Bitcoin Techno-Maximalism

A structural companion to Michael Saylor’s “The Four Ideologies of Bitcoin.”

Saylor sorts the Bitcoin community into four ideologies: Maximalist, Capitalist, Technologist, Fundamentalist. He argues that Bitcoin reaches its full potential through a synthesis of all four.

The Maximalist supplies conviction. The Capitalist drives adoption. The Technologist solves technical problems. The Fundamentalist guards first principles.

Saylor’s essay answered why, and who. This one is about how. When you list the ways to scale Bitcoin to global capacity, you do not find a long list of options. You find three.

One is honest about what it gives up. (TradFi) One hides what it gives up. (Monolithic State Machines) And one keeps the properties that made Bitcoin worth scaling at all.

There is no comfortable fourth option, and the choice among the three we have matters more than the market yet realizes.

What Scaling Actually Demands

First, what are we asking the system to do? The Bitcoin conversation often talks past itself here.

Scaling is not one problem. It is three, and they pull against one another.

The first is throughput. Moving enough value, often enough, for billions of people and the institutions acting on their behalf. This is the problem most people mean when they say scaling, and it is the easy one. Any central database solves it. Visa solves it. Moving the numbers quickly was never the hard part.

The second is verification. The ability of an ordinary participant to confirm, independently and cheaply, that the rules held. That no coins appeared from nothing. That no balance was spent twice. That the transaction they care about happened and cannot be undone. This is the property that makes Bitcoin money rather than a promise. Satoshi’s contribution can be read as a single idea: a way to reach global agreement on transaction history without anyone having to trust a bookkeeper. Remove cheap independent verification and you have not scaled Bitcoin. You have replaced it with a statement about Bitcoin, issued by someone you now have to trust.

The third is ordering. Deciding the sequence in which transactions are recorded. This sounds like plumbing. It is the most dangerous of the three, and it is the one the scaling debate steps around. Whoever decides the order of transactions decides which transactions succeed, which fail, and which never appear at all. We will come back to this. For now, hold the word.

A real scaling solution has to deliver throughput while keeping verification cheap and without gathering control over ordering into few hands. Scaling is hard, and is a question of architecture rather than engineering, because the obvious way to win TPS throughput quietly gives up on ordering integrity and sovereign verification.

The First Option: Traditional Financial Rails

The first way to scale Bitcoin to eight billion people is already running. Place Bitcoin inside the existing financial system. Custodians hold the coins. Banks hold the accounts. Exchanges match orders on private ledgers. Funds wrap exposure in a security. Payment processors net out millions of “Bitcoin” transactions that never touch the chain and settle the remainder now and then. The user sees a number on a screen that says they own Bitcoin, and most of the time they can redeem it.

This works, in the narrow sense that it moves volume and brings people in. The spot funds absorbed enormous demand. Custodians secure billions. For a great many holders this is already what owning Bitcoin means.

It is also what it is. An open return to trusted third parties. You do not hold keys. You do not verify. You hold a claim against an institution, redeemable at its discretion and the law’s. This is the arrangement Bitcoin’s opening sentence set out to remove.

The purists tend to hurry past the next part, and it decides the argument that follows. Traditional finance is not bare centralization. It is centralization wrapped in four centuries of compensating machinery. When a custodian fails there is a bankruptcy court. When a bank takes your deposit there is deposit insurance, a regulator, a chartering authority, capital requirements, and audited reserves. When an intermediary cheats you there is law enforcement, civil suit, discovery, subpoena power, and a judge who can pull the money back. The trust is named. It is priced. It is accountable. You know who you are trusting. You know what happens when they betray you. There is a number to call and a court that will hear you.

Carry this standard through the rest of the essay, because it is the measure that matters. The centralization of traditional finance is a known quantity with a known remedy. It is trust with a paper trail and a sheriff. Whatever its failures, and Bitcoiners can list them from memory, it does not pretend to be something it is not.

That refusal to pretend is what makes the second option worse.

The Second Option: The Monolithic State Machine

The second option dominates the conversation. It fills the venture decks and the developer mindshare and the daily argument on X. It is the world of programmable, monolithic, execution-first blockchains. Ethereum and the dense stack of rollups that settle to it. Solana and its high-throughput relatives. The long tail of chains built from the same template. When the first essay wrote “L2 and L3 innovation,” this is what the market heard, because this is where the builders and the capital and the noise are.

The offer is built for the Capitalist. You get programmability, throughput, and composability, and, the claim goes, you keep decentralization, because it is a blockchain, and blockchains are decentralized, right?

To see why this is the worst of the three, worse than the honest centralization of the first option, look at how monolithic state machines work. Observe the contrast between the manifestos and blog posts versus the future its architectural optimization path guarantees.

A monolithic chain bundles execution, ordering, and settlement into one global, totally ordered sequence. Every node, in principle, must process every transaction in the same order to arrive at the same state of the world. This single decision defines the category, and the rest follows from it. Because every participant must replay every transaction to know the state, the cost of taking part in consensus rises with the activity of the whole network. The more the chain succeeds, the heavier each block, the more expensive it becomes to run a node that genuinely verifies, and the fewer parties can afford to.

This is not a problem that faster hardware retires. It is a slope built into the architecture, and it runs one way, toward concentration. As demand for throughput climbs, block production gathers among the few operators with the capital and bandwidth and specialized machines to keep pace.

Zero-knowledge proofs are often offered as the escape, and they are a real achievement: a validity proof lets you confirm a state transition is correct without replaying it, which lowers the cost of checking the math. But producing those proofs is itself heavy, specialized work, so proving concentrates among industrial provers even as checking gets cheaper, and the proof attests only that the transactions in the block were executed correctly, never that the right transactions were included or that any were left out. Verification migrates from anyone with a laptop to anyone with a data center, and then, in practice, to anyone willing to trust the data center.

The rollup model states this openly and then softens it. Today the leading rollups run a single sequencer, one party that orders every transaction, and the ecosystem reassures itself that decentralizing the sequencer is on the roadmap. It has been on the roadmap for years. The architecture does not want to let go of it, because a single fast orderer is the efficient answer to the problem the architecture created.

Set the two options side by side. Both gather control into few hands. TradFi admits it and surrounds it with courts and regulators and recourse. Monolithic VMs deny the concentration of power while building toward it. You inherit the centralization of traditional finance with none of the compensating machinery. No deposit insurance. No regulator charged with protecting you. No bankruptcy court with reach over an anonymous sequencer or a multisig of pseudonymous signers. No law enforcement that can compel a foundation in an undisclosed jurisdiction to make you whole. The trust assumptions are not removed. They are moved to where no one can see them and no one answers for them.

This is the shell game. The marketing says trustless and decentralized. The architecture says trust the sequencer, the bridge multisig, the upgrade key, the foundation, and the few entities that can afford to produce blocks. The words and the structure point in opposite directions, and the gap between them is filled with the user’s risk, unpriced and unprotected. The bank, at least, told you it was a bank.

The Implications of Ordering

Return to the third problem, the one set aside earlier. Deciding the sequence in which transactions are recorded. In a monolithic, totally ordered state machine, someone has to choose that sequence for every transaction on the network. Whoever produces the block decides what goes in, in what order, and what stays out.

In the early and hopeful telling, this power is diffuse and almost accidental. Miners or validators drop transactions into blocks roughly as they arrive. On a high-value chain the reality is different. The right to order transactions becomes one of the most valuable franchises in the system, because the order itself is worth money. See a large trade about to execute, slip your own transaction in front of it, and you profit at that trader’s expense. Place a user between two of your own transactions and you take the spread. Reorder, insert, or delay at will, and the ledger’s sequence turns into a surface for extraction.

The industry, with some understatement, named this maximal extractable value. Most Bitcoiners have heard the term, filed it under Ethereum’s problems, and stopped listening. That reflex is the mistake this essay is about.

The dismissal goes like this. It is a few cents skimmed from some speculator’s trade, a cultural pathology of another chain, nothing to do with sound money. This badly understates what is being described. The skimming is the measurable, monetized symptom of a deeper fact. On a monolithic state machine, control over transaction ordering is a concentrated, valuable, and largely lawless activity. The sandwich trades are not the disease. They are its earliest and mildest visible sign. The capability underneath is the power to decide which version of events the ledger will record.

Follow that capability down the slope the architecture has already shown us. Ordering rights, like block production, gather into few hands. The parties that extract the most value from ordering can pay the most to control it, which lets them extract more, which lets them pay more. Firms appear whose entire business is capturing the value of order. The right to decide sequence consolidates among a small number of sophisticated, well-funded, and unaccountable operators. They are not chartered. They are not regulated. They are often pseudonymous. No law names what they do as a crime, because the law has not caught up and the jurisdiction is wherever the servers happen to sit. They are lawless in the exact sense of the word. They operate where no applicable law reaches, and those they extract from have nowhere to turn.

Now ask the question that turns a problem about fees into a problem about freedom. If a small number of unaccountable operators control the order of transactions, what else does that control allow?

It allows them to decide which transactions never happen at all.

The Road to Minitruth

Censorship on a monolithic chain does not require breaking the protocol. It is a native power of whoever orders the blocks. Control the sequence and you control inclusion. You can delay a transaction without end. You can refuse it. You can make a chosen address, a chosen kind of transaction, or a chosen person’s economic life simply fail to occur, again and again, while the chain goes on producing blocks that look perfectly valid to everyone watching. Nothing in the ledger marks the transaction that was suppressed. The censored event leaves no trace, because in a ledger that records only what the orderer admits, the suppressed transaction never happened.

We have already seen the rehearsal. When regulators sanctioned certain addresses, a majority of the parties ordering Ethereum blocks began, for a time, declining to include transactions that touched those addresses. Not because the protocol demanded it, but because the concentrated, identifiable, pressurable orderers chose to comply rather than stay neutral. The capability was demonstrated. The only open question is who points it, and at what.

This is the line of reasoning that “L2 and L3 innovation” waves past, and it is why scaling architecture is not a downstream concern but the whole of the matter for sound money. A monetary network whose ordering is controlled by a few unaccountable parties is a network with a built-in mechanism for editing economic reality. Today that mechanism edits reality for profit, in the sandwich and the skim. Pointed differently, the same mechanism edits reality for control. Who may transact. Who may not. And what the permanent record will say took place. The ledger stops being a neutral account of what happened and becomes an account of what the orderers allowed to happen, which is a different thing wearing the same clothes.

Orwell gave the institution that edits the record to suit power a name. The Ministry of Truth, whose work was to revise the past without rest, so that it always confirmed the present regime. The horror of that ministry was never the crude lie. It was that the record itself, the thing you would consult to check the lie, had become the instrument of the lie. There was no neutral ledger left to appeal to. The archive was the weapon.

A blockchain was meant to be the opposite of that ministry. The one record no one could quietly revise. The neutral account of what happened that no king or bank or office could rewrite. This is the whole reason a Bitcoiner cares about decentralization. It is not a taste for distributed systems. It is the refusal to let any party become the judge of what is true about who owns what.

A monolithic state machine, raised to global importance, rebuilds that judge. It does so by degrees, along a slope no single person chooses and everyone descends, until the power to decide what the record says rests with a handful of operators no law can reach. It arrives at the ministry not through malice but through architecture. And it arrives wearing the language of decentralization, which is what makes it more dangerous than the bank that at least asked you to sign a custody agreement.

This is why the monolithic state machine is the worst option for scaling Bitcoin. TradFi is a regression from Bitcoin but at least it gives you centralization with accountability. The monolith gives you centralization with deniability. The same concentration of power, the same trusted third parties, the same editorial hand on the record, laundered through doublespeak and shielded from every remedy that four centuries of financial law evolved to supply. It is not a scaling solution for sound money. It is a scaling solution for the thing sound money was built to resist.

If these were the only two options, the Fundamentalist’s instinct to refuse the whole business, to keep Bitcoin small and self-custodied and verified by hand, would be not just defensible but right. That’s why many people say “Bitcoin only.”

So if Bitcoin scaling doesn’t really work on TradFi or monolithic VMs, what is the solution?

The Secret Third Thing

The deepest mistake in the monolithic design is the gap between the minimum viable global coordination a network requires, to the ordering monolithic state machines impose. A monetary network has to agree on sequence: you cannot spend a coin before you receive it, and when two parties settle, the network must agree that it happened. That is the required final ordering.

The monolithic chain takes that narrow requirement and inflates it into a single scarce canonical line that every transaction must join, whether or not it has anything to do with adjacently ordered transactions or not. This forces artificial order where no causal relationships exist, actual relational information is destroyed, the scarce blockspace is expensive, and the power to determine order and censor activity concentrates. What concentrates gets captured.

Most transactions have no order to discover. If I pay you, and on the far side of the world two strangers transact, there is no meaningful fact about which came first. Forcing the network to invent one is costly fiction, and the cost is not just performance. The invented sequence has to be assembled by someone, and that someone now holds the power this entire essay is about. Order is real within a single account and at the moments accounts touch. Everywhere else it is manufactured, and every degree of manufactured order is an increment of valuable and capturable power.

The Block Lattice: Reloaded

This is the idea behind a different architectural lineage, a block-lattice paired with a meta-DAG. Each piece is routinely misread through monolithic habits, so take them one at a time. The meta-DAG is not the absence of global ordering. It is global ordering reduced to its minimum: enough shared sequence to make the network agree and prevent double-spends, and not one increment more than that. Imagine a notary that stamps your documents before anyone interprets or judges them.

The block-lattice is a relational data structure, not a trick for parallelism. Each account owns its own chain and is the only party that can add to it. Your history is yours to extend. Mine is mine. There is no shared global chain that every transaction must pass through, and so there is no single sequence for anyone to control. The structure records a partial order. It keeps the orderings that are causally real, the sequence of operations within each account, and declines to impose orderings that are causally invented, between transactions that never met. This is not a performance optimization that happens to look decentralized. It is a refusal, written into the data structure, to build the chokepoint that the second option’s total order requires.

The meta-DAG is the consensus layer that establishes agreement across these independent account-chains without collapsing them into a total order. Its task is to confirm and interlock the partial order, to make the concurrent transactions mutually verifiable and final, while keeping their concurrency rather than flattening it into one sequence owned by a sequencer. The meta-DAG is an ordering primitive that produces enough order to prevent double-spends and establish finality, and no more order than that. The “no more” is the point. Every increment of unnecessary global ordering is an increment of capturable control, and a partial-order architecture spends none of it.

The difference between blockchain and block lattice is categorical. The difference is not that one has less extraction than the other. It is that the precondition for ordering-based extraction and censorship, a single global sequence that some privileged party assembles, is absent by construction.

The MEV surface is eliminated. You cannot censor by reordering a sequence you do not assemble. You cannot consolidate ordering rights that were never gathered into a franchise to begin with. The architectural choices that created MEV extraction, that concentrated into unaccountable operators, that matured into the ministry, did not get reduced. It is absent. That is what categorical means here. Not a mitigation strategy, but the refusal to create the problem in the first place.

A partial-order architecture does not abolish every trust assumption, and it does not solve every problem in distributed systems. Nothing does, and any design that says otherwise is running the same marketing-over-architecture play as the second option. What it does is refuse the specific concentration the monolithic design treats as ordinary infrastructure, the global orderer, and in refusing it, close the specific path that runs from throughput pressure through ordering capture to censorship and the editing of the record. The trust assumptions that remain are smaller, more local, and more honestly stated.

Why There Is No Fourth Option

TradFi, EVM, and meta-DAG + block lattice. Three is a suspiciously tidy number. Surely the design space is wider than this?

I went looking, and the menu is short for structural reasons, not rhetorical ones. Strip the question to its core. How does a system reach agreement, at global scale, on the sequence and validity of transactions? There are only a few truly distinct answers, and they differ by where they place the authority to order.

You can put ordering authority in a trusted institution and make up for the resulting centralization with accountability from outside the system, through courts and regulators and law. That is the first option, and the entire traditional financial system is its proof. It is honest about being centralized.

You can put ordering authority in a single global sequence and let market forces assemble it, which produces, along a slope no participant controls, a small set of unaccountable orderers with an editorial hand on the record. That is the second option, the monolithic state machine that brags about validator count but quietly sweeps protocol-level centralization under the rug.

You can decline to manufacture stateful global ordering at all, blind ordering, record only what is causally real, and build consensus that preserves it. That is the third option, and it is the only one of the three that scales without concentrating ordering authority.

That is the finding, and it is why this essay exists. We are not choosing from a wide menu. We are choosing among three architectures, and only one scales Bitcoin to the Capitalist’s vision while also satisfying the Maximalist, Fundamentalist and Technologist. The four temperaments cannot be reconciled by negotiation. They can only be reconciled by architecture, because it is the architecture that decides whether adoption betrays preservation or lets it stand. Acta non verba.