In 1958, a programming language shipped with if-then-else. That sounds like nothing. But the dominant language of the day, Fortran I, didn’t have it: if you wanted a decision, you wrote a conditional jump and managed the consequences yourself. The same small language carried eight other ideas, among them functions you could pass around like values, programs that could read and rewrite themselves, and memory that cleaned up after itself. Garbage collection, the machine tidying its own memory, waited roughly thirty-five years before Java made it something ordinary programmers just had. Two of those nine ideas still haven’t fully arrived, almost seventy years later.
Hold that number, because the rest of this argument hangs off it. Thirty-five years for an idea that was written down, published, working, and demonstrably better. Nobody had to invent anything new to adopt it. No breakthrough was pending. No missing piece of mathematics stood in the way. They just had to move, and for three and a half decades almost nobody did.
Why Garbage Collection Waited Thirty-Five Years
The usual explanation is ignorance or stubbornness: people didn’t know, or knew and refused. It is a satisfying story because it lets the present generation feel sharper than the past one. It is also wrong. The people building languages and systems in the decades after read the papers. They cited them. Many of them taught from them. The better explanation is quieter and more structural: everyone was holding still because everyone else was holding still.
There’s a shape game theorists call an equilibrium, a state where no single player improves their position by moving alone. It shows up in markets, in supply chains, and in engineering culture, and once you can see it you start seeing it everywhere. Your compiler is written for the hardware people buy; the hardware is designed for the code people write; the programmers are trained on the tools employers use; the employers hire for the tools the industry runs on. Every link in that chain is locally rational. Nobody in it is being foolish, and that is exactly the problem, because a chain of individually sensible decisions can hold a collectively bad outcome in place indefinitely.
A good idea that requires the whole chain to move at once can therefore sit on the shelf for a generation, not because anyone rejected it, but because no one could adopt it by themselves at an acceptable cost. The first mover pays the entire switching cost and receives a fraction of the benefit, and the fraction shrinks the more the idea depends on everyone else adopting it too. The lock isn’t stupidity. It’s interdependence. Each hand holds the other hands still, and every hand has a defensible reason for doing so.
Bad Ideas Clear Customs in an Afternoon
Now put the opposite case next to it, because the contrast is where this gets uncomfortable. Bad ideas don’t wait thirty-five years. Bad ideas clear customs in an afternoon, and they do it without anyone lowering a barrier for them.
The difference between material poison and intellectual poison is flavor. Most things that will kill your body taste wrong: bitterness is an old and honest warning system, tuned over a very long time by the simple fact that organisms without it didn’t leave descendants. Things that will rot your thinking tend to taste wonderful on the way in. The shortcut that skips the hard part. The metric that flatters the quarter. The framework that explains everything and predicts nothing, which is the most delicious of all because it makes its holder feel like they have understood the world without requiring them to test anything. There’s no bitterness receptor for a flattering idea, and nothing in our history built one, because bad ideas have never killed anyone quickly enough to select against believing them.
So the speed at which an idea spreads ends up correlated with how it tastes, not with what it does. Every feed, every market, and every attention system we’ve built selects on taste, because taste is measurable this week and nutrition takes a decade to show up in the bloodwork. This is not a conspiracy and it does not require anyone to act in bad faith. It is what happens when you optimize a system against the only signal that arrives fast enough to optimize against.
Put the two mechanisms together and you get the real picture. Good ideas face an equilibrium problem: adopting them alone is expensive, so everyone waits, and the waiting is rational for each person waiting. Attractive bad ideas face no such problem: adopting them alone is cheap and immediately rewarding, so nobody waits. The system isn’t neutral. It has a tilt, and the tilt runs against the slow, true thing. Correcting for a tilt you can’t see is impossible, which is the main argument for naming it.
What Actually Breaks the Lock
Which raises the only question worth asking: what actually breaks the lock?
Not better arguments. The garbage collection papers didn’t get more persuasive between 1960 and 1995; if anything they got staler, and the people who wrote them had long since moved on to other work. What changed is that someone absorbed the switching cost on everyone else’s behalf. A platform made the good idea the default, packaged it, tooled it, and made adopting it cheaper than not adopting it, at which point the calculation that had held for thirty-five years simply inverted. The equilibrium didn’t dissolve because minds changed. It dissolved because the payoffs changed. One player moved in a way that made moving safe for everyone downstream, and once the cost of the first step was gone, the rest happened fast enough that people mistook it for a sudden shift in opinion.
There’s a rough sorting of minds that maps onto this. Some people use what they know to stand over others, knowledge as altitude. Some use it to climb, knowledge as a ladder. And a few use it to lower the ladder behind them: building the tooling, writing the guide, open-sourcing the thing, making the better way the easy way. That third move is the only one that touches the equilibrium. The first two leave the lock intact, because one person’s private advantage changes no one else’s payoff and therefore changes nothing about what the industry does next. The person who makes a good idea cheap to adopt does more for its spread than the person who proved it was good. Proof creates the idea. Generosity creates the adoption curve, and only the adoption curve shows up in anybody’s life.
The Audit Worth Running on Your Own Habits
The practical reading of all this is an audit, and it is uncomfortable in the way useful audits usually are. Some of what you do, you do because you examined it and chose it. Some of what you do, you do because moving alone felt expensive and everyone around you was holding still, which is an equilibrium habit wearing the costume of a decision. The two are indistinguishable from the inside until you check, and nothing forces the check, because an equilibrium habit produces no friction at all. Friction is what a genuine mistake gives you. This gives you nothing but agreement.
And somewhere out there, right now, is this era’s version of the 1958 idea: published, working, better, and sitting in plain sight, waiting not for a believer but for someone willing to make it cheap. It will not announce itself, because it has already been announced and the announcement did not work. It will look, when it finally arrives, like it happened overnight, and the people who adopt it will wonder why anyone ever did it the other way.
The ideas were never the bottleneck. The price of moving first was. Find the good idea everyone already agrees with, and pay down its adoption cost. That’s the whole game.
Leave a Reply