Whoever writes the standard

The most valuable object in a mature infrastructure market is rarely a product at all. It is a definition. Consider the intermodal shipping container: a steel box of agreed dimensions and agreed corner fittings, dull enough that most people have never thought about it once. The company that builds the best box does not win that market. The specification of the box wins — because the specification is what the crane, the ship, the chassis, the port, and the customs form are all built against. Once the world had settled on the corner casting, a rival box that was ten percent stronger or cheaper but a centimeter off was not an improvement. It was garbage, because it fit nothing. The lesson generalizes, and it is the most important thing an investor in an infrastructure category can understand: in these markets, the deepest moat is not the product. It is the standard.

This runs against the instinct that competitive advantage comes from building something better. In consumer and application markets, it often does. But infrastructure is different, because infrastructure's whole purpose is to be depended upon by things it did not build and cannot control. A better mousetrap wins on its own merits. A better rail gauge wins nothing if the track is already laid, because the value of a gauge is not in its engineering but in the thousands of miles of track, the rolling stock, the yards and the workshops that have already committed to it. Whoever set the reference that others built against holds a position that competitors cannot easily attack — not because the reference is technically superior, but because attacking it means asking every adopter to move at once.

Why the standard is the deepest moat

The mechanism is worth stating plainly, because it is what makes standards different in kind from ordinary products rather than merely stronger versions of them. Two forces compound. The first is switching cost: every party that builds against a definition sinks work into it — training, tooling, contracts, downstream systems that assume its shape. To abandon the standard, they must abandon that work, and they must do it in coordination with everyone else who assumed the same shape, because a standard adopted alone is no standard at all. The second is the network effect: each new adopter makes the standard more valuable to every existing one, because the reason to speak a common language is the number of people who already speak it. These two forces do not add; they multiply. A definition with many adopters is both more expensive to leave and more valuable to keep with every adopter it gains.

The pattern recurs wherever coordination matters more than any single participant's preference. Accounting standards let an investor in one country read the books of a firm in another and trust that "revenue" means the same thing in both — and the standard, not any individual auditor, is what makes the statement legible across the whole market. A credit score compresses a person's history into a number that a lender who has never met them can act on; the score's power comes entirely from the fact that everyone consults the same one. File formats, character encodings, the humble protocols underneath the internet — in every case the durable position went to whoever established the reference, and in every case that position proved extraordinarily hard to displace even by rivals with better technology. The standard outlived the products, the companies, and often the technical assumptions it was born with. That durability is the asset.

Why this governs decision accountability

Decision accountability is exactly such a category, and for a structural reason. The value of a decision record does not live inside the record. It lives in whether the parties who matter — a regulator, a court, an insurer, a counterparty, an auditor — will recognize it and act on it. A record that one company finds convincing and everyone else regards as a private artifact is worth almost nothing, because the entire point of accounting for a decision is to make it answerable to someone outside the organization that made it. And recognition across parties requires exactly what a standard provides: a shared definition of what the record must contain, what it means to be able to replay a decision, what counts as adequate provenance, what threshold of contestability earns a record the standing to be relied upon.

Whoever establishes that definition does something more consequential than shipping a feature. They shape what "accountable" means, and therefore what everyone in the market must produce. The public substance is already visible in the work — the Decision Receipt as an object, admissibility as the bar it must clear, provenance and replay as the properties that let a skeptic check it rather than take it on trust. But an object becomes infrastructure only when its shape is the shape others build against: when a court knows what to ask for, an insurer knows what to price, a vendor knows what to emit, and a buyer knows what to demand. The strategic prize is not to sell the best record. It is to be the reference that determines what an admissible record is. This is the argument the book Admissible Reality develops at length, and it is a claim about market structure, not about any standard being won.

Products compete; standards accrue — and the company that defines what a decision record must contain will still matter long after today's models are forgotten.

The open standard and its difficulty

There is a fork here that is both strategic and ethical, and the two considerations point the same way. A standard can be captured — held as a proprietary format, a walled garden that others may use only on the owner's terms. Or it can be open and independently verifiable, defined precisely enough that anyone can check a record against it without asking the definer's permission. The captured route looks like the stronger business, because it appears to convert the standard directly into rent. It is the weaker one. A standard's power comes from adoption, and the parties whose adoption matters most here — regulators, courts, insurers — are precisely the ones who will not build their processes on a definition they cannot inspect and do not control. A record's legitimacy depends on its being checkable by an adversary; a standard that is itself a black box forecloses the very thing that made the record worth trusting. The openness is not a concession that weakens the moat. It is the condition that lets the moat exist at all.

This produces a genuine tension, and honesty requires naming it. The commercial value of a standard and the openness that gives it legitimacy pull against each other at the margin. Too proprietary and it fails to be adopted; too open and it is unclear what, exactly, is retained. The resolution is not to own the definition but to be its most credible author and steward — the party others reference because the reference is good, maintained, and trusted, the way a widely-used open standard still confers enormous advantage on those who shaped it and understand it best. The prize is being the reference, not owning a toll booth on it.

None of this is easy, and it would be dishonest to suggest otherwise. A standard requires adoption you cannot mandate — you can define the box, but you cannot force the cranes and the ports to fit it, and no amount of technical merit substitutes for the coordinated commitment of people who each move only because the others are moving. Being early is necessary but nowhere near sufficient; the history of standards is littered with first movers whose definitions were displaced by better-timed or better-governed ones. Premature standards, set before a market knows what it needs, fail. Captured standards, set for the definer's benefit rather than the ecosystem's, fail. The task is to define the right thing, openly enough to be trusted and precisely enough to be built against, early enough to matter and late enough to be right — and then to earn the adoption one party at a time, which cannot be bought and cannot be rushed. That is the difficulty. It is also, precisely because it is difficult, where the durable advantage lives.

— Dispatches · Summit Cognitive