BIP-110 Lost. Bitcoin Worked. Here Is What I Learned.

A few weeks ago I published “BIP-110, the Spam War, and the Battle Nobody Wants to Name: Where I Stand.”

In that piece, I tried to do something very specific.

I followed the money on both sides.

I looked at the game theory rather than the tribal narratives.

I laid out the possible escalation path.

And, most importantly, I told you exactly where I stood before we knew the outcome.

I already ran Bitcoin Knots. That was an implementation choice I had made before BIP-110 became the centre of this fight. (Simon Dixon)

I supported what BIP-110 represented: node operators demonstrating that Bitcoin Core could be resisted, particularly when powerful corporate and institutional interests increasingly surround Bitcoin development and infrastructure. But I was also explicit that I had concerns about BIP-110 itself. (Simon Dixon)

I wanted the experiment to happen.

I wanted it to receive all the attention it deserved rather than being buried.

Because if there really is an attack vector around developer concentration, corporate influence, mining-pool coordination or economic-node centralisation, then I would rather Bitcoin discover how that attack vector works now, while we are paying attention, than discover it later under much worse circumstances.

That was one of the central arguments in my previous piece: if Bitcoin has a weakness here, we should run the experiment and learn exactly what that weakness is. (Simon Dixon)

But I also laid out my stop-point.

I was willing to support the experiment while we were fighting over the rules of Bitcoin.

I was not willing to follow a minority branch into becoming another coin.

That distinction mattered before the result.

It matters even more now.

Because BIP-110 lost.

And I am sticking to what I said.

Bitcoin Worked Exactly as Designed

The first conclusion is the simplest:

Bitcoin worked.

BIP-110 supporters were free to propose stricter rules.

Node operators were free to enforce them.

Miners were free not to follow them.

Other nodes were free to continue enforcing Bitcoin’s existing consensus rules.

The market is free to decide which chain it regarded as Bitcoin.

Nobody needed permission to run the experiment.

But nobody could force everybody else to participate.

That is Bitcoin.

The outcome was decisive.

The overwhelming majority of Bitcoin’s hashpower remained on the existing Bitcoin chain.

The BIP-110 branch received only a tiny fraction of hashpower, produced only a small number of blocks and rapidly fell far behind the Bitcoin chain.

That matters because of one very simple technical reality:

Nodes don’t produce blocks. Miners do.

Without sufficient hashpower, blocks become extremely slow and expensive to produce.

If people stop mining the BIP-110 branch completely, that chain simply stops advancing.

If people continue mining it and a market develops around it, it could eventually become a shitcoin .

And that is precisely where my previously stated stop-point applies.

I supported testing whether BIP-110 could change Bitcoin. I did not commit to creating another Bitcoin.

The Experiment Answered a Question

Before this happened, one of the most interesting facts was the enormous growth in Bitcoin Knots.

Knots went from being a relatively small alternative implementation to accounting for a substantial minority of publicly visible nodes. In my previous article, I described that as evidence of real conviction among node operators. (Simon Dixon)

I still believe that was significant.

But BIP-110 has now taught us how to interpret that number more carefully.

Node count is not a referendum.

Twenty percent of nodes does not automatically mean:

• 20% of Bitcoin holders;
• 20% of Bitcoin’s economic value;
• 20% of transactions;
• 20% of exchanges and businesses;
• or 20% of Bitcoin’s hashpower.

So you should not describe the Knots number as though it represented 20% support for BIP-110 across Bitcoin.

The more accurate conclusion is:

A substantial minority of node operators were sufficiently concerned about Bitcoin’s direction to change the software they ran.

That matters.

But BIP-110 then tested whether that constituency could translate its resistance into the power to change Bitcoin’s consensus rules.

It couldn’t.

That’s not something to explain away.

It’s something to learn from.

A Node Gives You Sovereignty, Not Votes

Interestingly, I had already made part of this argument in the previous article.

A node does not “vote.”

Bitcoin has no one-node-one-vote governance system.

A node simply validates according to the rules its operator has chosen and rejects anything that violates those rules. (Simon Dixon)

That distinction has now become much clearer to me.

Running your own node remains extraordinarily important.

Your node determines whether the Bitcoin you receive complies with your rules.

Nobody can make your node accept something it believes is invalid.

That is sovereignty.

But sovereignty is not the same thing as authority over everybody else.

Ten thousand nodes enforcing a minority rule don’t automatically force miners or other economic participants to enforce that rule.

So I think the cleanest way to express what we have just learned is:

A Bitcoin node gives you sovereignty over the Bitcoin you accept. It does not give you authority over the Bitcoin everyone else accepts.

Defensive Power Is Different From Offensive Power

This may be the biggest lesson of BIP-110.

Nodes are extremely powerful defensively.

Your node can say:

“I refuse to accept something that violates the Bitcoin rules I already recognise.”

But BIP-110 was attempting something harder.

It was saying:

“I am adding an additional restriction, and I want the rest of Bitcoin to converge on that stricter set of rules.”

Technically, BIP-110 was a soft fork.

But economically and politically, it was still an attempt to change Bitcoin’s consensus.

A minority is completely free to enforce stricter rules upon itself.

What it cannot automatically do is force everybody else to adopt those stricter rules.

I actually laid out this precise possibility before the fork point.

If only a minority of hashpower enforced BIP-110, I wrote that the enforcing nodes would find themselves on a minority chain while the looser chain continued. (Simon Dixon)

That is essentially what the experiment has now demonstrated.

Did Mining Pools Decide?

For this particular battle, largely yes.

The overwhelming majority of hashpower did not follow BIP-110.

That determined the immediate result.

And this is something Bitcoiners should not dismiss merely because they disliked BIP-110.

Mining-pool coordination matters.

We often speak about hashpower as though every ASIC represents an independent political actor.

It doesn’t necessarily.

There is an important distinction between:

• an ASIC owner;
• an individual miner;
• a mining farm;
• a mining pool;
• and whoever constructs the block template.

Those actors can have very different powers.

This episode therefore strengthens, rather than weakens, my interest in technologies that decentralise block construction and give individual miners more control over the templates they mine.

Mining decentralisation remains an important battlefield.

But I also think it would be a serious mistake to take the next step and conclude:

“Therefore miners control Bitcoin.”

They don’t.

2017 Proved the Opposite Side of the Equation

If miners controlled Bitcoin outright, SegWit2x should have succeeded.

It didn’t.

I lived through that fight.

The New York Agreement assembled major corporations and enormous mining support behind an attempt to change Bitcoin’s rules.

Yet the economic network refused to follow.

The corporations and miners ultimately backed down because mining a chain that the economy does not recognise as Bitcoin is economically pointless. (Simon Dixon)

That gives us an extraordinary symmetry.

2017: major corporations and miners could not successfully impose a contentious hard fork.

2026: a substantial node movement could not successfully impose a contentious restrictive soft fork without sufficient miner and economic support.

Taken together, those two experiments tell us something much more interesting than either one individually.

No single constituency has unilateral control over Bitcoin.

Soft Forks Don’t Simply Require 51%

There is another conclusion I want to be precise about.

It would be too simplistic to say:

“BIP-110 proves you need majority hashpower to soft fork Bitcoin.”

Bitcoin’s game theory is more complicated than that.

Economically significant users can potentially force miners to make difficult decisions.

Exchanges matter.

Custodians matter.

Businesses matter.

Wallets matter.

Holders matter.

Markets matter.

That was part of the reason I was willing to run this experiment in the first place.

In 2017, miner signalling changed rapidly as economic pressure approached the deadline. That historical experience was why I refused to predict the BIP-110 outcome before the game theory actually played out. (Simon Dixon)

This time, however, that economic weight did not materialise strongly enough.

So the more accurate lesson is:

A minority cannot simply impose a restrictive soft fork on miners that continue producing blocks valid under Bitcoin’s existing consensus rules.

That is much more precise than saying miners always decide.

Choosing the Right Battlefield

Another lesson comes from strategy.

You can believe your cause is righteous.

You can have intelligent supporters.

You can have clever tactics.

And still lose because you chose the wrong battlefield.

There is a hierarchy of difficulty in Bitcoin.

The hardest thing you can attempt is to impose a contentious hard fork and convince the economic network that your new chain is Bitcoin.

SegWit2x demonstrated just how difficult that is.

The second-hardest battlefield is imposing a contentious restrictive soft fork without overwhelming economic and mining coordination.

BIP-110 has now provided its own lesson there.

Below that are battles over mempool and relay policy.

Those are much easier because they don’t necessarily involve changing consensus.

And then there is something Bitcoin makes comparatively easy:

Defending the existing consensus against somebody else’s proposed change.

You simply refuse to adopt the new rules.

This asymmetry is incredibly important.

Bitcoin structurally makes changing consensus harder than defending consensus.

And that might be exactly what we want.

Bitcoin Consensus Is a Fortress

The metaphor I keep coming back to is a fortress.

Bitcoin’s existing consensus rules sit inside heavily fortified walls.

Trying to impose a hard fork means attacking the innermost citadel.

Trying to impose a contentious restrictive soft fork means attacking the walls.

Changing local relay policy is fighting on a much less fortified battlefield.

And refusing somebody else’s proposed consensus change is defensive warfare.

The crucial insight is this:

The fortress doesn’t know whether the army outside is righteous.

Bitcoin has no mechanism for determining that my proposed change is morally superior to yours.

It simply makes changing the rules extraordinarily difficult.

That gives me perhaps my favourite lesson from this entire episode:

Bitcoin’s consensus mechanism doesn’t know whether your cause is righteous. It simply makes the fortress incredibly difficult to capture.

The Mechanism That Defeated My Desired Change Is Also Protecting Bitcoin

And here is the uncomfortable part.

I wanted BIP-110 to receive attention.

I wanted Bitcoiners to run the experiment.

I wanted people to challenge the direction of Bitcoin Core rather than outsourcing their judgement to developers, corporations or institutions.

I wanted the community to understand the concentration risks I documented when I followed the money in my previous article.

BIP-110 lost.

Therefore I have to acknowledge something important:

The mechanism that prevented the change I supported is the same mechanism I rely upon to stop changes I oppose.

Bitcoin’s resistance to reform and Bitcoin’s resistance to capture are largely the same property.

You cannot celebrate ossification when somebody else wants a change and then complain about ossification when your preferred change fails.

If Bitcoin is supposed to be extremely difficult to change, intellectual consistency requires accepting that difficulty when it works against you too.

BIP-110 Losing Doesn’t Mean Bitcoin Core Was Right

This distinction is essential.

BIP-110 losing establishes:

There was insufficient consensus for BIP-110.

It does not establish:

“Therefore Bitcoin Core was right.”

Those are entirely different propositions.

Everything I wrote previously about following the money remains worth examining.

The incentive structures surrounding Bitcoin development remain worth examining.

The concentration of developer funding remains worth examining.

Institutional influence remains worth examining.

Mining-pool centralisation remains worth examining.

Implementation diversity still matters.

Self-custody still matters.

Running your own node still matters.

And Bitcoin Knots still matters.

In my previous article I argued that systems do not require a conspiracy to become concentrated. Align enough incentives and concentrated outcomes can emerge naturally. (Simon Dixon)

BIP-110 failing does not invalidate that concern.

What it does invalidate is any assumption that concern alone gives a minority the right—or the ability—to change Bitcoin’s consensus.

I Am Still Running Knots

There is another point I want to make very clearly.

The failure of BIP-110 does not mean I need to stop running Bitcoin Knots.

Knots and BIP-110 are not synonymous.

I ran Knots before this particular activation fight, and I can continue running Knots while remaining completely on Bitcoin. (Simon Dixon)

There is also a huge distinction between consensus rules and mempool/relay policy.

My node can still choose what it relays.

It can still refuse to assist certain types of traffic.

It can still express my individual policy preferences.

That’s node sovereignty.

And perhaps one of the post-BIP-110 questions we should now explore is:

How much resistance can node operators exercise at the policy layer without attempting to force a consensus change?

That may be a far more effective battlefield.

So Who Actually Controls Bitcoin?

I started this experiment asking who really controls Bitcoin.

I think the answer is clearer now.

It isn’t Bitcoin Core developers.

It isn’t node operators.

It isn’t miners.

It isn’t mining pools.

It isn’t BlackRock.

It isn’t exchanges.

It isn’t corporations.

Different actors possess different kinds of power.

Developers propose software.

Node operators choose software and validate.

Miners produce blocks.

Pools coordinate hashpower and block construction.

Exchanges and businesses create economic infrastructure.

Holders determine where their economic weight sits.

Markets ultimately price competing assets.

Each group has power.

Each group also encounters constraints imposed by all the others.

Maybe Bitcoin decentralisation doesn’t mean nobody has power.

Maybe it means:

Nobody has enough power to redefine Bitcoin alone.

The Lesson From 2017 and 2026

This is now the historical symmetry I find most interesting.

Bitcoin Cash and SegWit2x demonstrated that having corporations and miners behind you doesn’t automatically make your fork Bitcoin.

BIP-110 demonstrated that having a passionate movement of node operators behind you doesn’t automatically make your restrictive fork Bitcoin either.

And that leaves us with a much more sophisticated model:

Your node determines what Bitcoin is for you.

The emergent coordination between nodes, miners, holders, businesses and markets determines what the world converges upon as Bitcoin.

That is not neat governance.

It isn’t democracy.

It isn’t corporate governance.

It isn’t miner governance.

It is Bitcoin governance.

And deliberately, nobody is completely in charge.

My Position Hasn’t Changed. The Experiment Has Finished.

So let me finish where I started in the previous article.

I supported Bitcoin Knots.

I supported BIP-110 as an experiment.

I supported the ideology behind demonstrating that Bitcoin users can resist concentrated power.

I believed the attempt deserved attention rather than being quietly dismissed or buried.

I wanted to see exactly what happened when we ran the game theory all the way to the edge.

And before we knew how that experiment would resolve, I laid out what would make me stop.

Now we know.

BIP-110 lost.

I accept the result.

I am not following it into another shitcoin.

That isn’t abandoning my previous position.

It’s following it.

I can continue running Knots.

I can continue holding my own keys.

I can continue resisting concentration.

I can continue following the money.

I can continue scrutinising corporate influence, developer funding, mining pools and economic nodes.

And I can continue believing Bitcoiners should never outsource their judgement about Bitcoin to whichever institutions currently have the biggest balance sheets.

But I also have to respect Bitcoin’s consensus mechanism when it rejects a change I supported.

Because the exact same resistance that frustrated me this time may be the resistance that protects Bitcoin next time.

That may ultimately be the most valuable thing BIP-110 taught us.

Bitcoin isn’t decentralised because everybody gets their way.

Bitcoin is decentralised because getting everybody else to accept your way is extraordinarily difficult.

 

Read other Related Blog