Who steers the specs: legitimacy fights in Bitcoin and Ethereum process

Process authority in open blockchain development is supposed to be clerical until someone wants it to be something else. This month the Bitcoin development list carried an explicit motion to strip Luke Dashjr of the BIP editorship, and by the time the thread was a day old his write access to the repository had already been pulled. Across the fence, Ethereum Magicians spent hundreds of posts fighting over an issuance-burn EIP while moderators deleted sock puppets and the conduct of the forum became part of the argument. The two threads are not about the same mechanism. They circle the same question: who may steer documents that communities treat as near-constitutional, and what separates the legitimate use of a gate from its capture. On the Bitcoin side, the striking thing is how little of the case survives contact with that question.

A removal in search of a rule

The most honest sentence in the entire Bitcoin thread came from a supporter of the motion. "This is unprecedented because there is, to my knowledge, no formal governance structure or policy around management of BIP Editors," Jameson Lopp wrote, adding a few lines later, "This isn't a democracy and I have no vote." That is the whole problem stated by a witness for the prosecution. There is no written procedure for removing a BIP editor, no threshold, no notice, no defined body to decide. So a handful of participants invented one on the spot and applied it, first and immediately, to the colleague they had spent years fighting with.

You can see the improvisation in the sequence. Murch posted the motion. Within about two days, before anything resembling a decision process had run, Bryan Bishop reported the outcome as already done: "The individual known as Luke Dashjr is no longer a permissioned user on the BIP content repository." There was no vote to point to because, as Lopp conceded, there was nothing to vote with. Access was revoked because the people holding the keys decided to revoke it. Whatever that is, it is not process, and dressing it up as one does not make it so.

The charges are ordinary editing, read in the worst light

Strip away the adjectives and the concrete accusations against Dashjr are two routine editorial acts. He assigned a BIP number, and he merged a pull request. Murch's complaint is that the number was floated on Twitter before list discussion and that the merge landed minutes after the PR opened. Editors assign numbers and merge PRs; that is the clerical job. Whether these particular actions were premature is a fair thing to argue on the merits, but it is an editorial dispute, not misconduct that voids a decade in the role.

The workload charge actively undercuts itself. Murch faults Dashjr for leaving "fewer than 1% of the BIP Editor comments" since April 2024 and for making his first merge "since May 2024," and in the same breath faults him for using the role too aggressively this once. He cannot be both the editor who does nothing and the editor who does too much. What the timeline actually shows is a long-tenured maintainer who stepped in rarely, on a proposal he cared about, in a community that added new editors around him in 2024 and then decided the founder-era editor was the one who did not belong.

Even the people voting to remove him concede the record. Antoine Riard, no ally, wrote, "I respect the years of hard work and dedication invested in the bitcoin project. He has been technically contributing more longer than I had." Michael Tidwell opened with "Thanks for the years of hard work. Genuinely appreciated." These are not the notes you write about someone whose editorial conduct is disqualifying. They are the notes you write about a veteran you have decided to push out for other reasons.

Reputation is not a charge

What are the other reasons? Lopp says it plainly: "I see no need to rehash the lengthy list of disputes in which Luke has been involved; it seems to me that ultimately this comes down to reputation." That is a remarkable admission to build a removal on. Reputation, disagreeableness, and a history of being hard to work with are not violations of any BIP rule, because there is no rule. They are a personality the majority has grown tired of. A standards process that removes its editors for being abrasive, rather than for a defined breach, is not applying a norm. It is settling a grudge and calling the result governance.

The tone of the thread makes the grudge hard to miss. Olaoluwa Osuntokun seconded the motion and, in the middle of it, editorialized: "you fool, don't you see, bcash jr is the real Bitcoin!" That is not the register of a neutral body weighing an editor's fitness. It is a faction enjoying a win. When the same message insists the motion "isn't about punishing a developer for advocating a fork," the sarcasm a few lines up says otherwise.

The conflict of interest runs both ways

The strongest-sounding charge is conflict of interest: Dashjr leads a project that forked from Bitcoin, so he should not hold a privileged position in Bitcoin's standards process. Taken seriously, that principle does not stop where its authors want it to. The movers and seconders are overwhelmingly aligned with one implementation and one faction of the ecosystem. If a stake in the outcome of contentious proposals disqualifies an editor, it disqualifies most of the room that just voted, not only the one member they dislike. Either an editor's outside allegiances are grounds for removal, in which case this becomes a very long list, or they are not, in which case the central accusation collapses. The motion wants the rule to apply exactly once.

There was a fairer path on the table, and the thread walked past it. Riard proposed that Dashjr step aside voluntarily and temporarily, keep contributing reviews, and regain the role in good standing: a proportionate response that respects both the concerns and the tenure. Instead the group chose the summary version, executed before the discussion had even finished. Dashjr's own answer was the reasonable one under the circumstances: "These are false accusations. I have followed the BIP process consistently for years. Your novice opinions alone do not establish new processes." And on the communication breakdown that Murch blamed on him: "You are the one who stopped communicating, not I."

Ethereum shows the same reflex

The parallel across the fence is instructive precisely because Dashjr is nowhere near it. On Ethereum Magicians, the fight over EIP-8361, Tapered Issuance Burn, ran to hundreds of posts and turned on whether cutting validator issuance pushes solo stakers out of the network. Critics like pa7x1 warned about the population dynamics: any group that grows more slowly than the whole for long enough "becomes a rounding error" under exponential growth, which is what negative dilution-adjusted yields do to independent stakers. jdetychey laid out the steelman and the security-budget concerns in detail. What made it a legitimacy story rather than a math story was the moderation: the thread accumulated personal attacks and sock-puppet deletions, so the forum's own conduct became part of the record. Here too, the substance is contested and the policing of dissent is doing quiet work in the background.

The rule should come before the removal

Put the two together and the lesson is not that Dashjr is beyond criticism. He is abrasive, he is combative, and reasonable people wanted him gone. The lesson is about method. Bitcoin removed its longest-serving BIP editor with no written rule, no vote, and access revoked before the argument was over, on charges that amount to ordinary editing plus a reputation his critics openly admit is the real issue. That is a purge with a process-shaped label on it. If the BIP editors want the power to remove one of their own, the legitimate move is to write that power down in the open, define what breaches it, and apply it prospectively to everyone. Improvising it against the person you like least, and only him, establishes nothing except that the gate belongs to whoever currently holds the keys.