Block 961,632. That's where the mutiny ended before it began.
BIP-110 nodes — running a modified Bitcoin client — started rejecting blocks that lacked their activation signal. At that exact height, the network split. Eight hours later, the breakaway chain had produced exactly two blocks. The main chain, entirely unbothered, had advanced to 961,681.
Do the arithmetic. Ten minutes per block is Bitcoin's baseline cadence. Eight hours should yield forty-eight blocks. Two is 4.2 percent of that. Roughly one miner in twenty-four pointed hashrate at the fork. The remaining 96 percent of the network simply kept building blocks without the BIP-110 signal — blocks the rebel nodes now reject.
The irony is clean. The rebels separated from the network. The network never noticed them leaving.
This wasn't a fork. This was a footnote.
Context: What BIP-110 Actually Tried to Do
BIP-110 needs context before the numbers mean anything. The proposal targeted non-financial data writes on Bitcoin's block space — specifically Ordinals inscriptions and the BRC-20 token ecosystem that exploded onto Layer 1 starting in early 2023.
The policy argument is coherent. Bitcoin is a settlement network. The most secure chain ever built should not become distributed file storage for JPEGs and token launchpads. Inscriptions congest blocks, consume block space, and distort the fee dynamics that have held since genesis. For maximalists — the camp that views Bitcoin as pure digital cash, provably hard money — Ordinals is rent extraction dressed as innovation. It pollutes the ledger.
BIP-110 proposed a consensus-level cure: prohibit non-financial data at the protocol layer. Nodes running the updated client would refuse to validate blocks containing inscription-like data. No ambiguity. No opt-out. Code says no.
Here's where the governance disaster begins. Standard Bitcoin upgrades follow BIP-9, a miner-signaling mechanism where proposals activate only after roughly 95 percent of miners signal within a difficulty window. BIP-110 set its own threshold at 55 percent — then bypassed the mechanism entirely.
The activation path was UASF: a user-activated soft fork. Node operators impose new rules unilaterally. Miners are forced to choose between adopting the rule or facing perpetual block rejection. It's governance by ultimatum.
The signaling data made the outcome predictable before the first fork block was mined. In the previous 2,016-block difficulty period, BIP-110 appeared in 51 blocks. That's 2.53 percent support. The proposal demanded 55 percent. It was more than twenty times short of its own activation threshold.
2.53 percent. The authors knew. The node operators who triggered the split knew. And they still forced the fork at 961,632.
That's not a governance process. That's a temper tantrum with cryptography.
Core: Hashrate Is the Only Vote That Counts
Three data points define this failure. Prior-cycle support: 51 of 2,016 blocks, 2.53 percent against a 55 percent threshold. Post-fork production: two blocks in eight hours against a theoretical forty-eight. Implied hashrate share: roughly four percent — and dropping every minute the fork stayed alive.
A chain with four percent hashrate isn't insecure. It's effectively nonexistent. At that level, the fork is vulnerable to a successful 51 percent attack by any mid-sized pool on the main network. It's vulnerable to block withholding, double-spend races, and reorganization at will. The security assumption that protects Bitcoin — enormous, redundant, distributed hashrate — doesn't transfer to a breakaway chain that a weekend miner could bury.
The technical autopsy points to deployment failure, not code failure. BIP-110's cryptographic primitives were sound. Its implementation was operationally coherent. The chain failed because it carried no social license, no miner mandate, no economic constituency.
You cannot fork a social contract with a software release. This is the same lesson I learned during the 2017 smart contract audit sprint, when a critical integer overflow in the HotCo protocol taught me that code correctness means nothing if the surrounding system rejects you. Technical quality is necessary. It is never sufficient.
Now the UASF layer deserves deeper treatment, because it's the piece most post-mortems will gloss over.
BIP-9 activation is a coordination mechanism: miners signal, consensus forms, rules activate. The security model survives because the people securing the network chose the change. BIP-110 inverted this. The client enforced the rule first, asked questions later. The node layer unilaterally declared new consensus rules — the "code is law" narrative weaponized.
In Bitcoin, that approach fails every single time. Nodes don't produce blocks. Miners do. When the execution layer refuses to cooperate, the new rule is an empty declaration.
BIP-110 nodes were running what amounts to a flag-day trigger: after a specific height, unconditional enforcement, no further signaling required. The deadline was baked into the software. When it hit, the fork ignited. The market answered with two blocks and silence.
Compare this to actual successful forks. BCH in 2017. BSV in 2018. Whatever you think of those projects, they had real constituencies — exchanges, miners, capital, users. BIP-110 had a client fork and a blog post. The outcome asymmetry wasn't surprising. It was mechanical.
One regulatory thread deserves attention here. BIP-110 was, in part, an attempt to solve a compliance problem through consensus code. US regulators have scrutinized NFT-like assets; inscriptions and related marketplaces occupy a gray zone under the Howey framework. A protocol-level ban would have surgically removed the entire asset class. Problem solved by elimination. But the failure of BIP-110 means regulators will continue to evaluate Ordinals through conventional securities law — a messier path, with enforcement actions forming common law case by case.
Now the economic layer underneath the hashrate math — the part that actually matters going forward.
Miners didn't abandon BIP-110 because they love inscription technology. They abandoned it because BIP-110 was a self-imposed pay cut.
Ordinals inscription fees have become a non-trivial revenue stream since 2023. In the post-halving environment — block subsidies permanently reduced — transaction fees are the only growth vector miners have. Inscription-driven congestion pushed feerates into profitable territory for smaller operations competing against industrial-scale players. BIP-110 would have severed that revenue at the consensus level.
The economic logic is merciless: a proposal that cuts your income hits an immovable wall. Miners compute this in seconds. They don't need a governance forum. They vote with hashrate. This is the most efficient poll in financial history: 96 percent rejection, zero press releases.
And here's the hidden coalition forming in the ashes. Ordinals users want block space. Miners want fees. Both benefit from permissive data inclusion. The anti-inscription camp correctly identified their ideological opponent while completely misreading the economic battlefield. The proposal attacked the miners' revenue line. It lost before the first block was rejected.
There's also liquidation risk hiding in this story. A fork coin with four percent hashrate carries zero credible value. If any exchange mistakenly lists the breakaway token, its price approximates zero within hours — reorgs make settlement fictitious, liquidity dries up, holders eat the loss. This is a honeypot, not an asset. If you see BIP-110 "BTC" appear on a minor trading venue, remember the lesson of every failed token fork: a red candle doesn't lie, and neither does a stalled block height.
A low-probability scenario worth tracking: if even a handful of nodes continue mining BIP-110 blocks at a trickle rate, a zombie chain could persist — a symbolic outpost for the anti-inscription movement. It will carry no listings, no liquidity, no users. But its presence keeps the ideological conflict alive and provides a cost baseline for any future attempt. That's the blockade state of Bitcoin governance.
Contrarian: The Defeat That Strengthens the Next Assault
Here's the layer no one's reporting. BIP-110's failure actually strengthens the anti-Ordinals movement in the long run.
The direct assault failed. But the lesson will be operational: don't fight the consensus layer directly. Fight the incentive structure instead. The next wave of anti-inscription policy won't arrive as a fork. It will arrive as mempool policy. Mining pool feerate filters. Transaction ordering discrimination. Economic pressure that makes inscription transactions unprofitable without touching a single line of consensus code.
That's much harder for the Ordinals ecosystem to defend. You can't fork away a mining pool's business policy. You can't BIP your way out of an economic blockade.
The second contrarian insight concerns power structure. This event proves the mining layer holds de facto veto authority over any consensus change — even proposals with ideological support inside the developer community. That's a profound statement about Bitcoin governance. Developers propose. Miners dispose. Any party ignoring that feedback loop is burning capital on a political impossibility.
Surveillance isn't just watching blocks. It's anticipating the break before it happens. The break happened at 961,632. The anticlimax lasted eight hours. The next break will come from the fee market, not the consensus code. And when it does, the Ordinals ecosystem's protection narrative evaporates.
Takeaway: The Trap Just Closed
BIP-110 failed because it attacked the execution layer's bottom line. That's the most reliable veto in Bitcoin.
For BRC-20 infrastructure, this is a tail-risk removal — but the war continues on a different battlefield. Watch mining pool policies. Watch feerate structures. Watch for OP_CAT and CTV proposals that redirect developer energy rather than confront consensus head-on.
Yield is the bait; liquidity is the trap. In Bitcoin governance, the trap just closed on the impatient. Arbitrage is the market's tax on the slow. The next trade isn't in the ledger. It's in the mempool.