if >3 members total have dropped out since last election, the committee cannot continue until new elections; this prevents the majority being filled with non-elected members; as such, once 3 members drop out i would have a vote triggered to prevent halting of committee duties.
3. I agree with calurduran; there needs to be a fixed term if we are adding non-elected members, although i think staggering seats-particularly as we get more committees- may lead to snapshot fatigue. I would just have a vote for the full membership once a year.

B) my recs re: community status

1. Object_object was voted on before he was a paid contributor, so i think irrefutably grandfathered in

2. Ongoing, and this an unpopular minority opinion, I believe even a partial salary through oDAO needs to count as a team member. Even small gifts substantially affect behavior (see drug reps giving out pens to doctors or a 500 DAI dinner fundraisers for politicians). If RPL 10x, fun spending money (and aren’t we talking ~10k DAI per rocket scientist a year, which is like median income for the majority of the world) may become full time (Western European) salary with no specific responsibilities. Categorizing all paid members as non-community would also prevent co-option of a majority of committee members by the RP team in the future (eg, voting for things that are favorable to RP team increases chance of becoming RP scientist, creating positive feedback loop). And, since time spent in Discord is essentially campaigning for oneself, paying people to be on Discord gives them a substantial electoral advantage already.

As a compromise, I would be fine decreasing the “community members” to 3 or 2, but it’s otherwise easy to see a scenario where all committee members get reimbursed by RP in one way or another, which would obviously be a loss for pDAO independence.

3. Another possibility, not sure yet how i feel about it, is that there are always 2 permanent rocket pool scientist spots on each committee- they can decide amongst themselves who is the voting member for which committee.

C) bribe resistance, while we are talking about addending RPIP-10/15

1. there will be a substantial financial gain for approval of grants, bounties, etc, and obviously it is extremely easy to wash money in crypto. Consider putting in a “duty to report” clause for either suspected influence or communication from applicants outside the normal channels- if it is happening to you, it is certainly happening with the other committee members too.

[Valdorff]

A note on community status: this is currently only a call out on GMC membership, not on IMC membership or MCs in general.

This seems fine to me, as the GMC has a much more subjective job to do, and thus they need to represent the thoughts of the pDAO much more actively.

**Epineph B2 (who’s community) **
This is absolutely real money. There are a couple of fun complications:
(1) The oDAO isn’t the team. They can add/remove people without a single team vote, in theory. They are, however, significantly composed of the team and also team-correlated, so that makes it tough.
(2) We don’t want to punish service. The people being selected as RS are probably some of the people we most want on committees.

You gave me an idea however…
We could base the non-community tag on a dollar amount, and even base the dollar amount on something with meaning (poverty line, median income, etc).

Implementation thought:

• Get latest median wage in the US from a specific canonical source (any other metric works - this is an example)
• Ask folks to estimate expected payments from oDAO and team in the next year based on current token prices
• If above the line, say so (and get tagged as non-community for GMC purposes)
• If below the line, say so
  So this would be an additional requested nominee statement effectively. It’s a spin on Pieter’s self-reporting thought, but improved by (a) no subjectivity for the person making the statement and (b) some ability for others to cross-check.

Some other responses while I’m at it:

Pieter: illustrative examples

This is a great idea and easy to implement as “This MAY/SHOULD be done when…”

Epineph A1

This is presupposing we’re at minimum size. If we’re above minimum (like the IMC), then losing a member has no immediate operational impact and there is time to do a vote.

[epineph]

Community member thread reply to Valdorff

As with all things, I agree and I don’t. The oDAO can boot its own members (including the future RP scientist node); however, the oDAO cannot choose rocket scientists- this is (I believe) the sole discretion of the team. So the team will be able to choose which community members are paid (or not) through the oDAO, leading me to believe the RP scientists should be considered non-community.

Yes we want to reward our greatest contributors, but they are being rewarded with 1) formal recognition of their achievements 2) monetary compensation and 3) an enhanced voice to influence community members who would be voting. I hardly think it is punishment to prevent more than 3 of them from joining a given committee to prevent real or perceived conflicts of interest.

The GMC is a place where I expect general consensus between oDAO, team, and community, but our definitions now will certainly be precedent for future committees where true conflicts could arise. Just my two pennies worth.

[Pieter]

Lots of good points raised!

Vacancies

Combining our thoughts on vacancies and replacement, how about the following process:

• Hold a full committee vote 1 year after the last vote;
• Replace members who drop out (below the minimum required for the committee) with the next highest nominee from the latest vote (point above automatically limits vote validity to one year at max);
• If there aren’t enough remaining nominees available to to fill the minimum, trigger a new full vote;
• If half (rounded down) of the members originally voted in are replaced this way, trigger a new full vote.

@Valdorff re: replacement process. Kick vote followed by selection doesn’t work, I think. The existing committee can’t kick itself out it its entirety below the multisig minimum.
So you’d either have to add all new members first, then kick out the old. Or create an entirely new multisig with the new committee. The latter seems like a cleaner process, but does necessitate a transfer of the complete remaining balance after each full vote.

Edit: Valdorff pointed out to me that a Gnosis safe multisig allows for simultaneous kick and add. So the above is not an issue.

Community status

I’m coming around to @epineph’s view that any team- or oDAO-based compensation at all can lead to undesired influence on behavior. Even the perception of such should be avoided.
The dollar-cost cutoff proposed by Valdorff also seems hard to implement. How ‘substantial’ any dollar value is to a person depends a lot on where in the world they are located. Plus, being RPL-denominated, the actual dollar value can fluctuate wildly throughout a committee term.

To avoid these issues, I think we can go two ways:

1. Remove the hard team / oDAO / community clause from RPIP-15 entirely, and leave it completely up to the community to decide how to vote based on conflict statements;
2. Classify any team compensation or oDAO involvement as ‘non-community’ for the purposes of the GMC.

The first gives us the largest pool and flexibility for GMC candidates, but could be viewed as a loosening of the rules to the benefit of a small number of people.
Conversely, the second one further restricts GMC candidates, being a tightening of the rules to be more clear-cut.

At this time, I think the resilience and legitimacy gained from 2) outweighs the additional restrictiveness. Rocket Scientists and other paid contributors can still serve in any composition on the IMC and possibly other future committees. They can also still be on the GMC, just not in a majority with other paid contributors.

Another question is what to do with a change in community status during a term, potentially shifting the committee to non-community majority. For example, if an existing committee member becomes a Rocket Scientist. This is most obvious in the status quo and in 2), but even in 1) it’s an implicit issue as it could mean people wouldn’t have voted for that nominee.
Do we accept current committee members are grandfathered in for the remainder of a term? Or do we call for a new vote if changes shift the committee to a non-community majority?
We could encounter this situation immediately if these amendments are passed, as the current GMC has a team + paid contributor / RS majority.

Bribe resistance

A duty to report clause sounds like a good idea to me. What’s not immediately obvious is how attempted bribery / suspected irregular influence should be handled after reporting. Should we ban offenders from receiving a grants (possibly for a longer period of time) or should this be left up to GMC / community discretion? Will there be a possibility to appeal, e.g. in case of misunderstanding or even a false report?

[calurduran]

Vacancies: I like Pieter’s proposed process for vacancies. My only fear with holding an entirely new slate of elections every year is the chance that you have complete replacement of the committee and thereby lose institutional memory, but I also think if that happens it’s a sign that the community was quite unhappy with the previous group, or that no one wanted to run for re-election.

Community status: I’ll argue for the option of just removing the non-Team/oDAO clause from the RPIP. It is notable that the community had its first chance to vote for GMC, and the Team member finished first, with the three RS members finishing in the Top 8. If we had the stricter version of the rule in place we would have had to turn down a community-supported candidate due to their role as a Rocket Scientist. I’d prefer to just have people declare any conflicts or payments or oDAO roles and then let the community decide who they want to be on the committee. I become very uncomfortable with the idea that one day we might have to kick someone off the committee or hold snap elections because the Team decided to give the RS title to a GMC member, thereby triggering the clause.

[Pieter]

For vacancies, I think we can move ahead with drafting a RPIP edit based on the process I outlined.

Bur first, on community status, it seems we don’t have a clear consensus on the preferred course of action yet. So let’s do a poll. I’ll recap the problem and the main points that have been argued.

Problem statement

The 'shared oDAO seats' concept, introduced for the upcoming Rocket Scientist node, introduced unforeseen consequences to the clause below:

• The GMC SHOULD be composed of a mix of core team and community members, and SHALL at any point in time have a majority of the membership be community members
• Individuals that receive a salary from Rocket Pool Pty to work full-time on Rocket Pool or who are members of the oDAO SHALL NOT be considered community members for the purpose of this RPIP.

A strict reading would now include all Rocket Scientists as oDAO members. This limits the amount of Rocket Scientists that can be on the GMC. A ~13 way split oDAO reward, while by all means significant, does not correspond to a full-time salary. Discussion was had on if and how the clauses above should be amended to provide more clarity.

Alternatives

Several partial / subjective approaches were discussed. For example, nominees self-reporting their community status or allowing up to x% of a FTE compensation to still count as community. These approaches come with difficulties in case of disagreement and are complex to apply. Current thinking leaves us with two 'all-or-nothing' alternatives:

• Remove the team / community clauses and just leave it up to the community to judge nominee’s conflict statements.
• Stricter clauses: classify anyone receiving any financial compensation via team or oDAO as non-community for this clause.

Arguments for removal of the clauses:

• Larger pool of available candidates to form a GMC - active contributors like Rocket Scientists are well qualified to be on them.
• Better stability of the GMC (doesn’t trigger people having to leave or even a mid-term election as people are promoted to RS or become involved with the team / oDAO.)
• Voting still up to community preference: recall that the first vote ended with a team member in first place and three RS in the top 8.
• GMC is not a place one would expect disagreement between oDAO, team and community in general.

Arguments for stricter clauses

• Even small gifts / compensation can substantially affect behavior.
• oDAO rewards, even split 13 ways, are still significant at current value. If RPL appreciates, even more so.
• This is not too limiting on RS / non-community participation, they just can’t be a committee majority.
• Possibly better perceived legitimacy of governance by having hard rules in place that don’t play favorites (also as a precendent looking forward to new future committees.)

Poll

Without further ado, please cast your votes:

• Remove team / community clauses
• Strict clauses: any financial compensation = non-community
• Other (please comment)

[ramana]

I would amend the clause to clarify that it means full oDAO seats, or shared oDAO seats that are commensurate with full-time employment, but not smaller share oDAO seats.

[epineph]

While I previously made the case for strict clauses , I don’t feel strong enough to vote one way or the other on this- i think both options are reasonable, and both remove the ambiguity of current wording- thus limiting future disagreements. If i could vote for both, i would.

[calurduran]

I’ve come around to just removing it entirely. I agree that any attempts to split the baby by specifying things like “equivalent to FT employment” will still leave it up to the community to adjudicate if there are disagreements about whether a person fits that definition or not. And if we’re trusting the community to do that then we should also trust the community to be able to decide how much they care about having a majority of GMC be RS or team members or other oDAO-affiliated people.

Mostly I would be opposed to any change to the clause that prevented us from having a majority of the committee be Rocket Scientists. It feels like an own goal to limit the number of RS’s - who have achieved that role specifically because of their service and dedication to the protocol.

[dietz]

I agree with your comments.