Thirty-eight. Then forty-seven. Eight new signals in 392 blocks. BIP-110 has 217 blocks left before every block without versionbit 4 becomes invalid to its own nodes. The proposal's author says Bitcoin Core is about to be insecure. Michael Saylor says the backers should stand down. Adam Back calls it a stupid idea that fails technical consensus. They are looking at the same chain. A 2.6 percent signaling rate cannot mean three different things. It means one thing. All three men know it.
Charts lie. Liquidity speaks. On-chain, the signal is a flatline.
What Is BIP-110, Really?
BIP-110 is a temporary soft fork that caps the amount of data a Bitcoin transaction can carry. It ships in Bitcoin Knots, the smaller and more opinionated client maintained by Luke Dashjr. The rule is enforced through a versionbit — one flag in the block header. Miners must set versionbit 4. Blocks that do not carry the flag are invalid to BIP-110 nodes.
Dathon Ohm, the pseudonymous author, says the proposal fixes critical vulnerabilities. He also says miners and users should stop running Bitcoin Core because Core will become insecure once mandatory signaling begins. His warning is specific: miners who take block templates from Core may produce invalid blocks on an incoherent chain that keeps being wiped out, along with any earnings.
The problem is that the proposal's own website does not use that language. Bip110.org describes the change as a curb on arbitrary data. That is a policy preference, not an emergency. A critical vulnerability is one thing. A block-space opinion is another. The distance between those two descriptions is the whole BIP-110 story.
The Blockspace Argument, and the Ghost of SegWit
Bitcoin block space is scarce. Every transaction is bidding for the same bytes. The “arbitrary data” debate is as old as the whitepaper, but the framing has changed. After SegWit, the next frontier was cleaning up the remaining ways to store non-financial data in a block. BIP-110 is that urge made into code.
Opponents reply that the protocol should not care what data moves through it. Bitcoin is not a content moderator. That is a legitimate argument. It is not a critical vulnerability. The words “critical” and “vulnerability” imply a code-level exploit. They imply someone can steal funds, break signatures, or corrupt the chain. No one has shown that. The author says it. The author's own documentation does not.
The same gap applies to the “insecure Core” claim. A client is not insecure because another client disagrees with its consensus rules. Insecurity is something else. It is a hole that lets an attacker steal keys or forge signatures. Versionbit 4 is a header flag. A block without it is not a broken block. It is a block that a particular node population chooses to reject. That is not a security advisory. That is a political boundary.
The Signal Math Is Brutal
At block 961,022, Saylor counted 38 signaling blocks, a rate of 2.70 percent. At block 961,421, the monitor the project cites logged 47 of 1,806 blocks — 2.60 percent. Eight new signals in 392 blocks. The period rate slipped rather than climbed.
Early lock-in needs 1,109 signaling blocks. With 217 left in the current period, the highest reachable total is 263. No completed two-week stretch since December has finished above 1.29 percent.
Let me translate into trader language. This is not a rally losing steam. It is a bid with no size. If I looked at an order book where 47 contracts were bid out of 1,806 traded, and the bid was fading, I would assume the buy side had left. There is no hidden accumulation here. The sample is every block. It is not an opinion. It is a census.
The versionbit is also cheap to set. A miner can flip versionbit 4 with zero impact on his transaction set. It costs nothing in fees. It costs nothing in hash. If 97.4 percent of miners are not setting it, there is no emergency. There is either apathy or opposition. Both kill a soft fork.
I have watched this pattern before. In the 2020 DeFi Summer, I deployed $500 into a Uniswap arbitrage bot. I thought I saw a price gap. One hour later, a slippage error had cost me 20 percent. The lesson was simple: a quote is not a trade. Someone has to take the other side. BIP-110 has the same problem. A proposal is not consensus until miners pay for it, and the price is paid in headers, not tweets.
A Versionbit Is Not a Vote
Versionbit signaling is often described as a mining vote. That is too generous. A miner can change versionbit settings at any moment. A miner can signal one day and stop the next. A single mining pool can produce many headers with the same apparent interest. This is why Bitcoin activation thresholds are not meant to be easy. They are meant to make failure cheap. The early lock-in threshold for BIP-110 is 1,109 out of 2,016 blocks. That is not a majority of stakeholders. It is a filter.
The filter exists because activation is expensive. A soft fork changes the rules for everyone who validates, even if they never asked for the change. The threshold protects the silent majority from the loud minority. BIP-110 is so far under the filter that it is not even close to the starting line. With 217 blocks left, even a perfect, unanimous move by every remaining block yields 263 total signaling blocks. That is not a rounding error. It is a mathematical impossibility. No amount of social pressure can produce 846 extra blocks out of 217 remaining.
The Economic Reality of a Minority Chain
Even if BIP-110 miners produce a competing chain, they will face a liquidity problem. Block rewards on the minority chain will be denominated in coins that major exchanges might not recognize. Miners have real costs in fiat. They cannot pay an electrical bill with an orphaned header. This is not a subtle point. It is why minority chains almost never survive. In every contested fork I have analyzed, the coin with exchange liquidity wins. The coin with purity dies.
What Happens at Block 961,632?
At the cutoff, mandatory signaling begins. BIP-110 nodes reject every block without versionbit 4. The chain that currently carries the network, the chain with nearly all the hashpower, becomes invisible to them. The old chain continues producing blocks. The new nodes call them invalid. That is a split.
But a split is not an equilibrium. It is a transition to a single survivor. Dashjr says any miner who refuses to signal loses block rewards. That is true only inside the BIP-110 consensus set. On the majority chain, a non-signaling miner keeps the rewards that the majority chain recognizes. The two sides see each other's coins as invalid. The question is which side can produce blocks.
The BIP-110 side, if it produces any blocks at all, has about 2.6 percent of the hash rate. It will take hours to find a block, then days to adjust difficulty. Exchanges need quick confirmations. Custodians need finality. Institutions need clarity. None of those exist on a chain that produces two blocks a day.
This is not Bitcoin Cash. This is not a hard fork with two currencies from the same genesis. This is a flag day enforced by a minority node policy. Bitcoin's proof of work does not care about policy. It cares about work. The chain with the most accumulated work is the one that survives. A minority chain is designed to be wiped out. The author's phrase “incoherent chain that keeps being wiped out” is not a warning about a Core bug. It is a description of Bitcoin's intended behavior.
The “Insecure Core” Claim Is Security Theater
Security theater means making people feel unsafe without making them less safe. BIP-110's campaign has invented a new version: making people feel unsafe without making them less safe. The idea that Core becomes insecure after 961,632 presumes that the BIP-110 chain will be the real Bitcoin. If that chain never exists, Core is not insecure. If that chain exists but is promptly orphaned, Core is still not insecure.
Protocols do not argue. They reject.
Bitcoin Core will keep producing the chain that miners actually mine. It will keep accepting transactions. It will keep enforcing rules that have settled the network for years. The only thing that changes is the set of nodes that refuse to talk to Core. That refusal is not a vulnerability. It is a decrease in the size of the BIP-110 community.
The word “insecure” is doing a huge amount of work. It should not be. A critical vulnerability in Bitcoin would be reported privately, patched urgently, and only then announced. It would not be announced as a countdown on X. The mismatch between “critical vulnerabilities” and “curb on arbitrary data” is enough to tell any serious engineer that this is an opinion dressed in an emergency suit.
Why Miners Are Not Signaling
The absence of miner support is not a mystery. BIP-110 does not increase miner revenue. It does not protect against inflation. It does not improve finality. It imposes a rule that might reduce the set of transactions miners can include. From a miner's point of view, the upside is a cleaner block-space debate. The downside is a fight with the reference client and potentially invalid blocks. That is a bad risk-reward trade.
There is also a coordination problem. Even if some miners privately like the rule, signaling early does not help unless enough others do the same. A single miner who sets versionbit 4 before the deadline gains nothing. The threshold is not about good faith. It is about count. If the count is unreachable, early signaling is a donation to a monitor website. Miners are not charities.
This is why the BIP-110 campaign is asking miners to switch clients in the final hours. It cannot win by persuasion. It can only win by fear. Tell miners that Core is insecure, that blocks will be wiped out, that earnings will disappear. That is not a technical roadmap. That is a last-minute plea.
The Contrarian Read: The Split Is Not the Risk
Here is the part that most commentary will get wrong. The market is treating this as “Bitcoin could split into two assets.” Retail sees a potential free airdrop. Smart money sees something else: the cost of a governance process that lets a 2.6 percent minority force a flag-day test. The real split is not between two chains. It is between people who understand versionbits and people who ask “which bag do I hold?”
FOMO is a tax on the unobservant. Fear is its margin call.
Dashjr says there is no material opposition. Saylor says the backers should stand down. Back says the idea is the polar opposite of SegWit, which had technical and ecosystem consensus. None of these men are neutral. Dashjr maintains the client that carries BIP-110. Saylor holds Bitcoin on a corporate balance sheet. Back runs Blockstream and has spent years defending a particular route to SegWit. Their positions are predictable. The chain is the only witness with no position. The chain says 2.6 percent. The chain says early lock-in is unreachable. The chain says no completed two-week stretch since December has cleared 1.29 percent.
Be careful with the phrase “no material opposition” from a maintainer with a stake in the client. In my experience auditing protocol changes, the phrase “no material opposition” is often followed by a list of people who were never asked. The only valid measure of opposition in Bitcoin is the versionbit. The only valid measure of support is the versionbit. Tweets do not count. Posts do not count. A header is a signature. A signature is a statement.
The Institutional Cost Is Quiet but Real
Bitcoin is no longer just a node network. It is a settlement network with ETF products, corporate treasuries, and custody agreements. These counterparts are not watching versionbits. They are watching headlines. A headline that says “Bitcoin Could Split in Two” makes compliance officers nervous. It does not matter how small the fork is. It matters that the word “split” appears in the same sentence as Bitcoin.
Saylor understands this. His warning about neutrality is not a technical argument. It is a market-structure argument. He is saying: do not let the weekend's headlines redefine what Bitcoin is. In a post-ETF world, Bitcoin's gravity is institutional. The “peer-to-peer electronic cash” story is dead. What remains is settlement value. Settlement assets cannot have a “THIS IS NOT A DRILL” tweet from a pseudonymous author. They cannot have two competing claims of invalid blocks in one weekend. Institutions do not trade on the chain that wins the argument. They trade on the chain that wins the liquidity. Liquidity follows clarity. A 2.6 percent fork is not clarity. It is noise with confidence.
The Takeaway: Watch the Next 217 Blocks
The next 217 blocks are the only credible polling station. If the signal rate stays near 2.6 percent, BIP-110 is over. If it suddenly breaks to double digits, someone is buying signals off the books. Do not trade on the tweets. Do not sell because of the word “split.” Let the difficulty adjustment do its work.
For the weekend, set no stops around block 961,632 unless you enjoy being wicked out. This is a sideways market. A fork narrative will create a liquidity sweep, not a trend. Wait for the versionbit count to settle. A sub-3 percent rate is a sell signal for the fork narrative, not for Bitcoin. A break above 5 percent would be a different conversation.
The question is not whether Bitcoin splits this weekend. The question is whether Bitcoin can survive another round of people confusing a preferred rule with a security patch. The code will accept or reject the next 217 headers. The market will price the outcome. The rest is theater.
Watch the headers. They do not lie.