My history with Ethereum goes back to 2012, when I emailed Vitalik with the idea of "smart contractified companies" and got a same day (enthusiastic!) response.
In the following years, when I moved to Singapore, I spent a lot of time with the early Ethereans incl. Gavin Wood and for a while I was a Director in the Ethereum Foundation's (EF) Asian chapter.
The early Ethereans were still very accessible then and the EF was not a rich organization, so V and I teamed up looking for extra funding and budgeting for the next researchers he wanted to hire.
It was after Devcon Shanghai in 2016, where I first met Ming Chan, then Executive Director of the EF (who Laura Shin much maligned in her 2022 "The Cryptonians" book) that I started to feel things were getting too political and I'd rather just build on Ethereum.
The paradox at the heart of Ethereum
I felt then and still feel that there is a paradox at the heart of Ethereum between its values as a decentralized, permission-less technology and its centripetal, syndicate-style governance.
Plato's ideal form of State, a Republic lead by philosopher-kings, functions when its leaders are enlightened and act with integrity, and in all of the recent outburst of frustration within the community about how Ethereum is governed (see e.g. @MarkBeylin's "Ethereum's Social Layer is Broken" or @licuende's "Ethereum and its inner contradiction"), nobody including myself questions V's integrity.
However, for Ethereum to truly live its ideals in the next phase of its adoption (the next billion!), it needs to find a new equilibrium between its technology and social layer.
In this post, I want to propose a tentative design, based on the hundreds of Web3 governance experiments I've been involved with on the Otonomos client side and as Lead of two decentralized projects myself, otoco.io and a more recent project for a decentralized school.
The mushy social layer
Our starting point is that pure onchain voting is not only undesirable but also delusional, even for protocols with a high degree of technical decentralization which in theory could limit their governance to coin voting.
The "mushy human intuitions and feelings" should not be expunged in favor of completely algorithmic governance, as per Vitalik's 2017 "Notes on Blockchain Governance" and reinforced/refined in subsequent blogs on governance, crystalized in a 2021 post titled "Moving beyond coin voting governance".
But passions run high (especially when money is at stake) and animal spirits quickly take over from rational debate, as we've seen recently with the calls on X to remove the present Executive Director of the EF (I won't link to some tweets here).
People can tweet what they want ("as long as it is legal" as per Musk's X censorship policy) but when it comes to giving people an actual vote, a degree of structure around the venue and forum where governance decisions are taken (vs. merely debated) seems inevitable.
A bicameral model
We propose a bicameral model to introduce checks and balances in governance, with a first chamber acting as a Ethereum's townsquare and the second chamber consisting of an elected "Council of 100" (say).
As with democracy itself, bicameralism may be the worst idea except all others. However, combined with mechanisms such as proof-of-humanity (i.e. verifying individual voter identities via onchain attestations), quadratic voting (encouraging participants to allocate votes in proportion to their preferences) and an element of futarchy (allowing decisions to be made based on prediction markets e.g. using Polymarket), the Ethereum town square could be massively enlarged to basically all ETH holders.
The "Council of 100" would then acts as the representative body for the main stakeholders in Ethereum, such as the core devs (incl. the client teams, the L2/scaling teams, the researchers, infra/tooling teams, security sentinels), delegates from the product teams, the coordination teams (funding platforms, events, media, education, social networks), investment teams, etc.
Each faction could send delegates but in what proportion and how elections could be held is something that would need further study.
In this bicameral setup, the Council would be where Ethereum Improvement Proposals (EIPs) are drafted, discussed, tested and submitted for consensus and ultimate activation.
The EIP process in substance could be broadly preserved however the Council would formalize what is currently a rather opaque system, with intense back channeling in the corridors outside of the Ethereum Magicians Forum which is meant to be the main EIP forum.
Sub DAOs
In contrast with the "Council of 100", which is where EIPs get adopted, the town square would propose and vote on broad decisions related to the Ethereum ecosystem.
By using Sub DAOs, such enlarged town square of ETH holders should not result in unwieldiness or paralysis in governance.
For instance there could be a funding Sub DAO that looks at making research or developer grants, a marketing Sub DAO that appoints a third-party marketing agent such as Etherealize, a lobby Sub DAO (to V's horror!), a DevCon Sub DAO, etc.
Crucially, such sub-DAOs are not mere working groups (debating clubs) but can take decisions that bind the community, e.g. on how its treasury is managed, how (if?) marketing is done, how DevCon should look like, even what can be voted on in the town square - as long as a proposal to abolish the sub-DAO that vets proposals can never be vetoed by that sub-DAO!
Day-to-day decisions
Anybody who has ever built a business or lead a projects knows that every day, dozens if not hundreds of small decisions need to be taken to keep things going.
Believing any form of executive power and hence agency risk can be removed is simply naïve, however here too there are mechanism to prevent human capture.
These include the choice of a legal structure that does not require a Board of Directors nor has any form of shareholders, but can be ran by mere administrators under transparent remunerative arrangements.
This is why I'd personally like to see the EF decant itself into an Association in Switzerland or even in Wyoming, where the recent Decentralized Unincorporated Nonprofit Association (DUNA) would in many respect be superior to the staid, rigid Zug-based EF with its Swiss-resident Director(s) and its obligation to meet at least once a year in pretty but parochial Zug.
EF > EA
Such "Ethereum Association" (EA) would preserve all the advantages of the current Ethereum Foundation, most prominently its ability to make grants, but have the additional advantage of:
- being run only by an administrator(s) without the need for any Director or Board;
- extend membership and corresponding benefits to each and every ETH holder by what is legally called implied conduct, i.e. holding ETH in your wallet means you basically get a membership card, gas fees can be seen as a payment for membership, and all sort of rights and privileges can legally be bestowed on members, first and foremost the right to vote.
- the non-profit nature of the Association would mean it is exempt from taxes, but would not prevent it from owning for-profit subsidiaries.
Associations are as old as your local sports or coding club, and can be created even without formal filing.
The recent Wyoming DUNA goes a step beyond the legacy association structure by gutting all elements of centralization. But if the U.S. would be too contentious as a base of the EA, Switzerland also has an association structure that is used e.g. by professional services firms who create an association as the umbrella organization for their global brand, where shared services are being procured and paid for, with local chapters as paying members who each have their own P&L and separate liability.
An ecosystem at the edge of chaos
The strength of Ethereum is its community, and in order to maintain an ecosystem of such diversity and unbounded energy, its governance has to operate at the edge of chaos with about the right amount of centripetal forces to prevent the system from dissolving into well..., the ether.
Getting this balance right between form and freedom is a challenge for most Web3 projects, but few have as much at stake as Ethereum, if indeed it aspires to be the world computer that puts the world on decentralized rails.
Decentralizing the governance of a protocol with Ethereum's scale and ambition is best done via Popper-style piecemeal improvements, which is why our proposed design is not a radical overhaul but builds on the existing technical and social layer of Ethereum, aiming to reduce the potential for human capture and the emergence of Mafia-type factions lead by "made" Dons.
> Follow Otonomos' founder @hanverstraete on X