Michael Saylor didn't oppose BIP-110 because of what it restricts. He opposed it because of how it could be passed.
Over 110 reasons spilled out in a recent public statement—each one a cold, mathematical dissection of why a proposed upgrade to Bitcoin’s consensus layer poses an existential risk. But the core argument wasn't about script limits or witness data caps. It was about the activation mechanism: a 55% miner threshold with no FAILED state. A vote that could pass without supermajority, without a clean failure option, creating a new precedent for low-consensus changes. This isn't a technical debate; it's a narrative shift in security—and Saylor is warning us that the pattern matters more than the patch.
Context: The Proposal and Its Flaw
BIP-110 is a Bitcoin Improvement Proposal designed to introduce seven specific consensus rules limiting the use of script public key lengths, witness data quantities, and certain Taproot extension paths (BIP-342). Ostensibly, it targets data bloat on the blockchain—the kind of bloat driven by Ordinal inscriptions and other non-financial uses of block space. The technical changes are incremental, restrictive, and arguably address a real concern about resource consumption. But the proposal’s governance architecture is anything but ordinary.
Unlike the historical BIP-9 activation process, which required 95% miner signaling within a retargeting window and a clear FAILED state for expired proposals, BIP-110 lowers the activation threshold to 55% and eliminates the FAILED outcome. If 55% of miners signal support, the rules activate after a fixed period—no matter the opposition. There is no mechanism for the proposal to fail gracefully; it either passes or remains in limbo, a perpetual threat of non-consensus activation. This design, as Saylor highlighted, creates a new governance primitive: a low-barrier soft fork that could be deployed without broad community consent.
I’ve spent years modeling liquidity bottlenecks in DeFi protocols, and I can tell you that activation thresholds are the most fragile component of any consensus system. During my 2020 work on Curve’s CRV emissions, I learned that even a 1% shift in miner incentives can cascade into a new equilibrium. BIP-110’s 55% threshold is dangerously close to the point where a coalition of large pools—say, Foundry, Antpool, and F2Pool—could force a change against the will of smaller miners and the wider node network. Restaking isn't a narrative shift in security, but this activation mechanism might be.
Core: The Precedent Risk
Saylor’s opposition is not about the technical restrictions themselves. He explicitly argues that non-consensus solutions—higher transaction fees, Layer 2 networks, node-level policy filters—can address data bloat without tampering with the base layer. The real risk, in his view, is that BIP-110 establishes a new governance norm with long-term consequences. Once the community accepts a 55% activation and no FAILED state, future proposals with far more invasive changes could use the same mechanism. The Pandora’s box opens not with the content of BIP-110, but with its process.
This is a narrative shift in how we view Bitcoin governance. Historically, the network’s immutability was protected not just by cryptographic security but by a high bar for change. The 95% threshold ensured that upgrades were effectively unanimous among miners, reflecting deep consensus. BIP-110’s 55% threshold lowers that bar to a simple majority, vulnerable to capture by a coordinated minority. I’ve seen this pattern before: in 2022, the Terra collapse taught me that governance mechanics—not just code—determine resilience. Anchor’s yield mechanism failed because it lacked a safety valve. BIP-110’s lack of a FAILED state is the same kind of design flaw—a missing exit that turns a reversible discussion into a permanent fork risk.
From a technical perspective, the seven restrictions in BIP-110 are moderate. Limiting script public key lengths to 64 bytes or reducing witness data per input are defendable optimizations. They might even be beneficial if the goal is to discourage inefficient use of block space. But the cost of activating them through a flawed governance process outweighs any short-term gain. Saylor’s critique is structural: the proposal’s governance is more dangerous than the problem it solves.
Let me ground this in data. The real narrative shift lies in the threshold, not the restriction. Under BIP-110’s rules, a hypothetical coalition of just three major mining pools could activate the upgrade. Today, Foundry USA (28% hash rate), Antpool (22%), and F2Pool (15%) together control 65% of Bitcoin’s hashrate. They could easily reach 55% signaling. The remaining 35% of miners—including smaller pools and solo miners—would be forced to either follow the new rules or risk building on a minority chain. If enough nodes reject the change, you get a chain split. Without a FAILED state, the proposal never expires; it hovers as a constant threat, polarizing the community over months or years.
I encountered a similar governance fragility while modeling Ethereum’s migration to proof-of-stake in 2022. The inability to fail cleanly created uncertainty that suppressed validator participation. Bitcoin’s design has always avoided that trap by making upgrade failure explicit. BIP-110’s authors seem to have overlooked this principle, perhaps intentionally to fast-track the change. But as my simulation of slashing conditions for EigenLayer restaking showed in 2023, incentive alignment is everything. A low-threshold activation creates misaligned incentives: miners who benefit from reduced data costs may support it, while those who profit from high-fee inscriptions oppose it. The 55% threshold allows the former group to override the latter without negotiation.
Contrarian: The Case for BIP-110–And Why It’s Still Wrong
Proponents of BIP-110 argue that the 55% threshold is a pragmatic response to gridlock. Bitcoin’s governance has become ossified, they say; reaching 95% consensus for any change is nearly impossible. Lowering the bar allows the network to adapt to real-world issues like blockchain bloat, which threatens node operation costs and long-term decentralization. From this perspective, BIP-110 is not a power grab but a necessary evolutionary step. The absence of a FAILED state forces the community to engage seriously rather than letting proposals languish indefinitely.
There is some merit to this view. The 95% threshold has indeed slowed upgrades—witness the years-long debate over Taproot activation itself. But the counter-argument is stronger: ossification is a feature, not a bug. Bitcoin’s value proposition is predictability. Changing the activation rules for the convenience of one proposal sets a precedent that undermines that predictability. The 2024 ETF regulatory arbitrage I tracked in Australia taught me that market participants price in governance stability. Any sign of governance malleability reduces Bitcoin’s premium over other assets. BIP-110’s activation mechanism is a signal that the network can be changed by a minority—a signal that, once broadcast, cannot be easily retracted.
Moreover, the technical problem BIP-110 aims to solve can be addressed by non-consensus methods. Node operators can already reject large transactions by setting policy limits. Layer 2 networks like Lightning and RGB can absorb inscription activity without clogging the base layer. The market, through fee pressure, will naturally incentivize efficient use of block space. Saylor’s reliance on these alternatives is not just a preference—it is a defense of Bitcoin’s conservative upgrade philosophy.
Takeaway: The Next Narrative
Watch the miner signals. If BIP-110’s activation threshold becomes a topic of serious debate, the narrative will shift from 'what it changes' to 'how it passes.' The real fight is not about script lengths or witness caps—it's about whether Bitcoin can survive a change in its governance culture without fracturing. I suspect the community will ultimately reject BIP-110, not because of its technical content, but because the mechanism is too dangerous. But the battle itself has exposed a vulnerability: the network’s upgrade process is only as strong as its weakest threshold. If this proposal dies, a similar one with a more palatable activation process could emerge. The precedent Saylor fears is already being set by the discussion itself.
Restaking isn't a narrative shift in security—but BIP-110's governance mechanism might be the first step toward one. The next narrative will be about how we protect Bitcoin from its own protectors.