COMP community governance is not well organized. Multi-billion dollar protocols should operate better than this. While it seems like more individuals have been stepping up lately, there are bigger steps we can & need to take.
I wrote this up and am curious what others think. This is my articulation of the issues and possible solutions (and some case studies), but with ideas & feedback from: Getty Hill, arr00, Larry Sukernik, Mike Reinhart, the Polychain team, Robert Leshner, and others community members. Hopefully this discussion will help lead to some proposals/action.
Some great points made here. At Gauntlet, we’ve been working to share some thoughts on an asset onboarding framework, and plan to do so by EOW.
One note to possible contributors - it’s totally possible to have a narrow focus. One hard thing about jumping in to the protocol is where to start. If you have a clear plan for what you could do a as a governance facilitator, by all means, jump right in. However if you are wondering how you could help, focusing on driving value and taking ownership over one domain of the DAO’s efforts might be a good way to bootstrap the working groups mentioned in @JacobPPhillips’s doc.
Jacob your report seems to align strongly with some of the issues voiced on the community call yesterday. One critical issue I picked up on was around protocol governance and the grants program, specifically the balance between agenda and agency. As Getty mentioned, it is difficult to both gauge and enforce the caliber of contributors who pick up grants, however the frequency of grant adoption and execution is an important factor to maintain for the protocol’s growth. At a glance, hiring a governance facilitator, or general team centralization might be an apt way to solve this, to add culpability and oversight to “part time” contributors. However, if not structured delicately enough, could also lead to over-centralization risks and dilution of user voting power, potentially causing the protocol governance to function virtually like a public tech company.
I like your idea of building out a specialized team in charge of “serving community-driven initiatives” and hiring/overseeing the devs which execute it. I think an important issue as mentioned is culpability, and so structuring development teams with deliverables and specialized oversight would be effective to mitigate improper resource allocation.
I think there are some elegant solutions to balancing team efficiency, with some centralization, whilst democratizing and incentivizing voting activity amongst the larger community. Perhaps this could take place in conjunction with the user-whitelisting proposal. A specialized full-time oversight team, for example, could work intimately with whitelisted users - who’s proposals the community votes on - and see through the execution of their initiatives. A governance facilitator could then be in charge of providing consistent concise (and jargon-translated) community reports on those teams’ progress to the broader community, and users can vote on their satisfaction (with perhaps some comp incentive). This all leads to providing more metrics for larger voters to better track progress in the protocol.
I can’t think of a more important discussion to have right now. Thanks for bringing it up @JacobPPhillips.
I agree with pretty much all of the points mentioned in the writeup. That said, maybe I can chime in with some thoughts based on my experiences running Compound Grants for the last few months. Some of the thoughts may be redundant to what was previously mentioned; my intention is to share similar ideas from what is possibly another point of view.
The protocol needs vision. If Compound’s treasury is putting significant amounts of money to work, it needs to do so with an explicit long-term goal in mind. In my mind, the question that needs to be answered is “what does the protocol want to be in ten years?” If we can find an answer to that long-term goal, we can develop medium-term goals that achieve the long-term goal and short-term goals that achieve the medium-term goals. None of this is rocket science of course — that’s how companies, governments, and other organizations get stuff done! You can’t build a skyscraper by randomly laying down bricks; you need to have a blueprint that tells you where to place the bricks.
The protocol needs to hire the best people. Once the protocol figures out what it wants to be in ten years, it needs to recruit the people to make the vision a reality. In common parlance, that’s called recruiting talent. More specifically, the protocol needs to recruit people who will build the thing based on a blueprint. To date, Compound Grants has funded part-time contractors (“contributors”) to work on fairly minor improvements to the protocol. While these efforts go a long way, I don’t believe they’re enough. If the protocol wants to be great ten years from now, incremental improvements are not enough. What’s needed are major improvements, which I believe are best developed by people who are motivated, capable, and determined. In short, I strongly believe the protocol should hire the best people to work on major improvements on a full-time basis. I should also note that top talent is not cheap; if the protocol has the money (and it does), it should be paying above-market salaries to attract above-average people.
The protocol needs organization. After the vision is defined and the people are hired, an organization needs to be formed. Motivated, capable, and determined people working in an uncoordinated fashion don’t get stuff done — you can swat a fly with a finger, but if you want to take down a wolf, you’ll need to clench five fingers into a punch. Coordination and organization is how countries are formed, cities are built, and products made. I think everyone will agree that right now, protocol contributors are uncoordinated and disorganized. If the protocol wants to win big, it will need to form an organization around the talent.
Should the protocol continue doing what it has historically done, I’m afraid we’ll end up in a situation J Cole sung about: the good news is you came a long way; the bad news is you went the wrong way.
Thanks for bringing this issue @JacobPPhillips. As a new member to the Compound governance forum, I had the following issues:
No centralized location/summary of governance process – Any new member should have access to governance handbook/overview to quickly catch up to the process.
Vision – Went through the different topics @ comp.xyz and compound.finance website, and I could not really fine what the future vision or goals (both short/long term). Having a vision will make everything easier in terms of governance, planning and organizing. All future proposals will be geared towards accomplishing our vision/goals.
During my first community meeting, members were discussing collateral factors for the newly voted Markets. The process seemed very casual (although it appeared everyone was on board with suggested collateral factors) with no formal guidelines/procedures. Compound is known in our industry/community for being the most “conservative” and rigorous interest rate protocol, yet this only seems to be true on paper.
My suggestion is creating a Compound Governance Handbook. This handbook will establish our objectives/goals for the future, governance process, new market eligibility requirements, collateral factors for supply/borrow side and other items. Feel free to edit with google docs (Google doc. See draft below:
Compound (Protocol) Governance Handbook:
Motto:
Compound (protocol) is an algorithmic, autonomous interest rate protocol built for developers, to unlock a universe of open applications.
Objective:What is our objective/goals?
Short Term
Long Term
Governance:
Overview: The Compound protocol is governed and upgraded by COMP token-holders, who may delegate their voting rights to addresses. Proposals can modify system parameters, support new markets, or add entirely new functionality to the protocol.
Proposal: Addresses delegated at least 65,000 COMP can create governance proposals; any address can lock 100 COMP to create an Autonomous Proposal, which becomes a governance proposal after being delegated 65,000 COMP.
Delegate: Delegate votes from the sender to the delegatee. Users can delegate to 1 address at a time, and the number of votes added to the delegatee’s vote count is equivalent to the balance of COMP in the user’s account. Votes are delegated from the current block and onward, until the sender delegates again, or transfers their COMP.
Proposal Threshold: The minimum number of votes required for an account to create a proposal (currently 65,000 COMP) can be changed through governance.
Voting: When a governance proposal is created:
Enters a 2 day review period;
After which voting weights are recorded and voting begins;
Voting lasts for 3 days; if a majority, and at least 400,000 votes are cast for the proposal, it is queued in the Timelock; and
Implemented 2 days later (changes takes at least one week)
Market Eligibility Requirement:
Market’s Safety Characteristics
Market’s protocol has been reviewed and audited by independent auditors and/or consultants and has no known security vulnerabilities, including critical bugs, undue exposure to 51% attacks
Market must have sufficient liquidity across a variety of trading platforms
Market’s Supply and Volume Characteristics
Market capitalization of $100 million or above
Supply to be reasonably predicted over the next five years
Market’s descriptive Characteristics
Secured digital bearer instrument
Freely traded market and can be freely held for foreseeable future
Must not be considered a security by the corresponding authorities across different jurisdictions
Market Protocol is decentralized
Included by two or more third party custodian
Other
Market protocol must have significant usage and community
Collateral Factors for New Markets:
Collateral factor can range from 0-90%, and represents the proportionate increase in liquidity (borrow limit) that an account receives
Collateral factors can be increased (or decreased), as market conditions change
Starting collateral factor should be the average of the market in at least three other similar interest rate protocols