There are two ways to prosper in a system.

One is to build something better.

The other is to alter the rules until what you already possess becomes better by decree.

The first produces wealth. The second produces influence. Civilisation advances by confusing them as little as possible.

We are usually told that a technical foundation must remain changeable because change is progress. It sounds enlightened. Flexibility is one of those words, like virtue and community, which becomes more attractive as its meaning becomes less precise.

But the real question is not whether a system can change.

The real question is where human effort goes when change is possible.

A published design rule does more than describe a system. It replaces a conversation. It allows one person to build without asking another person’s permission. It permits strangers to coordinate without becoming friends, colleagues, subjects, voters, delegates or supplicants.

That is the quiet miracle of a stable rule: it performs the work of authority without requiring an authority.

The builder reads the rule, constructs against it and tests the result. She does not need to know who chairs the committee. She does not need to cultivate the maintainers, monitor the politics, interpret the roadmap or attend a conference in a city selected by men whose principal technical achievement is possessing an expense account.

A fixed interface is impartial in the only sense that matters: it is equally indifferent to everyone.

The moment the rule becomes contestable, this changes.

The rule no longer substitutes for governance. It becomes the object of governance.

Every builder must now form an opinion about what the rule will mean next year. Every serious investment becomes partly an engineering decision and partly a wager on politics. The participant must choose between improving the product and improving her position in the process that defines what products are allowed to exist.

Naturally, those who are best at the second activity will describe it as responsibility.

Power has always had a talent for presenting itself as maintenance.

The economic consequence is simple. If changing the rule is cheaper than competing under it, rational firms will invest in changing the rule. They will call this standardisation, safety, modernisation, governance, coordination or technical stewardship. The title is decorative. The return is what matters.

A better module yields an incremental advantage. Rivals may copy it, improve it or underprice it.

A favourable rule can erase rivals altogether.

The difference is between earning a profit and acquiring a privilege.

This is why the correct institutional question is not, “Can the foundation change?” Nothing made by men is metaphysically incapable of change. The correct question is:

What does it cost a well-resourced participant to alter the foundation, compared with what it costs an unknown participant to build at the edge?

Where changing the rules is cheap and building without permission is expensive, the system will attract politicians of technology.

Where changing the rules is expensive and building is cheap, it will attract builders.

This does not require villains. It requires only incentives.

A man need not be corrupt to prefer a strategy that rewards him. Indeed, the most dangerous forms of capture are performed by people who sincerely believe that their interests and the public good have achieved an unusually fortunate identity.

A committee member rarely says, “I wish to use the standard to protect my position.”

He says, “The ecosystem is not ready.”

He says, “We need alignment.”

He says, “The proposal lacks consensus.”

He says, “We must protect users.”

The language of public duty is particularly useful when the public has no means of checking the claim.