BIP-110: The 55% Threshold That Exposes Bitcoin's Governance Fracture

A proposal to cut Bitcoin’s soft fork activation threshold from 95% to 55% is not a technical tweak—it is a direct assault on the network’s most sacred consensus mechanism. On July 20, 2025, BIP-110 surfaced with a dual agenda: restrict arbitrary data in blocks (a nod to the ongoing inscription spam war) and, more importantly, redefine who gets to decide when the rules change. Within days, a consensus bug—dubbed BlockSlop—was discovered. The code that promised to ‘clean up’ the ledger instead introduced a path to chain divergence. This is not a debate about spam. It is a battle over Bitcoin’s soul. To understand why BIP-110 matters, you have to look past the surface. The proposal, authored anonymously, targets two pain points that have plagued bitcoiners since the Ordinals craze began in 2023. First, block space is being consumed by millions of tiny inscription-related transactions, inflating UTXO sets and slowing down low-end nodes. Second, the 95% miner signaling threshold—long considered the gold standard for soft fork activation—is now labeled as ‘paralyzing’ by a faction that wants faster upgrades. BIP-110 offers a package: limit new transaction data to 34 bytes, restore OP_RETURN’s 80-byte limit, temporarily disable Taproot’s data embedding capabilities, and lower the activation bar to 55%. In theory, it is a surgical strike against ‘non-monetary’ use of Bitcoin. In practice, it is a crowbar aimed at the network’s governance foundation. The systemic teardown starts with the numbers. According to pooled miner signaling data from early July, support for BIP-110 hovers below 3% of total hashrate. Compare that to the 95% threshold that SegWit and Taproot required—both of which enjoyed months of community testing and formal review. BIP-110 has neither. The BlockSlop vulnerability, independently confirmed by developer Dathon Pwn, means that any node running the modified client can diverge from the canonical chain when replaying old blocks. This is not a theoretical risk; it is a ticking consensus bomb. During my years auditing upgrade proposals, I have never seen a flagged soft fork that still had an unpatched critical bug being actively debated in the open. The fact that the authors pushed the code without basic fuzzing tells you everything about the technical rigor behind this initiative. Let me be specific: the 34-byte limit on new outputs is designed to cripple the inscription market without explicitly banning it. But the side effect is a potential freeze on UTXOs created under the current Taproot rules. If a user holds an inscription that relies on the existing 520-byte witness data structure, that UTXO becomes unspendable under the new rules—a silent confiscation of digital property. The proposal’s own documentation admits this is a ‘temporary limitation,’ but there is no mechanism to grandfather existing assets. In the Ethereum Parity heist forensics I led back in 2017, I learned that even well-intentioned code changes can orphan user funds when edge cases are ignored. This is the same pattern: a supposedly ‘clean’ rule that assumes all non-standard outputs are malicious. The blockchain does not lie—every frozen UTXO will be a scar on the chain. Now consider the governance angle. Lowering the threshold from 95% to 55% is not about efficiency; it is about power. Historically, 95% ensured that any change had overwhelming economic and hash power consent. SegWit’s activation saw over 90% support from miners and nodes before it locked in. Taproot reached similar consensus after years of discussion. BIP-110’s 55% gives a bare majority the ability to force a network-wide rule change on a minority that may still control significant hash and user share. This is a recipe for UASF (User-Activated Soft Fork) counter-reactions, which the proposal explicitly anticipates. It is, in effect, an invitation to a chain split. The author’s claim that ‘55% represents the will of the community’ is mathematically absurd when the same community is 97% opposed to the proposal. Opposition from established voices is sharp. Jameson Lopp, a long-time Bitcoin Core contributor and CTO of Casa, publicly stated that the proposal ‘creates a censorship precedent’ and would ‘freeze Layer 2 development.’ Michael Saylor, the largest corporate holder of Bitcoin, warned that any reduction in the activation threshold undermines the ‘digital gold’ narrative. Their combined weight in both technical credibility and market influence means that BIP-110 has zero chance of passing through the normal signaling process. But the proposal does not need to pass to cause damage. The mere existence of a formal BIP with a 55% threshold has cracked open a door that was previously sealed. Future proposals may now treat 55% as a plausible target, eroding the norm of supermajority consent and making Bitcoin’s upgrade process resemble that of Proof-of-Stake chains, where a 51% vote can change fundamental parameters. Here is the contrarian angle that most critics refuse to acknowledge: BIP-110’s supporters have a real point about data bloat. The current inscription ecosystem consumes over 10% of block space on average, with spikes during popular collections. This is not free speech—it is externalized cost on node operators and miners who must validate and store data that has nothing to do with monetary transactions. The UTXO set has grown by 40% since Ordinals launched, and pruning is impossible because the outputs remain unspent. If you are running a Bitcoin node on a Raspberry Pi, you have felt the slowdown. The argument that ‘anyone can use the block space they pay for’ ignores the fact that the same space is a shared resource. In my quantitative analysis of transaction fee data over the past 12 months, I found that low-value inscription transactions often outbid high-value payments during congestion, effectively taxing legitimate users. BIP-110, for all its flaws, is the first serious attempt to price in the externality of non-financial data. The bullish case here is that some form of data restriction is inevitable—the only question is how and when. But the method matters. A 55% threshold without a working implementation and with a known consensus bug is not a solution; it is a weapon. The proper fix—if the community decides one is needed—should follow the same rigorous path as any other soft fork: public discussion, reference implementation, testnet deployment, audit cycle, and then 95% miner signal. That process has served Bitcoin well for over a decade. Skipping it because of frustration with inscriptions is like burning down your house to kill a cockroach. The cockroach may die, but you lose the shelter. The takeaway is clear: BIP-110 will almost certainly fail. Miner support is negligible, core developers are united against it, and the technical flaws are too severe to patch before the August activation window closes. But the scar will remain. The ledger now bears a permanent record of a proposal that dared to lower the bar from 95% to 55%. Future debates will cite this as a precedent—not for success, but for the kind of instability that a poorly designed change can trigger. Bitcoin’s governance is not a democracy; it is a rough consensus of energy, code, and economic weight. BIP-110 tried to rewrite that balance. The numbers have no emotions, only consequences. And the consequence here is a wake-up call: the next time someone offers a ‘quick fix’ for spam, ask who really controls the 55%.