Eight hours. Two blocks. One corpse.
Block 961,632 was the flashpoint—the moment Bitcoin's quiet civil war stopped whispering and started bleeding. BIP-110 nodes, running a forced-rule variant that rejected any block lacking their activation signal, split from the main chain. By the time the market blinked, the rebellion had produced exactly two blocks: 961,633 and silence. The main chain didn't pause, didn't flinch, and reached 961,681 while the separatists' chain froze like fossilized ambition.
Run the numbers because the market won't. At Bitcoin's ten-minute average block interval, eight hours should yield 48 blocks. The fork produced two. That's roughly four percent of the hashrate required to maintain a pulse, and even that figure looks generous on a network operating at 500 EH/s. The previous difficulty period delivered 51 supporting signal blocks out of 2,016. That's 2.53%. The proposal's own activation bar was set at 55%. This wasn't a close contest; it was a ritual execution conducted with the compliance of everyone involved.
Let's be clear about what BIP-110 actually is, because the discourse has already confused it with a technical upgrade. It isn't. BIP-110 is a cultural declaration wearing protocol clothing. The proposal targets non-financial data writes on Bitcoin's Layer 1—the technical substrate for Ordinals inscriptions, BRC-20 tokens, and every data-dense experiment that turned Bitcoin's block space into an asset issuance and storage layer. It's the "Bitcoin maximalism" wing's attempt to purge the chain of everything that isn't a financial transaction.
The activation mechanism is where the story gets interesting. This was never a standard BIP-9 miner-activated soft fork, which requires 95% miner signaling over a difficulty adjustment period. BIP-110's own rules demanded only 55%—an implicit admission that it couldn't reach normal consensus. Even that lowered bar proved laughable. Instead, the node implementation executed a User-Activated Soft Fork variant: after a predetermined height, refuse all blocks without the BIP-110 signal. No negotiation. No patience. A flag-day ultimatum handed to the miners.
From my 2017 ICO audit work—when I reviewed 40+ ERC-20 whitepapers in a Vienna dorm room during the frenzy—I learned to smell the difference between a genuine technical contribution and a political weapon wrapped in code. The tell is always the same: when a proposal's activation path relies on coercion rather than consensus, its proponents know the merits aren't sufficient. I killed a €500k seed round back then by finding reentrancy vulnerabilities in a payment gateway's code. The founders were furious. The code was flawed. The market didn't care either way. BIP-110 carries that same scent—a rule change that needs a nuke because it can't win an argument.
The economic logic against it was never subtle. Ordinals transaction fees have become a supplementary revenue stream for miners—real dollars flowing through the fee market. A proposal that eliminates those fees isn't a protocol optimization; it's a pay cut. The miners' veto wasn't ideological. It was a line-item budget decision.
The real story is Bitcoin governance behaving exactly as designed—but not as the "code is law" crowd imagines.
Precision matters here. BIP-110 nodes executed their rule at block 961,632. Miners observed the new chain, computed the opportunity cost, and walked away. The fork chain's hashrate collapsed to near zero. This is Bitcoin's "automatic governance"—a multi-party game where miners hold the final veto through the simple act of not showing up. No committee vote. No civil war. No dramatic declaration. Just absence. The developer-nodes proposed. The miners disposed. The application layer—Ordinals projects, wallets, indexers—never even noticed.
Here's the part the market misreads: this was never a victory for decentralization. It's a demonstration of economic gravity. The failed fork is real-world confirmation that protocol changes on Bitcoin require miner buy-in first, user enthusiasm second, and developer intent a distant third. I sketched a governance triangle during DeFi Summer, when I tracked $2 billion in TVL flowing through Compound and Uniswap V2 yield farms. The lesson then was that liquidity follows incentives. The lesson now is that hashrate flees disincentives with equal mechanical certainty. In 2020, capital moved toward yield. In 2025, miners moved away from revenue destruction. Same physics, different asset.
The tokenomics lens makes the failure overdetermined. BIP-110 wasn't a neutral rule tweak; it was a redefinition of block space property rights. The current model: the highest fee wins the space. BIP-110's model: only financial transactions may rent the space. That's not a technical improvement. That's a redistribution of miner income from data-heavy users to—nobody. The proposal removed a revenue source without creating a replacement. Asking miners to vote for it is like asking a landlord to demolish a rent-paying unit for the aesthetic purity of the neighborhood.
Apply the behavioral modeling I use in my AI-agent payment research, and the outcome becomes even more deterministic. The market here behaved like a distributed agent system, not a human deliberative body. Each miner pool is an autonomous economic actor optimizing a utility function. The BIP-110 fork offered: a chain with four percent hashrate, no exchange listings, no wallet support, no user base, and a governance precedent that would permanently reduce future revenues. The utility calculation returns negative infinity. The pools didn't need to coordinate; each independently computed the same answer and refused. This is emergent consensus through aggregate self-interest—precisely the mechanism I flagged when I found 30% of transaction volume in an AI-agent micropayment protocol was generated by non-human actors exploiting latency arbitrage. It's mechanical, ruthless, and entirely predictable. Human ideology never stood a chance.
What the two-block fork proves is that Bitcoin's security budget is simultaneously a governance firewall. The hashrate that defends against double-spend attacks is the same hashrate that protects the block space economy from unilateral rule changes. Security model and governance model are a single integrated system. I spent the 2022 bear market mapping Terra's collapse onto shadow banking liquidity tightening, and the framework applies here in reverse. Terra taught me that leverage hides in interconnected balance sheets. BIP-110 teaches the inverse: legitimacy hides in hashrate distribution. A fork without hashrate isn't a fork—it's a suggestion.
Now the regulatory subtext, because my role as a cross-border payment researcher made me allergic to missing it. Had BIP-110 succeeded, it would have solved a compliance headache for Western regulators from within the protocol—eliminating Ordinals assets at the consensus layer so the SEC never had to rule on them. The proposal was self-censorship dressed as protocol hygiene. Its failure means the regulatory gray zone persists, BRC-20 assets continue circulating, and American securities enforcement remains a sword hanging over the ecosystem. The auditor in me notes that the fork chain's "Bitcoin"—code-identical, consensus-absent—has an economic value indistinguishable from zero. I flagged this exact trap after my 2024 ETF regulatory arbitrage work with five compliance officers across Europe: fork tokens are honeypots for the uninitiated. Exchanges that list them without a hashrate and liquidity audit are facilitating a value-destruction event.
Here's the take the pro-Ordinals crowd doesn't want to hear: the fork's failure is a short-term victory but a long-term warning. The BIP-110 route is dead for at least two years. But the idea didn't die. It will resurface as economic warfare, not protocol warfare.
Watch for the next generation of "block space purification" to arrive through fee structure reform or voluntary miner-side filtering. No fork required. No consensus signal needed. Just quiet coordination among major pools to deprioritize data-heavy transactions until inscriptions price themselves out of existence. That's the sophisticated threat to Ordinals, and it's harder to vote against because it never appears as a proposal. It appears as a mempool policy. The governance battlefield shifts from GitHub to block templates.
The other blind spot worth naming: the unidentified miners who produced those two blocks. They weren't economic actors—they were ideological ones, likely from the "clean Bitcoin" faction. Their two blocks are a canary in the coalmine. Miner consensus isn't monolithic, and the next confrontation may split pools along philosophical lines rather than profit lines. A split on economics is resolvable. A split on identity is not.
One more uncomfortable observation: the proposal's failure doesn't signal health for Bitcoin governance. It signals that the social contract now runs on raw economic power. The market should be mildly terrified that a 2.53% minority can trigger a network split event in the first place. The system worked this time. That's not a guarantee it works next time. The audit blinked once; the hashrate never did.
The auditor blinked; the market didn't. Narrative is not consensus. Hashrate is. Liquidity doesn't forgive miscalculation, and it won't remember BIP-110 by next quarter. Position accordingly: Ordinals' protocol-level tail risk has been retired, but the next battle won't be broadcast on GitHub. It'll happen silently, in fee markets and mempool policies, where the market isn't watching because the market is still staring at the last war. That's the trade. The next threat comes from the direction you're not looking.