The ENS Controversy
A Case Study in Decentralized Governance vs. Immutable Code
On an otherwise peaceful Saturday, a corner of Twitter populated by a fringe collection of developers, day traders, digital artists, and political theorists was abruptly disturbed by a heated exchange of tweets. Unremarkably, the altercation was sparked by a statement that had, also unremarkably, been dug up for renewed scandal. However unfortunate for those who find themselves suddenly under prosecution with little more than the right to a speedy trial, in our present time such events have become commonplace; worth little more than a passing shake of the head, or a bullet pointed rundown such as the one provided by Coindesk a few days later (Reynolds). Nevertheless, the events of February 5, 2022 have ramifications far beyond simple Twitter controversy that everyone concerned with the future of alternative tech and free speech on the internet should understand. The infrastructure is not yet built, but the conversation on how to build it is taking place now, and if we hope to preserve our future on the web we must make our voices heard.
Brantly Millegan, the subject of attack, was director of operations for True Names Limited — a non-profit behind the creation of the Ethereum Naming service (ENS), a key player in the growing conversation regarding web3, an evolution of web infrastructure which claims the ability to decentralize the web and do away with big tech content aggregators such as Facebook, Google, and Amazon through blockchain based systems. As Forbes writer Charles Silver explains, “Decentralized infrastructure and application platforms will displace centralized tech giants, and individuals will be able to rightfully own their data” (“What is Web 3.0?”).
ENS set its sights on creating a fully decentralized web3 protocol to replace the domain naming system of the traditional web. Towards that end, True Names Ltd. established a governing DAO (Decentralized Autonomous Organization) in November of 2021 to maintain ENS and allow token holders the ability to control its future through democratic government. While in the process of ceding full control of ENS, True Names Ltd. maintains a quasi-bipartisan relationship with the DAO. Millegan, as an active stakeholder in ENS, held leadership within the DAO in addition to his role at True Names Ltd.
Anonymity is a respected value in the web3 space, and many choose to act behind assumed twitter names, gamer tags, and faceless avatars. But, as the de-facto spokesman for ENS, Millegan made no secret about his public identity, including his devout catholic faith. Before coming to True Names Ltd. Millegan had completed doctoral work in Moral Theology following his conversion as an undergrad. His twitter bio reads simply: “Catholic, father, husband.”
Catholic doctrine is public knowledge, but in the current cultural climate it’s no surprise that putting it into words would spur controversy. Such was the case when a 2016 tweet surfaced wherein Millegan outlined catholic teaching on homosexuality, transgenderism, abortion, and other contentious issues.
The immediate demands were varied. Many were calling for ENS to disavow and distance from Millegan. Others, among them many who had been involved with ENS for years and knew Millegan as an active and respected community leader, appealed on Millegan to publicly recant his beliefs. However, on a public Twitter space hosted by Morgan Taylor of OnChainTV later that night Millegan reaffirmed his adherence to Catholic doctrine while expressing a belief in tolerance for all people.
“Web3 is for everybody; I really believe that… I really believe in tolerance. I’m willing to work with anybody regardless of who they are, what their beliefs are. But, the idea that like, adherents to the world’s largest religion are not welcome in the web3 industry I think is an extreme position” (Millegan, 2:24).
Some of Millegan’s friends and coworkers vouched for his character and right to his personal views outside of his duties to ENS. Notably, in the face of calls for his dismissal from True Names Ltd., the protocol’s founder and lead developer, Nick Johnson, spoke positively regarding Millegan’s ethical conduct while acknowledging that Millegan’s views were not his own or representative of ENS.
On February 6, Millegan was suspended from Twitter for violating rules against “abuse and harassment.” In a post on Discord, Millegan reiterated his belief that web3 should include tolerance for a wide diversity of beliefs.
A few hours later, Millegan’s fellow community stewards voted to oust him from his role within the DAO. The DAO had been organized into four working groups of five stewards each to facilitate the growth of ENS. For the first term, True Names Ltd. appointed two stewards to each working group. The remaining three had been elected by the DAO. After a five-week appointment, Millegan was removed from the Community Working Group by a majority vote of three stewards. Nine hours later, True Names Ltd. terminated Millegan’s employment.
For the following month, the DAO would continue to debate the future of Millegan’s role in ENS, including his remaining official position as one of three directors of the ENS Foundation (an entity formed to act as the real-world arm of the DAO largely as a shield against legal liability). But beyond ENS, a greater discussion regarding the narrative of web3 and how the ecosystem can and will develop has been taking place. The ENS controversy shines a light on growing disagreement over the meaning of core web3 talking points, including decentralization, DAOs, immutable code, censorship, and regulation.
Since its earliest days, web3 developers have spoken of a revolution to replace established Silicon Valley technocrats and government regulators with decentralized infrastructure and consensus-based decision making. But can we trust that the narrative will develop as originally intended, or should we fear that web3 can, or perhaps already has, been co-opted by those with other designs?
The Importance of ENS to Web3 Architecture
To understand the significance of the ENS controversy it is crucial to recognize the inherent threat web3 poses to established powers. Web3 is not simply proposing a new type of internet platform, but an overthrow of the centralized web in favor of a radically new paradigm. Behind the glitz and glamor of the modern web experience lies a base infrastructure that allows front-end user platforms such as YouTube, Facebook, and eBay to function. Using blockchain technology, web3 aims to rebuild this infrastructure with decentralized, parallel services that would lay the foundation for decentralized applications (dApps) that directly compete with big tech platforms.
ENS is essentially a dApp built on top of the Ethereum blockchain that allows users to register names (or identities) by interacting with coded smart contracts. The ENS protocol owns the .eth top-level domain and issues users second-level domain names (any name followed by the .eth suffix) as NFTs tied to a user’s Ethereum wallet address. This is similar to how the DNS system issues domain names for websites on the traditional web. For example, if Elon Musk were to create a personal website, he might request the second-level domain name elonmusk.com from a DNS compliant registrar. In the same way he can request an identity in the form of elonmusk.eth from the ENS protocol. The protocol will simply ensure no existing name has been registered before and then issue elonmusk.eth in the form of an irrevocable NFT directly to Musk’s crypto wallet.
There are many things that can be done with an ENS identity, including pointing it to a decentralized website hosted on a blockchain based storage service, such as IPFS or Arweave. Since the .eth top-level domain is not recognized by ICANN (the managing non-profit of the DNS system), ENS names cannot resolve to websites on the traditional web, but it is possible to import a traditional DNS domain name to the ENS protocol to get the enhanced benefits of an ENS identity. Imported domains will still resolve to traditional websites, but they are not censorship-proof like .eth names as ownership of DNS regulated domain names are ultimately controlled by the associated domain registrar and ICANN itself.
Some workarounds are available for .eth name holders who wish to link their names to a traditional website, but ultimately there are other decentralized protocols, such as Handshake, that intend to compete more directly with ICANN. ENS was designed primarily as an identity service for the web3 experience.
Identity is an interesting concept in web3. In web2, platforms usually require users to submit to some form of identity verification to interact with their services. In most cases a simple phone or email verification is enough, but more sensitive platforms may require users to provide stricter legal identification such as a photo ID. Identity verification is necessary to ensure users are who they claim to be and to prevent the creation of spam accounts, but it raises privacy concerns and has undoubtably enabled the creation of the data farming industry.
Beyond privacy concerns, the current identity verification model of web2 has balkanized the internet into platform-specific silos. Since each platform conducts their own identity verification, users can’t carry their online identities (including followers, digital clout, content, reputation, etc.) from one platform to another. Worse, users are increasingly forced to self-censor and stay ahead of ever-changing and sometimes morally questionable terms-of-service or risk losing accounts they have spent years building. Even users with the best intentions can find themselves suddenly suspended due to seemingly arbitrary decisions. In the web2 paradigm, users don’t own their own data; platforms do.
Identity verification is begging for an innovation, and ENS is seeking to do just that. Names registered with ENS are designed to act as a universal passport for all web3 based apps. An ENS name will be a singular identity for the web that allows users to carry their reputation and credentials with them wherever they go. If ENS succeeds, users will be able to verify their identity across web3 simply by linking their associated Ethereum wallet.
In the decentralized thought-space ENS inhabits, web users should own their own identities. A cohesive identity provider such as ENS would allow web3 users full control of their identities without fear of de-platforming.
“That is why we believe that Ethereum key-based accounts are the future of authentication…. They provide a unified online identity for users, where instead of holding platform-specific accounts, they could authorize platforms to connect to their unified pseudonym” (“Why ENS Names Are Much More than Domains”).
Web3: Code vs. Decentralized Governance
The vision of a fully decentralized web free from regulatory control has been going mainstream amid growing concerns of political censorship and collusion between big tech and government agencies. Free speech advocates, cultural dissidents, political protesters, and technological revolutionaries all have a stake in the developing narrative known as web3. As the social environment grows increasingly polarizing, the pool of web3 believers only grows larger. In many circles, talk of web3 takes on utopian overtones; an unstoppable force destined to re-build a fairer internet on the pillars of equal opportunity and self-determination.
“Web 3.0 will bring us a fairer internet by enabling the individual to be a sovereign. True sovereignty implies owning and being able to control who profits from one’s time and information. Web 3.0’s decentralized blockchain protocol will enable individuals to connect to an internet where they can own and be properly compensated for their time and data, eclipsing an exploitative and unjust web, where giant, centralized repositories are the only ones that own and profit from it” (Silver).
But the wheel of progress is still in spin, and narratives can quickly change. Behind the promise of web3 lies a battle between competing conceptualizations of the decentralized web and the ways in which it can and should solve web2’s problems. Of these competing narratives, two rise to the forefront: the idea of a web built on uncensorable protocols, locked in immutable code through the power of blockchain; and the idea of a web built on decentralized governance where protocols ultimately answer to the will of their users through DAOs.
The idea of uncensorable protocols has been spinning its narrative since the advent of Bitcoin in 2008; the first implementation of blockchain theory. Without a single overseeing administrator, the bitcoin protocol has been processing transactions as programmed ever since it was released to the internet by its anonymous creator. The protocol accepts and rejects transactions according to strictly coded rules verifiable via a public ledger.
Lawrence Lessig coined the phrase “code is law” to refer to the regulatory power of code to establish norms and values on the internet. Code can be changed, but it’s much harder to enact changes once the infrastructure is in place. Blockchain takes this concept a step further with the principle of immutability
Blockchains are distributed ledger systems that rely on a decentralized consensus network of computers, or nodes, that provide computational power and validate work completed on-chain. Once a node validates a change to the ledger it communicates changes to other nodes on the network. Each node will independently verify the communicated changes before adjusting its internal copy of the ledger. The entire network stores the ledger separately, preventing an unauthorized change to the ledger on one node from affecting the greater network. Code is law, and the immutability of blockchain coding theory means, once created, laws cannot be broken.
“Web 3.0 allows a global commons for state. Rather than state (meaning our data) being held by companies, we create a decentralized network where trust is established at the protocol level. It comes from mathematics and cryptography baked into the protocol. The protocol is broadly called blockchain…” (Brennan).
The idea of an internet running on pure code with no human oversight might sound scary, but that characterization is really only half the story. Behind these computers are very real human actors who choose to provide computing power in exchange for crypto rewards built into the protocol. At any time, a validator can choose to run their own network with an edited protocol, but they will lose out on the network effects of the greater blockchain. This concept of immutable code that relies on decentralized consensus is the heart of web3.
Highlighting blockchain’s immutability, in 2016, when an early Ethereum based dApp simply called “The DAO” (an investing fund utilizing an early iteration of decentralized governance) was hacked and drained of over $50 million of invested capital, Ethereum developers chose to effectively re-start the blockchain rather than let the hacker keep the stolen funds in a move known as a “hard fork.” The hacker had exploited a vulnerability in The DAO’s programing that was code-legal and therefore could not be reversed. A hard fork essentially splits the blockchain into two; a potentially disastrous move as it relies on users jumping ship on the old protocol. In the case of the 2016 hard fork, the community was small enough and the consensus was broad enough that the fork met with little resistance.
The fallout of the hack was overwhelmingly negative for Ethereum’s budding reputation. Some viewed the hack as proof that blockchain systems were unreliable and prone to bugs. However, contrary to popular reporting, Ethereum itself was never hacked. As an application built on top of Ethereum’s base layer, the dApp’s vulnerability affected investors of The DAO fund, but Ethereum holders at large were safe. Risks associated with using on-chain applications are arguably no different than risks associated with any other tech (even Twitter and Facebook have fallen victim to hacks).
More poignant criticism challenged the fundamental claims of Ethereum towards decentralization and immutability if a small handful of core developers could hard fork and essentially “reverse” transactions at will. While this debate continues today, counter arguments point out that the original blockchain still lives on. Validators who disagreed with the decision could, and did, choose not to update their code and instead continue validating the original chain. The Ethereum developers could not reverse transactions on the original chain due to fundamental blockchain mechanics, therefore proving immutability.
“Blockchains like Bitcoin and Ethereum allow people to conduct financial exchanges digitally and almost instantaneously, today, without banks. They do this by creating the global state that is secured not through social trust (accrued over hundreds of years to the third party institution of banks), but in code, through a protocol. In the same way we no longer have to speak to an operator to connect a phone call, blockchain makes financial transactions into a peer-to-peer protocol. Crucially, though, the protocol is decentralized and distributed — involving a global, public chain secured by the entire network of people running it” (Brennan).
As Ethereum grows it will only become increasingly more difficult to gather enough consensus to justify a hard fork for reasons of “reversing” executed code in the future as a split community will be detrimental to all. At least in theory. To echo the words of Lessig at the dawn of web2, to push the anti-big tech button “is not to teleport us to Eden,” and no utopian promises are fulfilled until the code is written (Lessig).
The truth is decentralized governance is much harder to achieve than the decentralization of physical data. While democratic forms of government arguably allow for greater decentralization than is possible in centralized governance, the idea that all participants are acting entirely independent of each other is naïve. Ethereum founder Vitalik Buterin questioned the efficacy of blockchain to withstand collusion attacks as early as 2017:
“Many communities, including Ethereum’s, are often praised for having a strong community spirit and being able to coordinate quickly…. But how can we foster and improve this good kind of coordination, but at the same time prevent “bad coordination” that consists of miners trying to screw everyone else over by repeatedly coordinating 51% attacks?” (Vitalik, “The Meaning of Decentralization”).
Buterin concluded that there is no easy answer to this question and that “tradeoffs are unavoidable” His concession echoes that of political theorists before him such as Alexander Hamilton who, although staunchly opposed to the development of political factions, contended in The Federalist №26 that “the spirit of party, in different degrees, must be expected to infect all political bodies” (Hamilton). As evidenced by Buterin’s reiteration of the problem in a more recent 2021 podcast, there have been no new developments to change his opinion.
“Once you drop the assumption that every single participant is making decisions independently and once you realize that in any kind of system people are going to form subgroups, people are going to talk to each other, people are going to collude with each other… there is no governance system that is completely robust against attacks” (“Vitalik Buterin “Spiritually, I Really Love DAO,”” 16:33).
Still, blockchain governance has advantages over traditional governance structures. Keeping a public ledger of all transactions, contracts, and operations executed on-chain certainly makes it more difficult to cut back-room deals and hide sensitive information. While some decry its lack of anonymity, blockchain protocols are transparent for this very reason. The public ledger allows us to trust the protocol to act as the third-party arbiter in a peer-to-peer network that has no governing authority.
But ultimately democratic governance is nothing new, and a web3 that promises to re-invent the internet on the back of an old horse can hardly be considered revolutionary. Nevertheless, that’s where web3 discourse has shifted in recent years. An outsider stumbling into a late-night crypto twitter space or discord channel might get the idea that DAO governance is inseparable from web3. Every dApp has its own DAO. The idea of building a dApp that doesn’t require a DAO is a seeming foreign concept.
The best methods and practices for running a DAO is a topic of ongoing debate, but most projects issue governance tokens that are bought and sold on coin exchanges. Theoretically, the more tokens a voter holds the greater their economic incentive to vote in the best interests of the project’s success, and granting users a voice in the decision making of the platforms they use is considered a more ideal arrangement than web2 currently affords. Whether or not current DAO structures are truly that superior is debatable, but regardless, shifting away from a narrative of immutable protocols that rely on decentralized consensus to a narrative of flexible protocols that can be modified at will by token holding community members is a significant shift.
The origin of this change in narrative is hard to pinpoint. In early theory, DAOs were thought of as organizations that could run “autonomously” on code without human direction. Whereas traditional corporations rely on a board of directors and managers to plan for the corporation’s needs, a decentralized corporation would theoretically be coded with the ability to analyze its own systems and incentivize help to support its needs through built in economic controls.
“But what if, with the power of modern information technology, we can encode the mission statement into code; that is, create an inviolable contract that generates revenue, pays people to perform some function, and finds hardware for itself to run on, all without any need for top-down human direction?” (Vitalik, “Bootstrapping a Decentralized Autonomous Organization: Part 1”).
However, this is not the narrative driving DAO theory today. The shift in narrative from autonomous code to democratic token governance is likely due to several factors: the recognition that blockchain-based consensus mechanisms are susceptible to the same faction forming tendencies of other, more organized communities, the desire to provide a level of decentralization while a dApp is still under development, and the desire to future-proof a dApp by allowing upgradeability without resorting to centralized ownership.
Future proofing a dApp that has no overseeing administrator is a difficult dilemma. Granting a DAO administrative rights to “update” the dApp’s smart contracts is a solution that provides some decentralization, but the unfortunate tradeoff is that future DAO leadership could potentially alter the code to the harm of its users. No matter how decentralized the DAO is, structured governance introduces new risks (collusion, powerful entities buying votes, hostile takeovers, etc.). Alternatively, since on-chain code is inherently open source, it’s always possible for future developers to create a “better” version of a dApp and, in theory, users will naturally migrate to the new “updated” version. In practice it is much more difficult to migrate users to a new platform, but arguably allowing users the choice to “update” or not is more aligned with web3’s values.
As to the other factors driving the narrative shift, it’s not clear that current DAO theory provides a real solution. Decentralized consensus is fundamental to blockchain design and may always be susceptible to collusive influence, but so is organized governance. Worse, organized governance may provide greater opportunity for controlling powers to exert negative influence than is typical in decentralized consensus models, especially if a DAO can alter a protocol’s code. It’s understandable that developers would want to delay locking the code until all kinks have been worked out, and vesting power in a DAO does at least provide a level of decentralization during these developmental stages, but the end form of a dApp would ideally be an un-updateable smart contract.
At least that would logically be the ideal if web3 is about building an un-censorable internet, and maybe this is where the real shift has occurred. We are not merely witnessing a war over the means to a common goal, but a war over the goal of web3 itself. To many, web3 is about realizing the original vision of the internet as conceived by Tim Berner-Lee of an open-source platform where information could be freely uploaded and accessed without censorship. But to others, web3 is about building a decidedly inclusive space on the internet by shifting power structures into the hands of the underprivileged. Both use the same terms, but their definitions can take on very different characters depending on the values one subscribes to.
There is a silent war between the balance of web3’s reliance on immutable code and decentralized governance, and it’s a war over the values of its developing populace. At the advent of web2, Lessig warned that the finality of coded architecture will decide the values of the new cyberspace (Lessig). It did, and those who fought for an uncensorable web found themselves on the losing side. The code of web3 is still in flux and its final values are yet undecided, but it won’t be that way forever. One narrative will win. We cannot be lulled into false irenicism, or else our peacemaking will allow for subversive interests to overtake the web3 narrative for their own purposes.
“Our choice is not between “regulation” and “no regulation.” The code regulates. It implements values, or not. It enables freedoms, or disables them. It protects privacy, or promotes monitoring. People choose how the code does these things. People write the code. Thus the choice is not whether people will decide how cyberspace regulates. People — coders — will. The only choice is whether we collectively will have a role in their choice — and thus in determining how these values regulate — or whether collectively we will allow the coders to select our values for us” (Lessig).
ENS: A Case Study in Party Spirit
The collision of these two, competing web3 narratives is at the heart of the ENS DAO controversy, and Millegan was caught in the middle. If there was no organized collusion against Millegan at the start of the controversy, the events certainly consolidated forces with a dogmatic vision for web3 diametrically opposed to Millegan’s.
While community stewards within the ENS DAO voted to remove Millegan from leadership, other influential members from across the web3 ecosphere held court on Twitter. The Twitter space, hosted by another employee of True Names Ltd. who goes by the pseudonym Alisha.eth (Twitter handle: @futurealisha) included leaders from web3 projects including Meta Mask, Entropy, Rainbow and Wonderverse. All speaking participants either identified in categories denounced by Millegan or showed support as allies. Throughout the nearly 4-hour discussion held on February 6, the alliance of web3 leaders outlined their future vision for ENS and the whole of web3. A selection of statements from that twitter space is included (“ENS Is for Everyone”):
“I think it’s a great show of true decentralization and the use of a DAO. Whatever happens moving forward here is really just going to be the blueprint moving forward in the space, and for things that just aren’t allowed. While we can all be pro freedom of speech and stuff like that I do feel that everything just does not need to be said” (@izzyalright, employed by MetaMask, 6:40).
“This for me is a test of, you know, DAOs, digital organizations, D-orgs as I’ve been calling them. This is a test of web3 technology. This is a test of our collective, like, humanity, really… This is a test of like what really is web3…. What we’re doing here is historic. There’s never been like a global and deeply diverse space created like this” (@cult_leader_en, employed by Wonderverse, 7:28).
“This is this new world that we’re in right now. This is the test y’all. This is the test; here it is! Because is it going to work? Is the DAOs, is the rules, is the things we have decided at this point, is this really what’s going to happen. So I’m really more excited than ever to really, really see if this new world is going to work the way we want it. Because if it does, I’m really excited because this is the way it should be. (@Swopes, 45:46).
“Web 3 is inherently queer is inherently tran[s], is inherently women is inherently sex work is inherently, you know, indigenous. It’s inherently those things, so there is no, there is no web3 for Brantly…. You can be in crypto, you can be in the blockchain world, but you are not a part of web 3” (@cult_leader_en, 1:30:41).
“When we talk about making space for people… what people don’t understand is that space is zero-sum. That, when you take away from some… you give it to someone else. When you take away voices from trans and queer people and you support…hate speech, you know, it makes it such that it’s much more difficult for trans and queer people to be involved in those spaces altogether…. its’s just a zero-sum game, that’s all it is. That’s all making space is” (@__tux, founder of Entropy.xyz, 2:13:00).
“…it has come at a cost, but we are here, and we can build something that is in the image of ourselves, and that is historic… That opportunity to be the first, to be the innovator, to be the one who’s voice is the loudest is such a blessing” (@cult_leader_en, 2:38:33).
Web3 as defined by Alisha.eth’s Twitter space doesn’t ultimately care about censorship or free-speech. These are the concerns of blockchain or are so malleable as to mean very little. To them, web3 is a means to end: the transfer of power from oppressor to oppressed. In their web3, Millegan must not hold power, delegated or otherwise.
“All I want to know is what is the method to remove them… I don’t need to hurt Brantly Millegan, I just need him removed so we can move on, and learn how to do it so that when we encounter this situation again, and we will, that we know how to do it next time” (@MikeMongo, 1:09:47).
Of course, rhetoric alone, however alarming, is simply hot air unless it can be implemented. Thankfully, True Names Ltd. made the decision sometime prior to lock the .eth root domain smart contract, meaning that not even the controlling owner (currently a multisig wallet controlled by 7 “trustworthy individuals in the Ethereum community,” but eventually the DAO) can edit its contents (“Frequently Asked Questions”). If future ownership wanted to edit the .eth root (even to fix a bug in the coding) they would have to create an entirely new protocol with a new ENS token, and even then the original protocol would remain in operation so long as the Ethereum blockchain exists. In effect, .eth names are irrevocable.
But what of the termination of Millegan’s employment by True Name Ltd.? Bowing to social pressures sets a dangerous precedent for web3 companies, but ultimately the private interest of a corporate entity is its own regardless of how much we might question undue influence. What’s important for the traditional vision of web3 is that the architecture of the code remains resistant to censorship in all its forms. Still, the power of centralized companies and the danger of their manipulation by external forces in these crucial development years for web3 should not go unnoticed.
The real danger is in the ability for collusive forces to exert power over the decisions of a DAO, especially in a web3 where DAOs are granted wide powers over coded architecture. Although the ENS DAO will never be able to alter the .eth root, other contracts, such as those managing pricing and registration for .eth names, are still under the control of the multisig and may one day be within the hands of the DAO. Some of these contracts may be locked in the future after bugs and future compatibility features have been worked out. Other contracts may never be locked, such as those that rely on changing oracles (information provided from outside the blockchain).
In the fallout of the Millegan incident, it was clear that the DAO was ill-prepared to deal with a divided community. In the weeks that followed and leading up to the first time the DAO would have an opportunity to weigh in on the issue through a public vote (by determining whether Millegan would keep his position as a director of the ENS foundation) discussion ensued over the DAO’s code of conduct, bylaws, steward appointment processes, and, perhaps most importantly, potential powers the DAO could wield against protocol users. Some were interested in identifying potential threats, others in potential opportunities to punish Millegan and use the DAO to enforce their own vision of web3.
The DAO cannot revoke .eth names, but a DAO seeking to punish specific ENS users could potentially affect the ENS protocol in other ways. For example, the DAO could raise registration/renewal fees to the point where ENS names become too expensive to maintain. It could also modify contracts through a controller that would allow for the refusal of any name’s renewal. It is also seemingly possible that the DAO could create address allow/block lists for name registration. These measures go against the DAO’s constitution, but they are code-legal in the protocol’s current form.
If a DAO can exert so much control over a protocol’s code, the security of the voting process becomes a primary concern. Like many DAOs, ENS relies on tokenized voting where each ENS token essentially equals one vote. To vote, token holders must go through a delegation process where they can choose to either delegate their tokens to themselves or another trusted party to vote on their behalf. Millegan is the largest delegate representing approximately 1.5% of all currently delegated ENS tokens and therefore also holds the largest vote within the DAO. Millegan’s voting power could be adjusted higher when accounting for the fraction of ENS delegates that actively participate in the DAO, but it would be a mischaracterization to attribute this power to failure of decentralization considering his vote represents token holders who entrusted their vote to him and are free to redelegate as they see fit.
Still, as the DAO prepared to vote on Millegan’s position as director of the ENS Foundation, questions were raised regarding whether Millegan’s conflict of interest should disqualify him from voting on the measure. Suggestions to nullify Millegan’s vote were ultimately rejected on the grounds that there was no conceivable way to do so without disenfranchising voters who delegated to him, but many still found the issue along with the DAOs unpreparedness for the situation problematic.
More concerning were discussions regarding how the DAO could deal with such issues in the future. Vocal DAO contributor Avneet Singh (posting under the username inplco), a PhD astrophysicist at the Albert Einstein Institute in Germany who was running to replace Millegan as director of the ENS foundation, suggested that it may be possible in the future to sanction targeted accounts by releasing delegated votes back to their original owners or blacklisting targeted addresses along with their delegated voters entirely. In response, ENS Founder Nick Johnson clarified that forcibly un-delegating votes will never be code-legal due to the .eth smart contract lock. Other sanctions against delegate voting power may be possible, but Johnson maintained he would “sooner leave the ENS DAO than see this implemented” (“Open Letter to Brantly Millegan”). While the strength of Johnson’s personal conviction is reassuring, it is concerning that there are many in the ENS DAO and web3 more broadly who do not feel the same. Singh for his part, maintained earlier in the topic that “Absolutist ideas like free-speech and right-to-associate (general absolutism) are arcane.”
The proposal to remove Millegan itself, which was incidentally drafted by Singh, moved on with just under three weeks of deliberations before the community agreed it was ready for vote. Although early drafts included two separate proposals (one to remove Millegan and another to replace him should the first proposal pass) the decision was made to create one proposal for a new director, giving voters the opportunity to vote for Millegan to continue in the role or for any of four DAO members running to replace him. A combined proposal was deemed simpler and shielded current ENS directors from legal liability resulting from a potentially vacant directorship position. Questions were raised regarding the appropriateness of forcing Millegan to essentially re-run for a position he was never removed from before his term-of-service was up, but the DAO bylaws had little to say on the matter and legal counsel found no technical issue.
Nevertheless, it could be argued that the combined proposal put supporters of Millegan at a disadvantage as they would be forced to decide if they could safely vote for Millegan or if their vote would be better spent ensuring a “next best” applicant is elected, effectively splitting the vote. Some candidates may not even have run if Millegan’s position were not in danger. In explanation of his candidacy, daylon.eth stated “I don’t want the position but I more don’t want it to go to someone that just showed up because of drama” (daylon.eth).
Whether it was apathy or ulterior motives on the part of the proposal’s writers isn’t clear, but the fact that DAO official discussion is dominated by a handful of reoccurring usernames doesn’t help. It is difficult to fault ENS for the low participation (DAO participation is notoriously low across web3), but the result is that the active contributors drafting proposals have a significant level of soft power over ENS. In such a small echo chamber it’s easy for groupthink to drive out contrary opinions, especially when those contrary opinions come from those without “active contributor” status. This is especially problematic considering the informal nature of the DAO’s proposal process. Whereas traditional democratic systems often require a specified number of signatures to put a proposal out to vote, no such requirement exists within the ENS DAO.
Yet another concern is the security of discourse in the public square. Communication regarding the ENS DAO currently takes place on Twitter, in the official ENS Discord, and the official Discourse forum. Millegan was suspended from Twitter following the controversy and disappeared from the ENS Discord server for the duration of the debate (as of recent he is back on Discord and the nature of his disappearance is unclear). Millegan was never banned from the ENS Discourse forum, but it is a possibility, especially now that the DAO has taken up a proposal for an official code of conduct. What are the consequences for web3 if the DAO, a rogue mod team, or even unaffiliated tech services have the power to silence DAO members?
In short, even if a DAO were to ensure the security of the voting process itself, the security of the proposal drafting process, including what proposals make it to vote and in what form, and the security of a public square free from undue censorship are significant concerns for a web3 that relies upon decentralized governance. At the very least, the events of the ENS controversy exposed vulnerabilities in current DAO systems open to exploitation. Is this the status quo we can come to expect from web3?
“…I would strongly say I think it’s a bad precedence for the DAO [voting to remove Brantly], but really web3 broadly speaking, that we’re going to exclude people who hold positions that are held by huge swaths of the world even if you think it’s wrong. I think that is a very narrow-minded position. I don’t think it’s practical. I think it would lead to the failure of web3” (Millegan, 33:45).
Which Way Web3?
For uncertain reasons, just days before the proposal was scheduled to go to vote several DAO members expressed concern over the proposal in its consolidated form. While concerns were initially rebuffed, it was eventually decided that the proposal would again be split. In response, Singh withdrew his candidacy. Two new proposals were quickly drafted (without the involvement of Singh) and went to vote on February 28: Proposal EP6.1 to remove Millegan from his directorship and proposal EP6.2 to elect a replacement dependent upon the results of EP6.1. The proposal to remove Millegan failed with a majority of 1.6 million ENS tokens, or 43.39% of the vote (19.1% abstained), including Millegan’s delegation of 363 thousand votes, and the results of EP6.2 were nullified.
The outcome was reassuring and is evidence that many in web3 still hold to its traditional principles, but it does not guarantee that the processes were secure or that decentralized governance will ever be a substitute for immutable code. The only way to ensure a protocol remains accessible to all without discrimination is to lock its code. To the extent that a protocol cannot be locked due to development or design, we must be realistic about the threats and set up governance accordingly. Fostering a sense of community spirit can help drive the success of a developing protocol, but unity at the expense of robust defenses against collusion are simply no longer viable. Whatever unanimity web3 may have enjoyed prior, it’s clear that there is now an ideological divide.
In a 2020 podcast with Palladium Magazine, Buterin hypothesized that U.S. tech companies benefited in early years from a level of public “goodwill” simply for being separate from government interests. Heavy regulation stifling innovation as well as general corruption in the lawmaking process was public enemy number one, especially in post 9/11 America. On the other hand, the internet itself was seen as an unregulatable force for liberation, and companies such as Facebook, Twitter, and Google, now pejoratively referred to as “big tech,” benefited from their association with it. Today that goodwill has by and large been lost amid privacy and censorship scandals.
“… I think that US tech platforms benefited from a lot of goodwill kind of in part because they were… not really touched by the government much, especially internationally, and now of course that’s changing… centralized tech ecosystems are definitely not going to be capable of… being independent from the political environment they’re based in. But people definitely do see blockchains as basically being… the last hope for an alternative in some sense” (Vitalik Buterin: Mechanism Design in Political Theory,” 16:33).
Buterin’s analyses confirms Lessig’s early warning that the removal of government regulation without an equal scrutiny of developing code would not guarantee a free web. “…so obsessed are we with the idea that liberty means ‘freedom from government’ that we don’t even see the regulation in this new space. We therefore don’t see the threat to liberty that this regulation presents” (Lessig). Blockchains have been described as trustless systems due to their decentralized consensus models that remove third party arbiters, but we must actively audit the code and the intentions of those writing the code the same way we would audit any other institution. For our own good, we cannot allow our common goal to free the web from corporate powers to blind us to other threats, even from within our own camp.
Today, as web3 overtakes web2 with new architecture, and therefore new regulation, we are at a critical junction point. We must interrogate our new regulators and ensure the architecture provides for the freedoms and values we desire. Web3 is still being written, and one day it will either free us from the shackles of big tech oligarchs or deliver us into the hands of new masters.
Which way web3?
Works Cited
Reynolds, Sam. “Ethereum Name Service Removes Core Team Member Brantly Millegan Over 2016 Tweet.” Coindesk, 7 Feb. 2022.
Silver, Charles. “Council Post: What Is Web 3.0?” Forbes, 6 Jan. 2020.
Millegan, Brantly. “ENS Domains twitter space 2/5/22 — Brantly Millegan defending his views before getting fired by TNL.” YouTube, Theology Meets Web 3.0, 8 Feb. 2022.
“Why ENS Names Are Much More than Domains.” Fleek Blog, 22 Dec. 2020.
Lessig, Lawrence. “Code Is Law.” Harvard Magazine, 1 Jan. 2000.
Brennan, Coogan. “Want to Really Understand Blockchain? You Need to Understand State.” ConsenSys, 20 Mar. 2020.
Buterin, Vitalik. “The Meaning of Decentralization.” Medium, 6 Feb. 2017.
Hamilton, Alexander. “The Federalist Papers : №26.” The Avalon Project, Lillian Goldman Law Library.
“Vitalik Buterin “Spiritually, I Really Love DAO.”” YouTube, uploaded by Stakeborg DAO, 19 Oct. 2021.
Buterin, Vitalik. “Bootstrapping a Decentralized Autonomous Corporation: Part I.” Bitcoin Magazine, 19 Sep. 2013.
““ENS Is for Everyone” — Twitter Space Hosted by ENS Domains Community Leader Alisha.eth, 6 Feb. 2022.” YouTube, Theology Meets Web 3.0, 8 Feb. 2022.
“Frequently Asked Questions” ENS Documentation, ENS Domains, 2021.
“Open Letter to Brantly Millegan, in Reference to EP6.1.” ENS DAO Governance Forum, 5 Mar. 2022, discuss.ens.domains/t/open-letter-to-brantly-millegan-in-reference-to-ep6–1/11164.
daylon.eth. “Nominations for ENS Foundation Director to Replace Brantly.eth.” ENS DAO Governance Forum, 8 Feb. 2022, discuss.ens.domains/t/nominations-for-ens-foundation-director-to-replace-brantly-eth/10634.
“Vitalik Buterin: Mechanism Design in Political Theory.” YouTube, uploaded by Palladium Magazine, 9 Jun. 2020.