Hook
On July 20, 2025, a single line in a Bitcoin Improvement Proposal changed the network’s political climate forever. BIP-110 proposes to lower the activation threshold for soft forks from 95% to 55% miner hashrate. A 40% reduction. Not a technical optimization — a constitutional amendment. The immediate reaction from core developers, leading holders, and even the mining community has been a wall of rejection. Yet the proposal remains alive, propped by a handful of anonymous contributors and a vocal faction that sees Bitcoin’s data space as polluted by inscriptions and runes. The numbers tell a clear story: miner support stands at less than 2% of the total hashrate. But a UASF — user-activated soft fork — is being whispered in Telegram groups. Fragility hides in the single point of failure: a governance system that has no formal exit strategy.
I do not trust the silence. I audit the code.
Context
BIP-110 is not a new idea. The tension between Bitcoin as pure money and Bitcoin as a data platform has existed since the first OP_RETURN transaction. But the current proposal is different. It bundles two radical changes: (1) a hard limit of 34 bytes on arbitrary data per transaction, effectively killing most inscription-based assets; (2) a temporary restriction on Taproot’s ability to carry complex data structures, pending further research. The technical implementation relies on a modified client that enforces these limits at the consensus layer. A vulnerability — dubbed BlockSlop — was discovered by pseudonymous researcher Dathon Pwn within days of the draft release. The bug allows a trivial fork if any node running BIP-110 encounters a block mined before activation that contains data exceeding the new limit.
Proof precedes value; provenance is the only art. A proposal that cannot pass basic code review has no right to redefine the social contract.
Core
The technical flaws are not the story. The BlockSlop bug will be fixed — maybe. The real analysis is governance. Bitcoin’s 95% threshold is not arbitrary; it is a product of nine years of evolutionary design. It forces any upgrade to demonstrate overwhelming miner consensus before being enforced. Lowering that to 55% is not a minor adjustment. It shifts the balance of power away from the widely distributed mining ecosystem and toward a coordinated minority. In practice, 55% of hashrate can be controlled by fewer than three mining pools. That is a single point of fragility. I have audited enough smart contracts to know that a system is only as resilient as its weakest governance path.
From my experience analyzing the 2017 CryptoKitties code, I learned that invisible vulnerabilities are the most dangerous. BIP-110’s vulnerability is not in the EVM — it is in the voting mechanism. The proposal’s proponents argue that the 95% threshold is a barrier to innovation, that it allows a tiny minority of conservative developers to veto change. That is true. But that veto is precisely what has protected Bitcoin from the kind of rushed, poorly-designed updates that plague other chains. The 55% threshold would turn every ideological disagreement into a potential fork. The cost of that flexibility is network cohesion.
The proposal also targets inscriptions and runes — assets that have generated billions of dollars in economic activity and millions of transactions. Whether one likes them or not, they represent a significant fraction of Bitcoin’s current utility and miner fee income. Freezing them is not a neutral technical decision. It is a regulatory action performed by the protocol layer. Truth is an oracle, not a price feed. Oracles do not censor.
Contrarian
Let me offer the counter-argument — because every good engineer must stress-test their own logic. The case for BIP-110 is not irrational. Inscriptions have bloated the UTXO set by tens of millions of entries, increasing node operational costs. The average transaction fee for non-inscription traffic has risen 15% since the peak of the Ordinals boom. Node operators — especially those in developing regions — report higher bandwidth and storage overhead. If left unchecked, the narrative of Bitcoin as a settlement layer could be eroded by its own success as a data layer.
But here is the contrarian twist: BIP-110 does not solve that problem. It imposes a crude data cap that will simply shift inscriptions to alternative L2 solutions, increasing complexity and centralization. The real solution is better fee markets and L2 scaling, not a return to 2013-era script limitations. The proposal’s advocates are fighting yesterday’s war. The block space is scarce, but scarcity is a feature, not a bug. It forces efficiency. The proposal’s true aim is not technical optimization — it is ideological purity. And ideology, when encoded into consensus rules, becomes censorship.
I do not trust the silence of the chain. I audit the politics.
Takeaway
By early August, the miner signaling period will close. Unless support miraculously rises above 5%, the proposal will fail to meet even UASF’s informal threshold. But the damage is done. The conversation has exposed a growing schism between those who see Bitcoin as a monetary system and those who see it as a programmable asset platform. That division will not disappear. It will resurface in future BIPs, perhaps with better code and more coordination. Bitcoin’s resilience lies not in its code, but in its ability to generate consensus from discord. The 95% threshold is a symbol of that resilience. Breaking it would break a fundamental trust. We do not buy pixels, we buy history. Let us not sell it for a quick political win.
Fragility hides in the single point of failure. The most fragile thing in this debate is not the UTXO set — it is the willingness to compromise on first principles.