[Forum Poll] Exploring Medium-Term Initiatives with GovAlpha

Governance
1

Exploring Medium-Term Initiatives with GovAlpha: Seeking Community Feedback for Prioritization

As GovAlpha pushes forward with various governance initiatives, we’re keen to involve you in directing our efforts for the medium term. We’ve described some options below, and we would appreciate a moment of your time to vote (poll below) or comment.

1. RPIPs Portal Revamp

Objective: Modify the RPIPs portal (https://rpips.rocketpool.net/) to better meet the most common use cases in the community.

Reasoning

As it stands, the RPIP portal is more or less a direct clone of the EIP portal. While this was the obvious starting point, there are differences in requirements between the EIP process and the RPIP process.

The portal can better support the two main use cases of an actively voting DAO:

  • Low-friction access to a list of binding RPIPs (living, final).
  • Low-friction access to a list of the most ‘current’ RPIPs (active votes, recent drafts, recently updated living).

We can also throw some branding and such up there to better match the other official Rocket Pool sites.

2. Snapshot Vote Templates

Objective: Standardize the language and format for snapshot votes for better comprehension and participation.

Reasoning

We have experience (doing this at MakerDAO), and there are a few low-hanging fruit here in terms of vote UX. Specifically:

  • Ensure that the summary lines displayed on the snapshot proposal list page summarize the vote effectively.
  • Ensure that outcomes are clearly and consistently present in the vote text.
  • Ensure that the process that has led to the vote is transparently communicated.

We would aim to keep templates fairly minimal. We went a little overboard with the Maker templates, so would be looking to find a happier medium between clarity and brevity.

3. Redraft RPIP-4 - Community Resolutions and Voting

Objective: Resolve the issues of ambiguity in RPIP-4 that are currently impacting Rocket Pool governance. Note this would be separate from the effective stake discussion, and more focused on having an approachable RPIP for governance procedure.

Reasoning

Again, we’ve done drafting of improvement proposals before, and have a strong understanding of the need to be as clear and unambiguous as possible.

As it stands, the wording in RPIP-4 could be improved, with the goal being the avoidance of similar debates over ambiguity issues in the future.

4. Updating RPIP Types and Statuses

Objective: Update the options of statuses and types to better suit Rocket Pool.

Reasoning

We found the current options for Type and Status of RPIP are not optimal for Rocket Pool’s governance. We believe the reason for this is that they were forked wholesale from the EIP process, which has different requirements than Rocket Pool governance.

To give some concrete examples:

  • An ‘Obsolete’ status would make it easier for readers to identify superseded RPIPs.
  • ‘Informational’ does not fit well as a type of improvement proposal, information is not a proposal, nor is it voted on. Existing informational RPIPs should be moved to documentation portals.
  • ‘Review’ appears to be unused currently, but is a useful status. Potentially could be renamed or usage encouraged more highly.
  • ‘RPRC’ appears to be largely unused, we would consider removal or merge with an existing type.

5. Propose Rocket Pool Operational Process

Objective: Create a formal operational process structure for the Rocket Pool DAOs.

Reasoning

As more responsibilities are moved to the Rocket Pool DAOs, more operational processes will be required. Currently, there is no strong way to distinguish, record, or structure these operational processes separately from RPIPs.

Some examples of existing operational processes are:

  • oDAO membership votes.
  • GMC / IMC elections.
  • GMC Adminstrator elections.
  • Grants Appeals.
  • Core developer funding.

These represent reoccurring processes (many of which involve token votes), that are not ‘improvement proposals’, they are a maintenance cost of existing structures.

Separating these processes creates clarity and allows better prioritization of time by voters.

6. Governance Participation Recommendations

Objective: Create a set of recommendations aimed at increasing participation in Rocket Pool governance.

Reasoning

This is a problem that is widely experienced across the crypto space which is very difficult to solve, but that is essential to build legitimate and effective long-term DAOs.

Using our time working in MakerDAO and @prose11’s Political Science background, we would seek to identify the current issues causing a drag on participation and convert this into actionable recommendations.

While Rocket Pool has a strong foundation in their voting history, we believe participation in decision making (which often goes beyond casting votes) is an area where we could provide tangible value.

7. Community Participation Recommendations

Objective: Create a set of recommendations aimed at increasing value-adding participation in the Rocket Pool community.

Reasoning

Similar to the above, this is an issue that is widely experienced and can be difficult to solve.

We think we could provide some valuable input on the community side of the DAO as well. While intertwined with governance, community participation is less about decision-making and more about identifying how Rocket Pool can better convert interested parties into contributors to the DAOs.

8. Decentralization of Permissions

Objective: Decentralize permissions to reduce single points of failure and improve coverage.

Reasoning

The DAOs currently rely on the Rocket Pool devs to moderate community coordination spaces. Safely transferring these responsibilities to the DAO:

  • Would further empower the community
  • Could result in increased moderation coverage.
  • Would reduce the workload of the Rocket Pool developers.

We’d like to recommend options for decentralizing these sorts of permissions based (in part) on our experience during Maker’s decentralization .

9. Facilitate Strategic Discussions

Objective: Hold bi-monthly calls or workshops to discuss progress towards Rocket Pool’s strategic direction.

Reasoning

Members of the Rocket Pool community appear to lack clear agreement over instrumental goals for the Protocol and the DAOs.

We think these conversations could help engage the community and lead to a focus of efforts towards Rocket Pool’s strategic goals.

What are your thoughts on these initiatives? We’re particularly interested in your views on which of these are most urgent and currently unaddressed. The results of the poll below will inform our course of action.

To help us gauge community interest, please select all the initiatives listed above that you would like to see us tackle.

  • 1 - RPIPs Portal Revamp
  • 2 - Snapshot Vote Templates
  • 3 - Redraft RPIP-4 - Community Resolutions and Voting
  • 4 - Updating RPIP Types and Statuses
  • 5 - Propose Rocket Pool Operational Process
  • 6 - Governance Participation Recommendations
  • 7 - Community Participation Recommendations
  • 8 - Decentralization of Permissions
  • 9 - Facilitate Strategic Discussions
0 voters
3 Likes
2

Nice work here. I feel like some overhaul of the governance/RPIP system will be helpful going forward. Those of us who have been here a while get it, but it can be hard to decipher it all for new members. Enough so that I fear they just don’t bother keeping up.

2 Likes
3

#3: Just for clarity, this refers to clarity beyond the effective stake discussion ongoing, correct? If so, would appreciate that being explicitly called out.

#6: While I agree that more participation would be even better, I do want to point out that we’re outlier good here. We have a very high 15% of vote quorum (which has yet to fail), and we near/pass 2x quorum when there’s excitement or vigorous debate. We also have active research chat on discord (over 90 people were in the RPL staking thread) and active folks on the forum. Still worth energy, just want to frame this as “build on a strength”.

#9: I find live interaction with many people to be a particularly challenging way to make progress. Also, on a purely personal note, I likely wouldn’t participate - it’s only asynchronous comms that enable me to be nearly as active as I am.

2 Likes
4

Appreciate these call outs and have made a few edits.

This is often true, particularly with calls that are high cadence and less organized. It is our belief however that a big benefit can be gained when people set aside time to come together and get to the bottom of thorny issues. Our intention would be to make these activities as public as is possible, so those like yourself who are not called to attend could still benefit from discussion they generate.

Happy to hear more prospective on this, perhaps this is a shared hesitation from the community that suggests our efforts would be better placed in a different objective from the list!

1 Like
5

Small bump for visibility. I think we’d appreciate more direction here. If you have a moment, please consider voting in the poll at the end of the first post.

(as a minor preview, here’s the staging version of the RPIP Portal revamp)

2 Likes
6

Thanks for this post!

I am most interested in #6 gov participation, #9 strategic discussion, #7 community participation, #5 operational process (but less elections and more day-to-day). Some thoughts, largely informed by my experience as a GMC member:

  • GMC members are time-constrained volunteers and there’s some sense that the GMC is not fully fulfilling our mandate, but also has no clear path to getting there. This is thematic for the entire DAO, although in other areas I think we have a bit more “heroism”, sometimes.
  • What the GMC currently is: a machine that (reactively) processes grant applications. What a high-functioning GMC could be: a (proactive) accelerator to the RP protocol and ecosystem,
  • Grants and bounties are one dimension of funding improvements, but as we’ve seen with the GMC admin hire, other approaches unlock new pDAO capabilities and they are largely unexplored.
  • There is no clear process around pDAO strategy. The pDAO has limited resources, it seems wasteful that we are not investing in a process that optimizes and aligns areas of investment. I suggested a committee: Roadmap & Strategy Ad Hoc Committee
  • So many contributions to progress are informal, hard to track, subjective, etc. Not sure if there are better ways we could recognize and incentivize contributions (https://sourcecred.io comes to mind but probably has problems).

In short, we have financial and community resources at hand but there are a lot of obstacles to fully leveraging them. I get that DAOs can be chaotic by their decentralized-nature, but I think that means the potential for improvement is fairly large. Thanks again, looking forward to seeing what comes out of this.

4 Likes
7

I do think the area this is trying to address is one worth looking at, but I also hear Val’s concerns about the particular suggestion. The team is in Australia and we have lots of community members in the US and Europe, so just finding a time slot that works for everyone is challenging.

1 Like
8

This is really interesting because it matches some of the general feel I’ve been getting over the last few months. I feel like the pieces are here, but they don’t quite fit together the way they should yet.

I’m moderately sure that you can handle goal-setting and strategy on a DAO level via voting with the right structures in place. We were very close to implementing something like this in Maker, but never quite got the time to prioritize it. A strategy committee is an easy and quick alternative, and likely shares most of the benefits.

We used SourceCred at Maker for a long time. It was moderately successful at first, though became less so over time. We also made the decision to exclude individuals paid by the DAO from the SourceCred distribution. I do think this was the right choice to avoid double-dipping, but since Maker had a habit of hiring all the effective community members to do structured work under a Core Unit, it left the SourceCred distributions open to gaming by some individuals. [Tagged Posts in Maker Forum]

With regards to this, I think what is missing is a quantative communication channel from pDAO governance to the GMC. Basically, there should be periodic votes in which the pDAO can provide a signal on the following:

  • What are the high priority grant areas for the next 3/6/12 months?
  • Should the GMC increase or reduce focus on the high priority areas over the next 3/6/12 months?

The problem basically comes down to direction: The GMC wants to know where and how much to focus their efforts. Only the pDAO can tell you this, but as-yet there is no structure or voting process to facilitate the communication of that signal.

Of course, the problem is bigger than this, because the pDAO doesn’t really know what it wants, or what’s important either. So the process needs to include a stage in which discussion happens, arguments are made for various focuses and vote-options are determined.

2 Likes
9

Interesting. I see the community discussion threads (eg September 1-9 2023 GMC Community Discussion of Submitted Applications) as a great place to provide the GMC with input. I’ve seen input about specific items and also general topics (eg, governance, documentation, dashboards…). The “results” threads, GMC nomination thread, and GMC selection threads also offer opportunity to provide feedback.

The issue I see seems more akin to what you touched on in your last paragraph – the pDAO doesn’t know what it wants. However, I don’t see much participation in terms of figuring that out either – instead that steering has (mostly) been delegated to the GMC. I suspect that if we discussed enough to be able to vote, the GMC would have gotten enough of a pulse even without the vote. (I will happily volunteer to help get a signal vote to happen if we have some idea what we want on it though)

2 Likes
10

I voted for 2, 5, 6, 7, and 9.

Personally I hit a brick wall with the stipend discussions. You and prose11 raised some really good ideas. I believe the general consensus is compensation is necessary, but no one really agrees on how to do it. Given the conflicts of interest from the GMC, a third-party coordinating this problem could have tremendous value.

While I can see the potential value of this in the future, for now, it doesn’t seem immediately necessary. We’re not constrained by budget concerns that force us to prioritize specific grant categories, and we also haven’t been around long enough to have a history of funding decisions in different areas. I understood Dondo’s comment as suggesting we focus on outbound marketing or initiatives to attract more applicants first. Once we’ve achieved that, we should definitely consider a more targeted approach to grant allocation.

1 Like
11

As a minor note (since Dondo and ShfRyn touched on it), I’ll note that my anti-stipend and anti-steering committee stances are softening over time. I do understand that we have some real challenges with engagement, and maybe “try something” is better than a holding pattern – even if I see pitfalls.

2 Likes
12

Thank you for reaching out!

The community (and team) have learnt a lot on our governance journey but it is always good to hear from experienced people, as it is often hard to know, what you don’t know.

Here are my thoughts:

5#
Each management committee has autonomy over their charter. Any help we can give to refine their internal processes based on experience / best practice, that respects their bandwidth would be ideal.

In terms of the existing processes listed, i am not sure what you mean by separating these process? Any help to make these clearer and straight-forward would be great but they are generally escalated to pDAO vote because they cannot be done at the committee level.

6#
As @Valdorff said, we have a good record in terms of vote participation. We would obviously like to keep it that way, so advice is welcome on this front. We have a good process for communicating and engaging the community around proposals. In saying that, getting information disseminated is always a challenge so advice here would be welcome.

7#
We rely very much on a small number of dedicated and passionate individuals. I do think it is important to save burning these people out and fostering wider community involvement. Ideally any solution would be a mixture of extrinsic and intrinsic incentives.

9#
We have started to develop a core team prioritisation process that includes strategic themes or research areas. We could easily expand this out to include the DAO outcomes. Again, advice on how to structure/improve this process would be helpful.

Like any complex system, i suspect Rocket Pool’s governance may never meet our expectation of complete. I would argue most governance systems are messy to some degree. Although it is important that we learn and evolve using the experience of others.

3 Likes
13

We’re not proposing that the processes that currently require a pDAO vote would no longer require one. More formalizing that these processes take can require votes without an accompanying RPIP, but that any other requirements for these operational process votes are recorded in a structured and clear way.

The general idea being that operational processes aren’t improvement proposals, and including them in that framework subverts expectations and assumptions around that process.

I think including the core team’s prioritization process to include DAO outcomes is in many ways the opposite of what you want. I don’t have a strong grasp of the prioritization work that has happened so far, but the implication there is that it would be led by the core team.

General advice would be to help and encourage members of the pDAO to develop this process without significant core-team input. It sounds grim, but at Maker we literally had: “We’re shutting down in six months, the DAO needs to work without us at that time.” I don’t think that is ideal, but the responsibility for function needs to be firmly shoved onto the DAO in order for the DAO to actually develop as its own thing.

At a high level I’d recommend the DAO-agent flow look something like:

  1. DAO decides its priorities in an agent-agnostic way.
  2. Agents view agreed DAO priorities, determine which of these they are able and willing to fulfil and make proposals to fulfil one or more priorities.
  3. Some time passes to allow a range of proposals from various agents.
  4. DAO determines which proposals to accept (by vote or committee), and agents are funded.
  5. Agents do work.

Obviously that’s very high-level, and is sort of a ‘perfect world’ flow. Currently RocketPool has elements of this in the form of the grants program, but is missing step #1, leaving agents to either:

  • Work with the DAO on a per-agent basis to determine priorities for that agent (not scalable)
  • Guess at the DAO’s priorities based on conversations with members and general observation (not efficient)
  • Gaslight the DAO into believing what the agent can deliver is what the DAO needs (actively harmful)
14

I don’t want to minimize the importance of this, because it is a hugely important step in the process. I believe it should be the first of two steps, though. With the second being the sort of voted-priorities I described.

The reason for this is that without this second quantative step, communication is still unclear. How much weight is a committee or an agent supposed to put on the two individuals responding to their grant / disucssion / etc? What if five take part, and it’s a 3-2 split? As much as these conversations are important to highlight key issues, provide direct and detailed feedback, and contribute to the ‘aliveness’ of the DAO, they are very difficult to justifiably use as direction.

So, once you have these conversations, you package the discussed options up into some sort of vote, and have the DAO vote on it. Now the committees and agents can justifiably use this as direction, because they know to what extent the DAO actually agrees with the handful of active voices on the forum.

This also provides committees and agents with a level of ‘cover’ in the event of unhappiness over specific decisions. A good example being grants. If your grant is refused, and the GMC says:

“We’ll there was a discussion in thread x on discord, and three people talked it over and the vibe was that the area your grant was in isn’t very important, the GMC read that and thought it made sense, so we refused that grant.”

…then this is infuriating! What if you didn’t see the conversation? What if you find four people that agree that it is important? The GMC screwed up! Except they didn’t screw up, they just aren’t being given the quantative signal they need to do their job effectively.

In the above scenario, if the GMC can point to a vote and say: “We like this grant in area X, but the pDAO has communicated a desire for a tight focus on area Y rather than area X.” then sure, maybe that’s still annoying, but its hard to say its unfair, especially if you were previously made aware of the outcome of said vote.

More likely, that grant would not have been proposed in the first place, or the grantee would have tailored it to better match the publicized priorities agreed upon by the pDAO. Potentially both saving the GMC’s time, and/or resulting in the pDAO getting more of what it wants, and less of what it doesn’t care about.

15

Hmmm… sounds like we might have a difference of opinion in terms of autonomy and trust. To me, the pDAO is selecting GMC members that we trust to “harness the community’s talent to further the goals of the protocol”.

Using this thread as an example, prioritizing 9 subject areas is non-trivial. I don’t expect we can get good quantitative results from 500 people, at that point we’d have a lot of voting based on the titles. I believe that conversations with involved people may well be a clearer signal than a quantitative vote. Ofc, if the GMC want to get more feedback and believe a snapshot is effective - let’s do it. Or if the pDAO want to push some feedback and believe that’s effective - ditto.

Insofar as cover goes, I hope we don’t need it. We delegated power to the GMC members. They’re exercising their power as best they can. We may disagree (and can vote for overturns where we see fit; we can provide our thoughts), but we shouldn’t be attacking the GMC or their members such that they need to defend themselves.

3 Likes
16

Possibly, I’m not sure the difference in opinion is on autonomy and trust. For me its more in this:

“harness the community’s talent to further the goals of the protocol”

GMC is given power and autonomy to further the goals of the protocol, but right now, I’m not sure those goals are well-defined enough for them to be effective at their role? (Though I would defer to the GMC on that).

Like, if you have high-level value-goals determined, and you have low-level grant-funding, you’re arguably still missing a middle layer of instrumental goals that represent ‘the current plans to achieve the high-level goals.’ I’m not sure producing that sort of mid-tier strategy is in the GMC’s role? Though I would say they have a role in actioning such plans through funding of grants.

Totally agree. In practice I’ve already seen it happening, and to some extent I believe it will always happen. We’re not even at a point yet where there is a limited pool of funds.

17

Appreciate all the votes that came in here. A quick update on what we’re working on.

  • Propose Rocket Pool Operational Process Framework - Draft Complete [forum post]
  • Governance Participation Recommendations - Started
  • Community Participation Recommendations - Not Started (Might merge with governance participation recommendations as there is a lot of overlap.)
  • Snapshot Vote Templates - Started
  • Facilitate Strategic Discussions - Not Started, may not pursue.
  • RPIPs Portal Revamp - Complete [Link] (Was done out of order because it looked fun)
  • Redraft RPIP-4 - Community Resolutions and Voting - Not Started, may not pursue.
  • Updating RPIP Types and Statuses - Kind of started. It’s getting done piecemeal as part of quick-wins on RPIP content.
  • Decentralization of Permissions - Not Started, may not pursue.
4 Likes