The Two-Block Rebellion: What BIP-110's Collapse Reveals About Who Really Governs Bitcoin

The most consequential fork in Bitcoin's recent history produced exactly two blocks. In eight hours, the BIP-110 chain — a would-be rebellion against Ordinals inscriptions — generated exactly two blocks at height 961,633, while the main chain, indifferent as a mountain, climbed to 961,681. Forty-eight blocks of separation. A chasm disguised as a crack. This was not a technical failure in the way software fails. It was an economic statement, made in the only language Bitcoin's consensus layer truly understands: hashpower. The nodes that activated the fork believed they were enforcing a rule. The miners who ignored them were demonstrating something far more consequential — that in Bitcoin, the right to propose rules and the right to execute them belong to different constituencies, and the second right always prevails. I have spent the better part of a decade watching governance experiments fail in crypto. From the smart contract audits I performed during the 2017 ICO boom to the liquidity frameworks I built during the 2020 DeFi summer, one pattern keeps asserting itself: whoever controls the economic bottleneck controls the outcome. BIP-110 is the purest expression of that law I have seen in Bitcoin since the Blocksize War. It deserves a careful post-mortem, not because it changed the network, but because it revealed who actually governs it. For readers who missed the story, here are the relevant facts. BIP-110 proposed a change to Bitcoin's consensus layer — specifically, a restriction on the amount of non-financial data that could be written into transactions. This was, without ambiguity, a direct attack on Ordinals inscriptions, which have come to occupy a significant share of Bitcoin's block space as vehicles for NFT-like digital artifacts. The proposal's activation mechanism was unusual. It set a threshold of 55% — at least 1,109 blocks out of 2,016 in a two-week period needed to signal support. This was lower than the 80% threshold associated with miner-activated soft forks such as BIP 91, and higher than the undefined bar of a pure user-activated soft fork. It was a middle path, designed to give miners time to respond while preventing a small minority from vetoing the process. The response was silence. In the preceding cycle, only 51 blocks — 2.53% — signaled support for BIP-110. When the deadline passed without the threshold being met, a subset of node operators triggered the UASF logic anyway. At block height 961,632, their nodes began rejecting blocks that did not contain the signal. Two blocks were mined on the resulting minority chain before it effectively stopped existing. The contrast with historical precedents is instructive. BIP 148, the SegWit UASF, succeeded because miners eventually signaled — not because they loved SegWit, but because the alternative was a network split that would have destroyed value for everyone. Bitcoin Cash became a durable fork because miners actively switched hashpower to a competing chain with different rules and a different vision. BIP-110 occupied a strange middle ground: it had the coercive structure of a UASF without the economic backing that made BIP 148 credible, and it had the rule-differentiation of a hard fork without the miner migration that made Bitcoin Cash viable. It was, in effect, a fork that nobody showed up to. The interesting analysis begins after the facts. Here is what I believe this event actually tells us. Let me be precise about what happened technically, because the details matter more than the headlines. The fork was not triggered by miners switching protocols. It was triggered by nodes unilaterally refusing to accept blocks that did not include the BIP-110 signal. This is the defining feature of a user-activated soft fork: enforcement happens at the node level, not the mining level. In theory, this works. If enough economic nodes refuse to validate non-signaling blocks, miners face a choice: comply and keep the network unified, or continue mining and risk producing blocks that a significant portion of the economy will reject. BIP 148 succeeded precisely because the threat of economic rejection forced miners to compromise. BIP-110 failed because the threat was never credible. The nodes that activated the fork represented a narrow slice of the network's economic weight. More importantly, the miners who ignored the signal faced no meaningful cost for doing so. Their blocks continued to be accepted by the vast majority of the network. Their revenue continued to flow. The few nodes enforcing BIP-110 were, in the most literal sense, shouting into a vacuum. The result — two blocks in eight hours — tells us something essential about the nature of that "alarm clock" mining. Those blocks were almost certainly produced by enthusiasts willing to burn a trivial amount of electricity to prove a point. No pool with real hashpower pointed even a fraction of its capacity at the fork chain. If they had, we would have seen dozens of blocks, not two. The two-block chain was a gesture, not a competitor. Its security was, for all practical purposes, zero. An eight-hour stretch with two blocks against a main chain producing forty-nine means the chain was not merely weak; it was non-viable. Any exchange, wallet, or application that had built on it would have been exposing users to a 51% attack risk of catastrophic proportions. There is also a quiet forensic detail in the numbers that deserves attention. The 51 signaling blocks in the previous cycle were not distributed evenly across the network's mining pools. They clustered — a signature pattern that suggests a small, coordinated group of operators rather than a broad grassroots movement. In my 2017 audit work, I learned to read such clusters the way a detective reads fingerprints. A signal that appears in tight geographical or pool-level concentrations is not consensus; it is a caucus. BIP-110 had a caucus, not a consensus. Based on my experience auditing smart contracts in 2017, I can tell you that the difference between a "technical protest" and a "real fork" is visible in exactly this kind of data. A real fork has mining infrastructure, pool coordination, a roadmap for exchange listing. BIP-110 had none of that. What it had was a signal — a way for a community of node operators to say: we do not accept the transformation of Bitcoin's block space into a general-purpose data layer. There is something philosophically interesting about BIP-110 that most commentary missed. Bitcoin improvement proposals are almost always additive. SegWit added a new segregation structure. Taproot added new signature schemes and script flexibility. The history of Bitcoin's protocol evolution has been about expanding what the chain can do. BIP-110 was subtractive. It sought to remove a capability — the ability to write arbitrary non-financial data into transactions — that had been made practically available by Taproot's activation in 2021. This is a different category of change, and it carries a different risk profile. Additive changes can be adopted incrementally, tested in parallel, and abandoned if they fail. Subtractive changes are inherently coercive: they take away something that already exists, which means they create immediate losers. The Ordinals ecosystem was the loser that BIP-110 proposed to create. And here is the critical economic detail that the proposal's supporters appear to have miscalculated: Ordinals fees had become part of the mining revenue structure. The miners who would have had to enforce BIP-110's restrictions were the same miners collecting fees from inscription traffic. Asking them to support a proposal that would cut off a revenue stream — without offering any compensatory incentive — was asking them to act against their own economic interest. Follow the money, not the noise. The noise around BIP-110 was about the purity of Bitcoin's monetary network, about the philosophical question of whether block space should be reserved for financial transactions. The money was simpler: miners were collecting fees from Ordinals activity, and BIP-110 threatened those fees. The 2.53% signaling rate was not primarily a philosophical statement. It was an invoice. Let me push deeper into the miner's decision calculus, because this is where the real insight lies. There is a common assumption in crypto governance debates that miners are passive executors of node preferences. The BIP-110 episode demonstrates that this assumption is dangerously wrong. Miners occupy a unique position in Bitcoin's power structure. They do not propose rules — that prerogative belongs to developers and node operators through the BIP process. But they do execute rules, and more importantly, they can decline to execute rules that do not serve their interests. This is not corruption; it is the design. Bitcoin's original vision was that economic participants would act in their own rational self-interest, and that the protocol's rules would be those that survived the crucible of that self-interest. BIP-110 failed the crucible. Miners looked at a proposal that would reduce their fee income, that had attracted only 2.53% support in signaling, and that promised no compensatory benefit, and they concluded — rationally — that ignoring it was the optimal strategy. The two blocks mined on the fork chain were the equivalent of a protest sign. The forty-nine blocks mined on the main chain during the same period were the actual verdict. This is the deeper lesson: Bitcoin's governance is often described as messy, informal, and ambiguous. Beneath the mess lies a hard economic filter. Proposals that align with the incentive structure of the network's economic actors can pass. Proposals that do not align with those incentives do not pass. The filter is not democratic, and it is not fair. It is economic. BIP-110 hit the filter and was stopped cold. There is a further nuance worth considering: the mining fee tipping point. Ordinals-related transactions have, since their inception, contributed a fluctuating but persistent stream of fee revenue to miners. In periods of inscription hype, that stream can represent a meaningful supplement to the block subsidy. This has quietly changed the political economy of Bitcoin mining. A decade ago, miners depended almost entirely on the block subsidy; today, the largest pools have tasted diversified fee income. That taste creates a constituency — not for Ordinals as a cultural phenomenon, but for the fee diversity that Ordinals provides. BIP-110 threatened to eliminate that diversity at a stroke, which is another way of saying it threatened the miners' newfound revenue resilience. In my 2024 analysis of Bitcoin ETF approval and its impact on liquidity distribution across altcoins, I noted that institutional capital had begun reshaping the incentives of every actor in the ecosystem. This is a related phenomenon: the growing weight of Ordinals fees has reshaped miners' incentives with respect to protocol changes. The miners who would have enforced BIP-110 are the same miners who have come to depend on a diversified fee structure that includes inscriptions. The result is a silent but powerful political bloc against restrictions — a bloc that operates not through lobbying or governance forums, but through the simple act of not switching hashpower. What makes BIP-110 analytically precious is what it reveals about the split between what I will call the "rule initiation right" and the "rule execution right" in Bitcoin. The rule initiation right belongs to anyone who can write a BIP and gather a community of node operators. It is deliberately open. Anyone can propose a change to Bitcoin's consensus rules, and if they find enough supporters, they can attempt to enforce it. This openness is one of Bitcoin's most celebrated features — it is what distinguishes a permissionless network from a corporate-controlled one. But the rule execution right belongs to miners. A rule change only becomes reality if miners are willing to mine blocks under the new rules. This is the asymmetry that BIP-110 exposed. The node operators who activated the fork were exercising their initiation right with perfect legitimacy. They signaled. They activated. They began rejecting non-compliant blocks. And none of it mattered, because the execution right was held by miners who had no interest in the outcome. This split is not a bug in Bitcoin's design. It is arguably the most important feature of the system — the separation of powers that prevents any single constituency from hijacking the network. But it creates a specific form of frustration: a community can be right about a problem, can articulate a solution, can even activate that solution — and still lose, because the economic center of gravity refuses to move. I saw this dynamic in a different form during the 2020 DeFi summer, while producing a report on how unstable stablecoin pegs affected cross-border remittances in Latin America. The communities building stablecoin alternatives had the right diagnosis — pegged assets were failing real users — but they lacked the economic gravity to force a change in the networks they were building on. The protocols that succeeded were the ones that aligned the incentives of the people who controlled the rails. The ones that failed were the ones that expected the rails to move out of principle. For the Ordinals ecosystem, the immediate reading of this event is positive. BIP-110's failure means that, for now, no protocol-level restriction is coming. Inscription traffic can continue to use block space as it has been. This is a meaningful reprieve, and it is rational for Ordinals-related assets to trade with slightly lower protocol risk. But I want to offer a note of caution. The reprieve is temporary, and the structural vulnerability that BIP-110 exposed has not gone away. Ordinals exists on Bitcoin only because Bitcoin's base layer tolerates its presence. That tolerance is not guaranteed. It is a function of miner incentives, which can change. Consider the scenario that should worry Ordinals holders, not the one that just failed. If inscription traffic grows to the point where it causes sustained mempool congestion — where ordinary financial transactions face meaningful delays and fee spikes — the political calculus among miners could shift. Miners would face a trade-off between the fee income from inscriptions and the broader economic health of the network on which they depend for their own long-term viability. In that scenario, a compromise proposal — one that limits large inscriptions while allowing smaller data writes, or one that reserves specific block space for financial transactions — could attract real miner support. This is the risk that the BIP-110 failure should not blind us to. The proposal that died this week was the maximalist version: a blanket restriction on non-financial data. The proposals that might succeed in the future are the incremental ones, the ones that carve out the most extreme uses while preserving the fee-generating middle ground. And those proposals will be much harder to oppose, because they will come with economic logic attached. There is also a longer horizon to consider, one that connects this governance skirmish to the technological currents I have been tracking in my recent work on AI-crypto convergence. We are entering an era in which autonomous agents will increasingly need to write data to public ledgers — attestations, credentials, provenance records, machine-payment trails. The question of who gets to write to Bitcoin's block space will soon become the question of whether AI agents can participate in the network's economy at all. BIP-110 was an early warning shot in that larger war. The factions aligning today around inscription policy will be the same factions negotiating the terms of machine-readable ledger space tomorrow. The ledger remembers what the headlines forget. Underneath the technical and economic analysis lies a philosophical conflict that will not be resolved by any single vote or fork. Bitcoin is undergoing an identity crisis that has been building since the first inscription was written: is it a pure monetary network, a sovereign digital currency with a singular purpose? Or is it a global settlement ledger capable of carrying any data that users are willing to pay for? BIP-110 was the pure-monetary-network faction's answer, and it failed. But the faction remains. The nodes that activated the fork did not disappear when their chain stalled. They are still running nodes. They are still signaling their displeasure. They have demonstrated a willingness to take costly, even futile, action to express that displeasure. That is not the behavior of a constituency that will quietly go away. The pure-monetary faction's failure reveals something important about where Bitcoin has actually landed. As a network, Bitcoin has become something more than a payment rail. It has become a substrate for digital artifacts, for timestamped data, for a strange new economy of inscriptions and the infrastructure building around them. This evolution was not planned by the core developer community, and it is not universally loved. But it has the most important property that any protocol feature can have in a permissionless system: users are willing to pay for it. And that, ultimately, is why BIP-110 failed while Ordinals continues to grow. In a system where miners are rational economic actors, the question "should this data be allowed on the blockchain?" is always answered by another question: "is someone willing to pay for it?" The BIP-110 episode also casts light on Bitcoin's broader data ecosystem — the alternative protocols that do not rely on direct inscription. Projects like RGB, Counterparty, and Stacks have long argued that Bitcoin should not become a general-purpose data layer, precisely because the base layer cannot offer cheap, flexible storage. BIP-110 was, in a sense, their argument made into code. The proposal's failure does not validate the inscription model; it simply demonstrates that the market prefers it for now. RGB and its peers continue to develop client-side validation and layer-two approaches that keep data off the main chain entirely. If inscription congestion ever becomes a genuine problem, these projects will be positioned to capture the overflow. The BIP-110 failure may, paradoxically, be the best marketing they have received in months — a reminder that the "safe" path of base-layer restriction is blocked, and the alternatives are worth examining. From a regulatory perspective, this event deserves attention precisely because it is a non-event. The fork involved no new token issuance, no securities offering, no KYC/AML implications, and no exchange exposure. The two-block chain had no users, no liquidity, and no institutional footprint. There is nothing for the SEC, FinCEN, or any other regulator to act on. But the non-event status is itself informative. It reminds us that Bitcoin's governance conflicts are, at their core, private ordering problems. They are resolved by economic actors using economic means, not by legal institutions using legal means. The BIP-110 supporters did not file a lawsuit to stop Ordinals; they activated nodes. The miners did not lobby Congress; they declined to switch hashpower. This is a form of governance that operates entirely outside the regulatory frame, and it works. It is worth remembering for anyone inclined to believe that the future of crypto governance will be decided in courtrooms and legislative chambers. Now let me offer the counter-intuitive reading that I believe most market commentary will miss. On the surface, the winners of this event are clear: the Ordinals ecosystem, which avoided a potentially existential protocol restriction, and the miners, who retained their fee diversity. The losers are equally clear: the BIP-110 supporters, whose coercion attempt collapsed. I think the opposite is true in ways that matter for the next two years. First, consider what the BIP-110 supporters actually achieved. Yes, the fork failed. But the act of triggering it — of going through the full UASF choreography and forcing the network to respond — accomplished something that two years of BIP discussions could not. It made the conflict visible. Every node operator, every pool, every analyst examining this event is now discussing the question that the BIP-110 supporters wanted discussed: should Bitcoin's block space be used for non-financial data? The fork was a communications strategy disguised as a technical event. And from that perspective, it worked. Second, consider the normalization of coercive action. Before BIP-110, the lesson of BIP 148 was sometimes read as "UASF works." The BIP-110 failure adds a crucial qualifier: "UASF works only with economic backing." But it also demonstrates that node operators are willing to attempt coercion even without that backing. This lowers what I would call the activation threshold of desperation in Bitcoin governance. The next group that feels passionately about a rule change will look at BIP-110 not as a failure, but as a proof that the mechanism is available — and that the cost of failure is surprisingly low. A proposal that dies quietly at 2.53% signaling is one thing. A proposal that forces the conversation, generates analysis, and enters the historical record is another. The BIP-110 pattern — propose, signal, activate, fail, but be remembered — is likely to be repeated. Third, and most importantly, the real long-term threat to Ordinals is not the maximalist restriction that just failed. It is the middle path that BIP-110's failure makes more likely. If a blanket ban on non-financial data is politically impossible, then a targeted restriction on the most extreme uses becomes more plausible. Such a proposal would be harder for Ordinals supporters to oppose, because it would not threaten the core inscription use case. And it would be easier for miners to support, because it would preserve most of the fee revenue while addressing congestion concerns. If such a proposal emerges in the next eighteen months, BIP-110 will be remembered as its precondition. The maximalist failure opens the door for the incremental success. I am also skeptical of the "miners will always protect Ordinals" thesis. It rests on a static view of miner incentives. My experience in cross-border payment research has taught me that economic actors do not have fixed preferences; they have context-dependent ones. A miner that supports Ordinals fees today may support inscription limits tomorrow, if the alternative is sustained congestion that threatens the network's broader adoption. Volatility is the tax on impatience, and the Ordinals ecosystem's impatience — its desire to treat today's victory as permanent — could cost it dearly when the tax comes due. Let me turn briefly to how this event shapes the broader narrative and market psychology, because in a bull market, narrative often leads price. The BIP-110 failure adds a powerful data point to the story that Ordinals and the broader "Bitcoin as a programmable asset" movement have been telling: Bitcoin's base layer will not be restricted. The "pure monetary network" narrative has taken a significant hit, not because it was defeated in debate, but because it was defeated by economics. This matters for how institutions and retail participants perceive the future of Bitcoin's utility. If the base layer is open to data-bearing transactions — and if attempts to close that door are structurally doomed — then a whole class of applications becomes viable, and a whole set of investment theses becomes more credible. For traders, the market signal is relatively muted. Bitcoin's price impact from this event is less than half a percent, in my estimation. The fork was too small, too invisible, and too quickly resolved to move the macro narrative. But for Ordinals-linked assets, the short-term signal is more tangible: a policy risk that was priced as a tail risk has been removed, at least for now. I would expect modest positive drift in inscription-related tokens in the weeks ahead, though I would caution against extrapolating from that drift. Follow the money, not the noise. The money in this event is in the fee streams, the miner incentives, and the long-term evolution of block space allocation. The noise is in the headlines about "Bitcoin splitting" or "Ordinals winning." In a bull market, the temptation to read every event through the lens of price is strong. But the events that shape Bitcoin's next decade are rarely the ones that move its price this week. What should a thoughtful observer take from the Two-Block Rebellion? First, Bitcoin's governance is not broken. It is working exactly as designed — as an economic filtration system in which rule changes must survive the test of miner self-interest. BIP-110 failed that test, and the network continued producing blocks as if nothing had happened. That is resilience, not dysfunction. Second, the conflict that produced BIP-110 will not disappear. The block space question — who gets to write what into Bitcoin's ledger, and at what cost — is now a permanent feature of the network's political landscape. The groups that took opposing sides in this skirmish will meet again, probably with more sophisticated weapons. Third, watch for the middle path. The most important proposals of the next cycle will not look like BIP-110. They will look like compromises — targeted limits, fee-based allocation mechanisms, layered data solutions — and they will be much harder to dismiss. The question for the Ordinals ecosystem is whether it uses its temporary victory to build durable economic and political capital within the mining community, or squanders it in the belief that the war is over. The two blocks mined at height 961,633 will one day be a footnote. But the signal they sent — that a determined minority can force Bitcoin to confront its own identity — will echo through every future attempt to reshape the network. Bitcoin has absorbed this challenge, as it has absorbed so many others. The question is whether the absorption strengthens the system or merely postpones the reckoning. Volatility is the tax on impatience. And in Bitcoin's governance, as in its markets, the impatient keep paying.