Two Blocks, Zero Hash: BIP-110's Failed Fork as a Study in Bitcoin's Economic Veto

At block height 961,632, a group of node operators executed what they believed was a legitimate user-activated soft fork. Eight hours later, their fork chain sat at block 961,633. The Bitcoin main chain, in the same window, produced 49 blocks, reaching height 961,681. Two blocks in eight hours is not a blockchain. It is a press release with a hash appended. We do not build for today. BIP-110, however, failed to build anything at all. This was never a scaling debate. It was a constraint-based proposal designed to limit non-financial data written into Bitcoin scripts, aimed squarely at Ordinals inscriptions. The activation threshold required 1,109 blocks signaling support within a 2,016-block epoch — that is 55 percent. In the previous cycle, only 51 blocks signaled support. That is 2.53 percent. The gap between fifty-five and two-point-five-three is not a negotiation. It is a verdict. The fork was triggered by nodes unilaterally rejecting blocks that lacked the BIP-110 signal. Textbook UASF logic. In 2017, BIP 148 succeeded because miners capitulated: SegWit activation aligned with their interest in avoiding chain split. BIP-110 offers miners nothing but a revenue cut. Ordinals fees have become a meaningful component of Bitcoin's fee market. Restrict non-financial data writes, and you restrict miner income. Miners voted with their hash by doing nothing. No pool with meaningful hashrate moved. The two blocks on the fork chain were likely clock miners or hobbyists. Let me be precise about the sequence, because the order of operations assigns responsibility. First, nodes acted. They implemented UASF-level rejection of non-signaling blocks. This is within their rights as full node operators. It is also performative. Nodes do not mine. They cannot allocate hashrate. Second, miners observed and ignored. The 2.53 percent support rate was not discouraging; it was prohibitive. Third, the fork chain produced two blocks and stalled. There is no security model under which a two-block chain with negligible hash is viable. A 51 percent attack on such a chain would require one competent hobbyist with a rented rig. The core insight emerges when comparing BIP-110 to BIP 148. Both deployed UASF mechanisms. BIP 148 succeeded because it enjoyed convergent interests: miners feared network division more than they feared SegWit itself. BIP-110's failure stems from divergent incentives: miners profit from the very inscription activity the proposal would ban. This is governance's dirty secret. The 'node sovereignty' narrative is real, but proof-of-work remains the final court of appeals. Nodes can write rules. Miners decide which rules receive security. The art is the hash; the value is the proof. BIP-110 provided neither. My years auditing smart contract reentrancy vectors taught me something relevant here. Reentrancy exploits do not depend on syntax; they depend on the order of operations and the reward structure that surrounds them. BIP-110 is a reentrancy-vulnerable governance proposal in exactly that sense. It entered the protocol with a state change the economic majority would never ratify. The reentrancy this time is the fee market itself: inscription transactions flow in, miners collect the fees, and any attempt to cap that flow collides with rational self-interest. When I reviewed the parity multisig library back in 2018, I learned that logic flaws are rarely in the code. They live in the assumptions about who benefits from execution. The same principle governs protocol politics. There is also the question of technical maturity. Two blocks in eight hours, versus 49 on the main chain. A healthy chain in that window produces roughly 48 to 50 blocks. The fork chain achieved less than four percent of that rate. This is not a liveness failure. It is an absence of a chain. Without hashrate, the fork cannot survive a single meaningful reorganization. It cannot secure one transaction. BIP-110 was a subtractive proposal — it added no capacity, introduced no virtual machine, and constrained existing script usage. That class of change demands overwhelming consensus. It received 2.53 percent. The scrutiny from rational miners was immediate and final. The dangerous narrative emerging from this episode is that Ordinals 'won' and the threat is over. That reading is emotionally convenient and analytically lazy. BIP-110 died because miners chose fees over purity today. That is a preference, not a permanent law. If inscription traffic congests the mempool to the point of displacing legitimate financial transactions, or if it begins affecting block propagation under stress, the same miners who ignored BIP-110 could pivot. Miner economic rationality is not a moral commitment. It is a function of the fee schedule at any given moment. The subtler risk is what comes next. The next proposal will not be a frontal assault. It will be a compromise: restricting only large-scale data blobs while preserving small inscriptions. Such an intermediate BIP would hold materially higher odds of miner support. BIP-110's failure does not close this chapter; it educates the next generation of proposers. They now know precisely which thresholds fail and which reward structures survive contact with the hash market. The orthodox camp has demonstrated it will burn political capital on failed attempts. That willingness lowers the expected cost of future attempts. BIP-110 was not a fork. It was an audit of Bitcoin's governance assumptions, and the audit found that economic incentive outranks protocol ideology. The next proposal will be smarter, narrower, and more surgically aligned with what miners actually protect. Ordinals holders should treat this as a stay of execution, not a pardon. We do not build for today — and the same discipline applies to protocol politics. Two blocks are an epitaph, but they are also a rehearsal.