Signal or split: what the BIP-110 fight reveals about how bitcoin changes its rules

BIP-110, formally titled the Reduced Data Temporary Soft Fork, is a proposal to limit how much arbitrary, non-monetary data a bitcoin transaction can carry.

Published 13 July 2026 - Bitcoin Poland Conference editorial team

In July 2026, two of the most prominent figures in bitcoin came out against a proposal to change the network's rules. Michael Saylor and Adam Back both rejected BIP-110, a plan to restrict non-monetary data on bitcoin at the consensus level. The dispute is framed as an argument about spam. It is better understood as a live test of a harder question: in a network with no chief executive and no board, who actually gets to change the rules, and how.

TL;DR

What is happening: BIP-110, a proposal to limit arbitrary data on bitcoin, is heading towards a contested activation window in August 2026.
The opposition: Michael Saylor and Adam Back publicly opposed it on 11 July 2026, arguing the cure sets a more dangerous precedent than the problem.
The proposal: It would restore the 83-byte OP_RETURN limit and cap other data fields as a temporary, one-year consensus rule, while leaving existing coins untouched.
The support: Backers, associated with Bitcoin Knots and the Ocean mining pool, argue that unlimited data storage burdens node operators and dilutes bitcoin's monetary purpose.
The state of play: Miner signalling has stayed near zero, and no major mining pool backs the proposal.
Why it matters: The episode is a case study in how bitcoin coordinates, or refuses, a change that a determined minority wants and a large part of the network does not.

What is BIP-110, and what would it change?

BIP-110, formally titled the Reduced Data Temporary Soft Fork, is a proposal to limit how much arbitrary, non-monetary data a bitcoin transaction can carry. It would restore the historical 83-byte limit on OP_RETURN outputs, cap new transaction outputs at 34 bytes and restrict several other data-carrying methods, including large Taproot control blocks. The stated aim is to discourage the use of bitcoin block space for images, text and token metadata, the activity associated with Ordinals, inscriptions and similar schemes since 2023.

Two design choices matter for readers worried about their own coins. First, the rules are temporary: they would expire automatically after roughly one year unless a later change extended them. Second, they are not retroactive. Coins held in outputs created before activation are grandfathered and remain spendable, so no existing bitcoin would be frozen.

The proposal has an unusual provenance. It was published under the pseudonym "Dathon Ohm", with long-standing developer Luke Dashjr credited for advice, and its reference implementation is Bitcoin Knots rather than Bitcoin Core. Bitcoin Core, the dominant node software, has not endorsed it. There have also been unproven allegations about undisclosed backing, which the author has denied.

The immediate trigger was a policy change the other way. In October 2025, Bitcoin Core version 30 relaxed its default data-carrier policy, raising the relayed OP_RETURN size from 83 bytes to 100,000 bytes. BIP-110 is the consensus-level response to that relaxation: where Core changed what nodes relay by default, BIP-110 would change what the network treats as valid at all.

Why do Michael Saylor and Adam Back oppose BIP-110?

The opposition does not rest on a defence of spam. Both men have criticised non-monetary data use in the past. Their argument is about method and precedent.

Michael Saylor, executive chairman of Strategy, framed the fix as the larger threat. In a post on 11 July 2026 he wrote that there are "110 things more dangerous to Bitcoin than spam", a play on the proposal's number, and argued that turning a spam dispute into a consensus change would invalidate transactions that are valid and paying fees today. The danger, in his account, is the precedent itself. Days earlier he had argued that the free market has historically resolved bitcoin's block-space questions without intervention.

Adam Back, co-founder of Blockstream and the inventor of the hashcash system cited in the bitcoin white paper, made a related case. He described the proposal as an attempt to "police other people" and argued that it conflicts with bitcoin's ethos as permissionless, censorship-resistant money. His point about mechanism was blunt: a participant can change their own software but cannot impose rules on everyone else, and forcing the change without broad agreement would produce a separate, minority chain rather than an upgrade.

Other established voices have taken similar positions. Jameson Lopp, who has elsewhere argued for aggressive protective changes to bitcoin, called BIP-110 reckless and predicted it would have little lasting effect. Developers including Greg Maxwell and Peter Todd have raised technical objections; Todd went so far as to embed the proposal's own text inside a transaction that complied with its rules, to demonstrate that determined data storage would survive it.

What do supporters of BIP-110 argue?

The case for the proposal is not trivial, and a fair reading has to state it. Supporters, associated with Bitcoin Knots users and the Ocean mining pool, begin from a real externality. Bitcoin block space is scarce and running a node has costs. Large volumes of non-monetary data, they argue, raise storage and bandwidth requirements for every node operator and compete with ordinary payments for space, which can push fees up.

From that starting point, supporters make a cultural and technical claim: that bitcoin should prioritise its use as money, and that declining to standardise data storage at the consensus level is a legitimate way to defend that focus. They stress that the measure is deliberately temporary and narrow, that it preserves every known monetary use case and that it freezes no existing funds. They also point to history. The 2017 activation of SegWit through a user-activated soft fork is cited as precedent that user and node pressure can carry a change over miner reluctance. Opponents reject that comparison, arguing SegWit had far broader backing.

How would BIP-110 activate, and where does support stand?

BIP-110 does not use bitcoin's traditional path of overwhelming miner approval. It relies on a user-activated soft fork, a mechanism in which nodes enforce a new rule whether or not miners agree, set to a 55% miner-signalling threshold rather than the customary 90 to 95%. A mandatory signalling window opens in early August 2026, near block 961,632, with activation, if it locks in, projected for around 1 September 2026.

The support data is the plainest fact in the debate. As of mid-July 2026, miner signalling for BIP-110 sits near zero in the active period and has not exceeded roughly 1% at any point since May, with no major mining pool behind it. Node adoption is in the low single digits, concentrated among operators running Bitcoin Knots. At the reduced 55% bar the proposal is designed for, backing remains far below the line.

What is the risk if it activates without broad support?

The core risk is a chain split. Because a user-activated soft fork lets nodes enforce a rule that miners and much of the economy may not follow, a scenario is possible in which some nodes enforce BIP-110 while most miners and users continue under the existing rules, leaving those nodes on a smaller chain. The result would be two competing chains rather than a single upgraded one, with the attendant confusion for exchanges, wallets and custodians about which chain to treat as bitcoin.

This is why opponents frame the precedent, not the data limits themselves, as the hazard. A contested consensus change that reaches its flag day without broad economic support does not simply fail quietly; it can fragment, however briefly, the very network it aimed to protect. Supporters counter that a proposal with almost no miner backing is far more likely to produce a small, short-lived minority chain than a lasting split, and that this outcome would show the system rejecting a change it did not want.

What does this fight reveal about how bitcoin changes its rules?

Strip away the specifics and BIP-110 is a demonstration of bitcoin's governance in motion. No developer can ship a rule change into force. No single mining pool can enact one. No prominent holder, however large, can decree it. Saylor himself described bitcoin as an equilibrium of holders weighted by coins, nodes weighted by commerce and miners weighted by hashrate, with no one group able to rewrite the rules alone. Whatever one thinks of the proposal, the current state of play illustrates the point: a change can be written, assigned a number and given an activation date, and still go nowhere without the voluntary coordination of developers, miners, node operators, businesses and users.

That is frustrating to anyone who wants bitcoin to move faster, and reassuring to anyone who values its resistance to unilateral change. Both readings are defensible, and both will be argued at length before the August window closes. The neutral observation is the durable one: bitcoin's high bar for changing itself is not an accident. It is the mechanism by which the network stays hard to capture, and BIP-110 is the latest test of whether that mechanism holds under pressure.

Two scenarios into the activation window

The proposal fizzles. Miner signalling stays near zero through the August window. BIP-110 fails to lock in, or produces at most a small minority chain that the economy ignores. The episode is remembered as evidence that a contested change without broad support cannot be forced, and the underlying block-space debate returns to policy and relay defaults rather than consensus.

A late surge tests the network. A mining pool or a bloc of node operators moves towards the threshold in the final blocks, and the flag day arrives with enforcement uncertain. Exchanges, custodians and wallet providers are forced to state which rules they will follow, replay and deposit risks come into focus, and the network briefly confronts the coordination problem that opponents warned about. Even a resolved split would leave a precedent behind.

Neither scenario is a price call, and both can be overtaken by a single pool decision. The signal to watch is not sentiment but signalling: whether any major pool crosses into meaningful support before the window closes.

Key terms

Bitcoin Improvement Proposal (BIP): A formal document proposing a change or standard for bitcoin. A BIP is a suggestion, not a decision; it takes effect only if the network adopts it.
Soft fork: A backward-compatible rule change that tightens what counts as valid. Nodes that do not upgrade still accept new blocks, which lowers but does not remove split risk.
User-activated soft fork (UASF): An activation path in which node operators enforce a new rule regardless of whether miners signal support. It shifts weight towards nodes and users.
Consensus rule versus relay policy: Consensus rules define what the whole network treats as valid. Relay policy is a local default for which transactions a node forwards. A transaction can breach policy yet remain valid.
OP_RETURN: A transaction output type that marks a small data field as unspendable, used to attach arbitrary data to a transaction.
Ordinals and inscriptions: Methods for embedding images, text and other data into bitcoin transactions, popularised from 2023 and central to the block-space dispute.
Chain split: A situation in which incompatible rules leave two competing chains, each claiming to be bitcoin.
UTXO: An unspent transaction output, the discrete unit of bitcoin a wallet spends. Under BIP-110, UTXOs created before activation would be exempt from the new limits.

Frequently asked questions

Will BIP-110 activate in 2026? As of mid-July 2026, miner signalling is near zero and no major pool supports it, which makes activation through miners unlikely on current figures. The user-activated design means the picture can still change before the August window, which is why signalling from a large pool is the number to watch.

Does BIP-110 freeze anyone's bitcoin? No. The proposal grandfathers all coins held in outputs created before activation, and there is no deadline to move existing funds.

Is this the same as the 2017 SegWit fork? Both use a user-activated approach. Supporters cite SegWit as precedent that node and user pressure can work; opponents argue SegWit had far broader agreement than BIP-110 has shown.

Who decides whether bitcoin changes its rules? No single party. Developers propose, miners signal, node operators enforce and the wider economy chooses which chain to value. A lasting change requires those groups to coordinate voluntarily around the same rules.

Bitcoin's capacity to resist a change that a committed minority wants, and to do so without any central authority making the call, is one of the properties that defines it as a monetary network. How that process works, and where its limits lie, is among the questions examined at Bitcoin Poland Conference.

About Bitcoin Poland Conference

Bitcoin Poland Conference is a bitcoin-only conference taking place on 5-6 October 2026 at the Poznań Congress Center. The programme covers bitcoin as a monetary network and the technology, security and governance questions that surround it.

Follow Bitcoin Poland Conference