Michael Saylor does not write blog posts to make friends. On July 17th, the MicroStrategy chairman published a 4,200-word rebuttal to BIP 110, a Bitcoin improvement proposal that, on the surface, aims to limit script size, Taproot control blocks, and undefined witness versions. Saylor's thesis was simple: this proposal is not about reducing data or preventing denial-of-service attacks. It is about protocol overreach.
Saylor's criticism landed like a ledger audit. He called the technical solution a "crude proxy for an unmeasured cost." He argued that the proposal's bundled rules—each with different technical motivations—violate the principle of neutrality. He warned that BIP 110 would permanently cut off future innovation paths, including BitVM, a theoretical framework for Turing-complete computation on Bitcoin.
The timing matters. We are in a bull market. Bull markets amplify FOMO and suppress critical thinking. Saylor's intervention forces a pause. As an independent investigative journalist with a PhD in cryptography and fifteen years of blockchain forensics, I have seen this pattern before. Teams bundle controversial changes with unobjectionable ones. They exploit low activation thresholds. They claim to solve a problem that has never been quantified. BIP 110 is a textbook case of technical governance risk masked as optimization.
Context: The Proposal and Its Mechanics
BIP 110 is authored by an anonymous or little-known developer. It reached "Complete" draft status on the Bitcoin Improvement Proposal repository, but it is far from community consensus. The proposal imposes three restrictions: - Maximum script size for new UTXOs is capped. - Taproot control block size is capped at 520 bytes. - New witness versions (v1 and above) must not exceed 10,000 bytes.
These restrictions are enforced via a soft fork. Activation requires 55% of miner signaling over a difficulty adjustment period—a low threshold compared to the historical 95% for previous upgrades like SegWit or Taproot. The proposal includes a one-year automatic expiry. But the low threshold, combined with the bundled rules, creates a governance loophole: a minority of hashpower could force a controversial change onto the network.
Proponents argue that BIP 110 reduces UTXO set bloat and mitigates DoS attacks. They claim it lowers node operational costs. They frame it as a maintenance upgrade.
But the data is absent. No comprehensive measurement of cost savings exists. No formal security analysis has been published. The proposal relies on an anecdote: that large scripts and large witnesses are wasteful. This is not engineering. It is ideology dressed as optimization.
Core: A Systematic Teardown
1. The Technical Flaw: Proxies Without Metrics
Saylor identified the core problem: the proposal targets proxies—script size, control block size, witness size—rather than the actual cost it claims to reduce. In cryptography and protocol design, you never optimize for a proxy unless you have proven the correlation. BIP 110 provides no such proof.
From my experience auditing ICOs in 2017, I learned that whitepapers often invent problems to justify solutions. In 2020, during the DeFi rug pull investigation, I traced liquidity manipulation back to a hidden backdoor. The pattern repeats: a technical measure that sounds reasonable, but lacks empirical foundation.
2. The Bundling Problem
BIP 110 bundles three unrelated restrictions into one soft fork. This is poor engineering. Each restriction should have its own BIP, its own rationale, its own activation signal. Bundling forces the community into an all-or-nothing vote. It is a political tactic, not a technical one. Saylor's blog post implicitly accuses the proposal of "horse-trading": attaching a minor, acceptable rule to a major, controversial one to force a passage.
3. The Future Cost: Cutting Off BitVM
BitVM is a proposed method to verify arbitrary computations on Bitcoin using Taproot trees and fraud proofs. It does not require a soft fork. It uses existing Bitcoin script capabilities—but within the bounds of current witness size limits. BIP 110's caps on control blocks and witness versions would kill BitVM. The innovation would be permanently blocked. This is not theoretical. BitVM's developers have confirmed that the proposal would render their work infeasible.
4. The Activation Threshold: 55% Is a Dangerous Precedent
Bitcoin's governance tradition requires overwhelming miner consensus—95% for soft forks—to ensure network stability and prevent chain splits. BIP 110 lowers this to 55%. This is a de facto change to Bitcoin's social contract. A 55% threshold means that a small majority of hashpower could impose a change that a large minority plus node operators oppose. It creates a slippery slope: future proposals could use the same low bar to push through even more controversial rules.
5. The Game Theory: Who Benefits?
BIP 110, if passed, reduces the maximum size of certain transactions. This could reduce block space demand and lower fees. Lower fees benefit users who send small transactions. But they hurt miners who depend on fee revenue post-subsidy halving. The proposal does not incentivize miners to signal for it—it actually reduces their long-term income. So why would miners support it? Possibly because the proposal is backed by a faction that prioritizes low fees above miner sustainability. Or because the proposer hopes miners will signal without understanding the economics.
Saylor's model handles the same problem—large UTXOs and potential spam—without a soft fork. He advocates for variable fee rates and voluntary relaying by nodes. This is market-based, not rule-based. It preserves neutrality. It allows innovation (like BitVM) to proceed if users choose to pay for it. It does not require a chain change.
Contrarian: What the Bulls Get Right
Despite my skepticism, I must acknowledge the valid concerns that motivate BIP 110.
First, UTXO set growth is a real issue. Bitcoin's unspent transaction output set has grown from 50 million in 2020 to over 80 million today. Larger wallets and Taproot usage have increased storage requirements. Node operators, especially those on low-resource hardware, face higher costs. If running a full node becomes too expensive, decentralization suffers. BIP 110's advocates are right to be concerned.
Second, DoS attacks using large scripts are possible. An attacker could craft oversized transactions that are valid but expensive to validate. The network has experienced fee spikes from such attacks. Limiting script size is a partial deterrent. However, the better solution is to let nodes set their own policies, as Saylor proposed, rather than enforce a hard rule.
Third, the proposal includes a one-year expiry. If it does not gain enough miner support, it dies automatically. This safety mechanism reduces the risk of a permanent failure. The proposal is not irreversible.
But these valid points do not justify the bundled approach or the low threshold. The problem is real; the proposed solution is flawed.
Takeaway: The Accountability Call
BIP 110 will likely stall within weeks, as Adam Back predicted. The lack of data, the bundling, and Saylor's forceful opposition create too much friction. But the debate reveals a deeper vulnerability in Bitcoin's governance. Bull market euphoria makes us complacent. We assume that Bitcoin's stability is inherent, not contingent on active defense.
Ledger balances do not lie; they only wait. This proposal is a test. If it passes, Bitcoin loses its neutrality. If it fails, the community reaffirms that protocol changes must be backed by measurable data, not ideological hunches.
The question is not whether BIP 110 activates. The question is whether the next proposal will be better designed—or whether the 55% threshold becomes a new norm. Hype evaporates; receipts remain. The receipt for BIP 110 is code with no proven cost, bundled rules, and a governance loophole. That is the record.