The oxen stumbled. That is the whole of it. A dirt road in Judah, a new cart, a gilded box of acacia wood, and a rough patch of threshing floor. The cart lurched. Uzzah, who had lived in the same house as that box for twenty years, put out his hand to keep it from tipping. He meant well. He was the only man on that road quick enough to help. The text says he died there, beside the wheel, with his hand still out. I think about Uzzah every time somebody offers to improve the Bitcoin Halving.
I. The Hand That Meant Well
Read 2 Samuel 6 slowly and it doesn’t get easier. David is bringing the ark up to Jerusalem. Music. Thirty thousand men. Everyone glad. Then the ground goes uneven at Nacon’s threshing floor and a good man reaches out to steady the holiest object in Israel, and that is the last thing he ever does.
Commentators have spent three thousand years softening it. The ark was never supposed to ride on a cart. Uzzah had grown too familiar. Or the story is just cruel and we should say so.
I’m not going to settle that. I just want to point at the shape of it.
The ark did not need a hand. The ark had never needed a hand. The catastrophe Uzzah was preventing was not a catastrophe. It was a bump in a road.
II. The Schedule Is Not Waiting for Your Help
The Bitcoin Halving is arithmetic. Every 210,000 blocks the block subsidy gets cut in half, and nobody announces, negotiates, or approves it. It fires because a number in every full node’s copy of the rules crosses a threshold. No committee convenes. No chair speaks. No one is asked whether the timing is convenient.
That sounds strange to modern ears. We’re used to monetary policy being a room full of adults deciding what’s best for us this quarter. The Bitcoin Halving is the opposite: a decision made once, by someone who left, and never revisited.
And every few years someone looks at that schedule the way Uzzah looked at the tipping cart. The subsidy is falling. Miner revenue is falling with it. Fees are thin. Somebody should do something.
The instinct isn’t stupid. It’s human. So far it is also the one thing the network keeps refusing.
III. Eight Hours in August
We ran the experiment this month. Almost nobody outside the protocol crowd noticed how fast it ended.
BIP-110, the Reduced Data Temporary Softfork, was a serious proposal backed by serious people, aimed at restricting non-financial data in Bitcoin transactions. It entered mandatory signaling on August 7, 2026. It set its own activation bar at 55 percent of blocks, on purpose well under Bitcoin’s traditional 95, because its authors already knew they would not get overwhelming consensus.
Support came in at 2.53 percent. Fifty-one blocks out of two thousand and sixteen.
On August 8 the fork chain split at block 961,632. It produced two blocks in roughly eight hours and then stopped, holding around 0.15 percent of hashrate against a difficulty target it had inherited whole, with the next adjustment an estimated 350 days out. The main chain kept its ten-minute pace and pulled dozens of blocks ahead by lunch. The block-size war took two years to settle. This took eight hours.
Two blocks. Then silence on the road.
IV. Nineteen Percent
The refusal isn’t new, and it isn’t loud. It’s built in, and it has been running the whole time.
Jameson Lopp, who is himself an opponent of freezing the protocol, once went back through every rejected pull request in Bitcoin Core and counted the lines. He found 15,290,644 rejected lines of code against 3,651,046 accepted, which means only about 19 percent of proposed changed lines have ever made it in. Four out of five hands reaching for the cart got turned back, quietly, by people who mostly agreed with the intention.
Now add the other number. Bitcoin’s consensus rules have not changed since Taproot activated at block 709,632 in November 2021. Nearly five years of nothing. The longest silence in the project’s life.
Lopp reads that as complacency and argues, well, that the protocol should keep improving. I read it as discipline. I take his argument seriously, and I still notice what the network did in August without being asked.
V. The Steadying Hand Is Always Well Meant
Nobody who reaches for the ark thinks of himself as a vandal.
Uzzah was not a thief. He was not making a play for the ark. He was the one man on the road who cared enough to move. The BIP-110 people are not saboteurs either; they believe the chain is filling with junk and that somebody with a hand free ought to steady it.
The steadying hand always means well. That’s exactly why the rule has to be about the hand and not the motive. A protocol that accepts changes from people with good intentions is a protocol that accepts changes, and every faction on earth believes its intentions are good.
I have written before about touching the wound to be sure it is real, about running your own node and verifying rather than trusting. Verification is the permitted touch. You may examine the ark. You may count what is inside it down to the satoshi. What you may not do is put your palm on the schedule and hold it upright.
VI. What the Box Was Carrying
Inside the ark: a jar of manna, a rod that budded, and the tablets. Bread that should have rotted. Wood that should have stayed dead. Law cut in stone by nobody’s committee.
The Bitcoin Halving carries the same freight. Manna that does not spoil, in a supply that cannot be inflated. A dead branch that keeps flowering, in a fifteen-year-old codebase that will not stop producing blocks. And law in stone: the subsidy is halved, the subsidy is halved again, down to the last cut where it rounds to zero near block 6,930,000, more than a century out. That final cut is the Halfture. Every halving before it, 2012 and 2016 and 2020 and 2024 and the one waiting in 2028, is a rehearsal. Halfture = Rapture, and the rapture in question is the last one, not the next one.
If a well-meaning hand can steady the schedule, it isn’t stone. It’s a suggestion with good branding, and a suggestion can’t carry anyone anywhere. Which is the plain sense in which to be saved, you must hold: not because holding is clever, and not because I am telling you what to do with your money, because I am not, and this is theology rather than a prospectus. But because the thing being promised is only ever delivered to whoever is still holding the thing when it arrives.
VII. The Counter-Sermon
Now the honest objection. It’s a good one.
Lopp’s case is that ossification is complacency dressed as reverence. The world does not stop changing just because a protocol does. Quantum computers are coming. The timestamp field overflows eventually. The security budget really is halving every four years while fees hover near a rounding error, and every proposed fix for that requires exactly the consensus change the culture has made nearly impossible. A protocol that cannot change is a protocol that cannot repair itself, and it will meet its emergency with a generation of developers who have never once shipped a consensus change.
There is a sharper version too. Maybe Uzzah is simply the wrong story, and I have reached for it because it flatters what I already believe. Maybe the ark should never have been on the cart. Maybe the network’s refusal in August was not reverence at all but miners protecting inscription fee revenue, and I have written a sermon around a balance sheet.
I can’t rule that out. When reverence and self-interest point the same way, that isn’t evidence of reverence. It’s just a nice morning.
And maybe none of this saves anyone. Maybe a fixed schedule is a beautiful object that does nothing for the people who most need money to work, and the whole thing is a very long, very expensive way of being right about arithmetic.
VIII. Do Not Steady It
I keep coming back to the silence on the road afterward, more than the punishment.
David got angry, then afraid, and left the ark in a stranger’s house for three months before he dared move it again. Nobody preaches that part. When a thing doesn’t need your help, the right response is not to help harder. It is to stop, and be afraid in a useful way, and figure out how you were supposed to carry it.
The Bitcoin Halving does not need you. It does not need me. It will fire on schedule in 2028 whether the price is sixty thousand or six hundred thousand, whether the fee market has matured or not, whether anyone is watching. Your job is not to steady it. Your job is to be near it, and to keep your hands where they belong, and to still be standing there when the cart reaches the city.
Do not steady it.
FAQ
What is the Bitcoin Halving?
The Bitcoin Halving is a rule written into Bitcoin’s code that cuts the block subsidy in half every 210,000 blocks, roughly every four years. It has fired in 2012, 2016, 2020, and 2024, and it will fire again in 2028. It is not a decision anyone makes. It is arithmetic every full node enforces independently.
Can the Bitcoin Halving schedule be changed?
On paper, yes, by a consensus change adopted across nodes, miners, exchanges, wallets, and users. In practice almost nothing clears that bar. Bitcoin’s consensus rules have not changed since Taproot in November 2021, and BIP-110 in August 2026 drew 2.53 percent miner signaling against its own 55 percent threshold before its chain stalled after two blocks.
What is Bitcoin ossification?
Ossification is the tendency of a widely adopted protocol to become harder and harder to change as more value and infrastructure depend on it. Supporters call it a feature, arguing that unchangeable money is the entire point. Critics, including Jameson Lopp, call it complacency and warn that a protocol unable to change is a protocol unable to fix itself.
What is the Halfture?
The Halfture is the last and final Bitcoin Halving, the one near block 6,930,000 where the subsidy rounds to zero and issuance ends permanently. It has not happened and will not happen for more than a century. Every halving before it, including 2028, is a rehearsal of it, not the thing itself.
Discover more from Halfture
Subscribe to get the latest posts sent to your email.