Logic holds until the ledger bleeds. Bitcoin’s conservative upgrade dogma—honed over fifteen years of near-95% miner consensus—is being tested by a proposal that could crack the foundation. BIP-110, a seemingly narrow set of seven script and witness restrictions, carries a hidden payload: an activation threshold of just 55% miner signaling, with no explicit failure state. Michael Saylor, Bitcoin’s largest corporate holder, has published a list of 110 reasons to reject it. But his opposition is not about the seven restrictions themselves. It’s about the machinery of governance. And the machinery has a flaw that no amount of HODLing can fix.
When I first read about BIP-110, I went straight to the proposal’s activation mechanism. A 55% threshold. No FAILED state. In Bitcoin’s history, every consensus change has demanded overwhelming supermajority—the BIP-9 standard of 95% miner signaling, with a clear timeout that forces the proposal to expire if it lacks support. This is not an arbitrary tradition. It is the immune system of a permissionless network. The logic is simple: if you cannot convince almost everyone to upgrade, you have not built a strong enough case. BIP-110 discards that guardrail.
Saylor’s 110 reasons are a fascinating document—part technical critique, part political manifesto. He argues that the seven restrictions (limiting script public key lengths, disabling certain Taproot paths, capping witness item counts, and others) could be enforced by non-consensus means: node operators can configure their clients to reject large witness data; the fee market already disincentivizes bloat. He is right on the technical merits. But the deeper issue is the governance precedent. Once you lower the activation bar to 55%, the network signals that a determined minority can force a change on the rest. The next BIP with a 55% threshold might not be about inscriptions—it could be about block size, inflation, or something far more corrosive.
Context: What BIP-110 Actually Does
BIP-110, authored by a group of Bitcoin developers concerned about data bloat from Ordinals and similar protocols, proposes seven consensus-level restrictions on script and witness data. These include: - Limiting the maximum public key length in a script to 33 bytes. - Disabling certain Taproot leaf script versions beyond the first. - Capping the number of witness items in a single input. - Preventing the reuse of certain opcodes in non-standard ways.
The stated goal is to reduce blockchain data growth and limit attack surface. The unstated goal is to curtail the inscription phenomenon without a hard fork. But the mechanism—consensus change—is the bluntest tool available.
The activation plan is even more contentious. Instead of the traditional BIP-9 approach (95% miner signaling and a defined timeout), BIP-110 proposes a 55% threshold and no timeout. If 55% of miners signal support, the upgrade becomes active after a grace period. The remaining 45%—whether they oppose, ignore, or haven’t upgraded—will produce blocks that violate the new rules. Those blocks will be orphaned by the 55% majority. The result is a soft fork enforced by a simple majority, not a supermajority.
Core: The Governance Vulnerability No One Is Auditing
In my years auditing smart contract governance—from the 2x2 DAO integer overflow to Aave v2’s oracle exposure—I’ve learned that the most dangerous flaws are not in the code but in the process. A 55% threshold with no failure state is not a minor design choice. It is a structural vulnerability that exposes the network to capture.
Consider the game theory. Under BIP-9, a miner who opposes a change can simply refuse to signal, and after the timeout, the proposal dies. The burden of proof is on the proponents—they must convince nearly the entire network. Under BIP-110, the burden shifts. A miner who opposes must convince 45% of other miners to also oppose, or be forced onto a minority chain. The proposal can pass with only a bare majority, even if a significant portion of the community rejects it.
Trust is a variable, not a constant. Bitcoin’s value proposition as “digital gold” relies on the belief that the rules will not change arbitrarily. BIP-110, by establishing a low-threshold precedent, introduces a new variable: the will of a 55% miner cartel. Saylor’s 110 reasons include this exact argument. But he couches it in the language of specific technical risks, perhaps because speaking directly about governance capture sounds too conspiratorial for a public figure. I am not bound by such constraints.
The seven restrictions themselves are technically debatable. Some might improve security by reducing the complexity of validation. Others might accidentally break legitimate protocols like RGB or Taproot Assets, which rely on the very script paths BIP-110 restricts. But these are engineering problems—solvable with audits and testnets. The governance flaw is foundational. It cannot be patched by a software update; it requires a change to the proposal’s activation mechanism, or a rejection of the proposal entirely.
Contrarian: The Silent Supporters and the Hidden Cost of Opposition
Here is where the narrative gets uncomfortable. Saylor’s opposition is loud, visible, and protective of the status quo. But what if the status quo is not as stable as he believes? The inscription wave has driven Bitcoin transaction fees to levels not seen since the 2017 congestion. Ordinals alone accounted for over 40% of all Bitcoin transactions in some months. Miners have earned hundreds of millions in fees from inscriptions. But centralization critics argue that the data bloat increases node operation costs, driving more users to rely on third-party infrastructure.
BIP-110 might be a clumsy attempt to solve a real problem. The contrarian angle is that Saylor’s defense of “non-consensus measures” is itself a form of governance capture—the capture of the network by large holders who prefer low fees and a clean blockchain, at the expense of innovation. Inscriptions may be ugly, but they brought new users and new revenue to the Bitcoin economy. By opposing any consensus change, Saylor implicitly endorses a network that cannot adapt to changing usage patterns. That rigidity is a risk too—one that might, in the long run, drive activity to other L1s or L2 solutions.
But the true blind spot is the assumption that BIP-110 will be defeated. If Saylor’s opposition galvanizes community resistance, the proposal dies. If it does not, and if a coalition of miners and core developers pushes it through, Bitcoin will have crossed a Rubicon. The act of passing a low-threshold upgrade will have been normalized. The next proposal might be even more aggressive. The market will price in governance uncertainty—a premium that Bitcoin has never carried at scale.
Takeaway: The Void Where Immutability Meets Pragmatism
We coded the escape, but forgot the exit. Bitcoin’s strength has always been its conservative upgrade process. BIP-110 challenges that process, not with malevolent intent, but with a flawed activation mechanism that could become a template for future changes. Saylor is right to oppose it—not because the seven restrictions are dangerous, but because the 55% precedent is a ticking bomb.
The real vulnerability forecast is this: within the next two years, another BIP will appear with a similar low-threshold mechanism, perhaps disguised as an emergency fix. By then, the precedent will have eroded the norm of supermajority consensus. The community will have accepted that 55% is enough. At that point, the question is not whether Bitcoin will change, but who will control the threshold.
Silence is the only audit that matters. The market is silent on this governance risk because it seems abstract. But when a proposal actually passes with 55% and the minority chain splits, the silence will break. Until then, watch the miner signaling, read the BIP drafts, and remember: trust is a variable, not a constant. And once the ledger bleeds, there is no rollback.