Tracing the invariant where the logic fractures. Over the past seven days, a single version bit has flickered across 2.64% of Bitcoin blocks. The signal is for BIP-110, a soft fork proposal that aims to cap transaction data fields and choke Ordinals inscriptions. The forced signaling window—a hard deadline set at block height 876,000—is roughly 12,000 blocks away. The code has been written, the binary is compiled, but the network’s hash power has already delivered its verdict: this fork will not activate. Yet the story is not about failure. It is about the structural fault lines in Bitcoin’s upgrade mechanism, exposed by a proposal that was never designed to succeed.
Context: What BIP‑110 Actually Does
BIP‑110, officially titled “Reduced Data Temporary Softfork,” modifies two consensus rules. First, it limits the maximum number of witness stack elements in a transaction to five, with a combined size of 80 bytes. Second, it restricts OP_RETURN output scripts to 80 bytes. These constraints directly target the storage of large data payloads in SegWit’s witness fields and OP_RETURN outputs, the primary vehicle for Ordinals inscriptions and other NFT-like metadata. The proposal is a temporary soft fork—it would automatically expire after roughly 12 months unless extended.
The activation mechanism borrows from BIP‑8’s “lock-in” model but adds a forced signaling window. After the window opens, nodes running BIP‑110 code will treat any block that does not set the required version bit (bit 4) as invalid. Miners must explicitly signal support by setting that bit; otherwise, their blocks are rejected by upgraded nodes. This is a binary enforcement: either the entire network adopts the new rules, or a minority chain emerges where upgraded nodes follow only signaling miners, while non-upgraded nodes follow the longest chain (which may contain non-signaling blocks).
Currently, only a handful of miners—Ocean and a few small participants—are signaling. The support rate hovers at 2.64%. Major pools like Foundry USA, Antpool, and F2Pool have remained silent. Their silence is not neutrality; it is a tacit rejection. Given that activation requires 95% hash power to signal before the forced window ends, the proposal is effectively dead on arrival.
Core: Dissecting the Code and the Governance Trap
From my review of the BIP draft and the corresponding Bitcoin Core pull request, the implementation is clean and minimal. The logic amounts to a single validation check: if the transaction has more than five witness stack elements or if any element exceeds 80 bytes, the block is rejected. The same applies to OP_RETURN outputs. This is a simple parameter adjustment—what engineers call a “soft fork by constraint.” No novel cryptography, no new opcodes, no consensus complexity. The technical risk is low, and the deployment cost is near zero.
But the activation mechanism is where the fractures appear. Let me trace the invariant: the forced signaling window is a governance weapon disguised as a technical deadline. Unlike BIP‑9, which required 95% signaling within a retarget period to activate (and allowed miners to remain indifferent), BIP‑110’s forced window creates a binary choice. After height 876,000, any miner that does not signal is effectively producing blocks that will be orphaned by upgraded nodes. This is not a vote; it is a mandate.
The problem is economic coordination. Even with 2.64% hash power, if a critical mass of exchanges, wallets, and node operators upgrade to BIP‑110, they could enforce the rule on the network. Non-signaling miners would see their blocks rejected by the economic majority, forcing them to either signal or redirect power to a minority chain. In theory, the economic majority (users, exchanges, miners) always favors the status quo because a minority chain has lower security and liquidity. But the forced window changes the game: it creates a short-term incentive for small miners to signal (to avoid being orphaned), while large miners can simply wait out the window by continuing to mine non-signaling blocks on the side.
From real data, the risk of a persistent split is negligible. Even if all nodes upgraded, the non-signaling hashrate (97.36%) would quickly outpace the signaling minority, and the network would converge on the longest chain. However, the forced window introduces a temporary period of uncertainty. Exchanges might suspend BTC deposits to ensure chain integrity, wallets might see reorgs of a few blocks. The market hates uncertainty, even if short-lived.
Friction reveals the hidden dependencies. Here, the dependency is the assumption that miners will voluntarily signal for upgrades they oppose. BIP‑110 assumes cooperation; the forced window demands compliance. When cooperation is missing, the compliance mechanism becomes a coercion tool. This is the core technical insight: the forced window is a vulnerability in the governance protocol, not just a minor upgrade.
I ran a simulation of the forced window scenario using a simplified Python model (available on my GitHub). The results show that with less than 10% signaling, the upgraded chain produces blocks at a rate proportional to its hash share. After 100 blocks, the non-signaled chain is ~97 blocks ahead. Exchanges following the heaviest chain (cumulative difficulty) will naturally favor the non-signaled chain, rendering the upgraded chain irrelevant. But if an exchange chooses to follow the upgraded chain for policy reasons, it creates a temporary arbitrage between two versions of BTC—a scenario reminiscent of the 2017 SegWit2x split attempt. In that case, market forces quickly resolved the split, but not before causing a week of price volatility.
Contrarian: The BIP‑110 is Not About Ordinals
The conventional narrative frames BIP‑110 as an attack on Ordinals, a tool to “clean” Bitcoin of non-financial data. But the contrarian angle is that Ordinals are a pretext for a more dangerous experiment: testing whether a minority can force a consensus change through code rather than social agreement. The forced window is a radical departure from Bitcoin’s tradition of soft fork activation, which historically required broad miner and economic node support. By making the window mandatory, BIP‑110’s authors are essentially saying: “We will enforce our rule even if the majority disagrees, because we control the code that nodes run.”
Reverting to first principles to find the break. Satoshi’s whitepaper emphasized proof-of-work as the ultimate arbiter of truth: the longest chain is the valid chain. The security model depends on rational miners following the most profitable path, which in turn aligns with economic majority. A forced signaling window disrupts this equilibrium. It creates an artificial incentive to signal (or else be orphaned), but if the majority does not signal, the system splits. This is not a flaw in the soft fork; it is a feature of the activation design exposed by low support.
The deeper risk is precedent. If a forced window can be used to push an unpopular soft fork, future proposals with stronger support (say, 30% hash power) could use the same mechanism to force adoption. The network would then be governed not by hash power or economic weight, but by who can write code that gets adopted by node operators. This shifts Bitcoin’s governance from open consensus to code-centric authoritarianism. BIP‑110, even in failure, opens the door to that path.
Takeaway: The Forced Window Is a Bug, Not a Feature
BIP‑110 will expire without activation. The forced window will open, non-signaling blocks will continue, and the upgraded nodes will despair. The proposal will be remembered as a footnote in Bitcoin’s governance history. But the metadata is memory, and code is truth. The truth here is that Bitcoin’s upgrade path is becoming increasingly fragmented. The forced window template is now available for future uses. Expect other factions to adopt similar mechanisms for their own causes—L2 extensions, opcode changes, or even fee market adjustments.
The market should not price this event, but developers should note the signal. The 2.64% support is a canary. It shows that a non-negligible part of the mining community is willing to use aggressive activation measures. If that number grows to 10% or 20% in future BIPs, the risk of a real split increases. For now, the chain is safe. But the governance fracture is real. Watch the signaling rate after the window opens; if it jumps above 10%, prepare for short-term volatility. Otherwise, ignore the noise and focus on the code.
Precision is the only reliable currency. BIP‑110 fails on precision: its activation assumption is broken, its support is marginal, and its mechanism introduces unnecessary risk. The network will move on. But the governance lesson remains: forced signaling is a fragile invariant. And when that invariant fractures, Bitcoin’s unity fractures with it.