The Dopamine Trap on Servers: Why Some Hook You and Others Don't

Two servers. Two nights at three in the morning. Two completely opposite feelings when you switch off the computer.

On the first you had spent hours opening crates, chasing the legendary item that never dropped. One more key, one more crate, "this time for sure". When you finally stopped, you didn't feel satisfaction: you felt a strange emptiness, like getting up from a slot machine. On the second you had been building. A house, a redstone system, a tidy mine. And when you switched off, even though it was also very late, you felt something completely different. The moment of recognition sums it up: when I left the sandbox at three in the morning I felt I had built something; when I left the pay-to-win I felt I had lost.

Both had kept you glued to the screen for the same number of hours. But they weren't the same. And the difference has a chemical name.

Not all dopamine is equal

We tend to talk about dopamine as if it were one single thing — "the pleasure chemical" — and as if everything that releases it were equally good or bad. It is subtler than that.

There is an enormous difference between two ways of triggering it. One is the dopamine of anticipation: the one that fires at the expectation of a reward that may or may not arrive. It isn't the prize that switches it on, it's the "let's see if this time". The other is the dopamine of achievement: the one that comes when you complete something real, when effort sets into a result that is yours. The first keeps you pulling the lever; the second leaves you full and lets you stop. One is a loop that never closes; the other, a satisfaction that does.

The crate server exploits the first. The building sandbox feeds the second. From the outside both keep you playing; on the inside, one squeezes you dry and the other nourishes you.

The engine of the lever

Where does the dopamine of anticipation, the one of the loop that never closes, come from? From a mechanism that has been known for almost seventy years.

B. F. Skinner described variable-ratio reinforcement: when a reward arrives unpredictably — sometimes yes, sometimes no — the behaviour that chases it becomes extremely intense and extremely hard to give up. It is the exact principle of slot machines. And it is also the principle of crates and a good many pay-to-win mechanics: you open the crate and maybe the killer item drops, you buy the key and maybe this time you'll be lucky. It isn't the reward that hooks you. It's the uncertainty.

A sandbox with natural progression doesn't work like that. When you build a house, there is no chance involved: you place a block and a block appears, you invest effort and get a result, predictably and cumulatively. It is the opposite of the slot machine. That is why you can stop when you want: there is no "maybe next time" pulling at you, just a creation that is already yours and will still be there tomorrow.

The question that defines the design

This is where the heart of the matter appears, and Misha Klimovsky put it very well in his essay The Architecture of Attention: if families delegate to algorithms optimised for engagement metrics, he argues, they are in practice handing formative authority to systems that do not care about developmental outcomes. Platforms do not ask what a child needs in order to become someone. They ask what keeps that child hooked.

That is the line that separates a design that squeezes you from one that respects you: the question in the head of whoever built it. A crate server is designed around "what keeps you pulling the lever?", and to answer it, it turns to Skinner's uncertainty. A building server is designed around "what leaves you satisfied?", and for that it offers real progression and achievement. Both generate dopamine, but in the service of opposite goals: one, to keep you; the other, for you to enjoy yourself and leave whole.

The principle: what counts isn't how much it hooks you, but how it leaves you

Step away from servers, because this works as a measure for almost anything you consume.

The time you spend with something doesn't tell you whether it does you good. A slot machine and a good book can hold you for the same three hours. The question that truly tells them apart isn't how much it hooked you, but how it left you when you finished: full or empty, with something, or with the feeling that something was taken from you. The dopamine of anticipation keeps you going; that of achievement completes you. And almost always, what squeezes you dry can be recognised by that emptiness afterwards, while what nourishes you leaves a residue of "that was worth it".

That is the principle: to know whether something does you good, don't look at how much it hooks you while you're doing it; look at how you feel just after letting go of it. Being hooked is a poor judge, because the "during" is precisely what can be manipulated. What is honest, what can't be faked, is the "after": the hangover or the fullness, the "again, and I don't even know why" or the "that did me good". Learn to read that moment and you'll have a compass for almost everything — screens, food, relationships, work.

What can really be built

The consequence, for anyone designing entertainment, is a moral choice disguised as a technical decision: you can build for the "during" or for the "after". So that people can't stop, or so that they leave happy.

Building for the "after" means giving up the lever of uncertainty — crates, random prizes, pay-to-win — and betting on honest progression: effort that sets into results, achievement that stays, a world where what you earn is truly yours. It retains less and monetises worse, but it leaves people whole. It is the difference between a business that lives by squeezing its community and one that lives by serving it.

If there were a server in Spanish built for the "after" — no crates and no pay-to-win, with real progression instead of slot machines in disguise — it would be called this: KobiiCraft. Not a place designed so that you can't stop, but so that, when you do stop, you're glad you were there.

The principle, in any case, is bigger than any server: faced with anything that hooks you, don't ask how much you like it while you're doing it; ask yourself how it leaves you when you let it go.

The two three a.m.s

Go back to those two nights from the beginning: the crate one and the building one, both very late, both opposite on the inside.

The difference wasn't in the hours you put in, but in what you were left with. One left you chasing something that never came; the other, a house that was still there when you woke up. And that, at heart, is the whole difference between a design that uses you and one that respects you.

Don't count the hours something steals from you. Count what you are left with when you finally switch it off.

Frequently asked questions

Why are crates and pay-to-win so addictive? Because they rely on the variable-ratio reinforcement described by B. F. Skinner: when a reward arrives unpredictably (sometimes yes, sometimes no), the behaviour that chases it becomes very intense and hard to give up. It is the same mechanism as slot machines: it isn't the prize that hooks you, it's the uncertainty of "maybe this time".

What is the difference between the dopamine of anticipation and of achievement? Anticipation dopamine fires at the expectation of a reward that may or may not arrive, and keeps you pulling the lever without the loop ever closing. Achievement dopamine comes when you complete something real through effort, and leaves you satisfied and able to stop. Both release dopamine, but one squeezes you dry (a loop that never closes) and the other nourishes you (a satisfaction that does).

Why does a sandbox feel different from a crate server? Because its reward isn't unpredictable, but cumulative: you place a block and a block appears, effort sets into a predictable result that stays. There is no "maybe next time" pulling at you, so you can stop whenever you like, with a creation that is yours. It feeds the dopamine of achievement rather than that of anticipation.

What is ethical video game design? Design built with how it leaves the player in mind, not just how long it keeps them. As Misha Klimovsky points out, the problem arises when a system is optimised for engagement without caring about the development of the person using it. Ethical design gives up the levers of uncertainty (crates, pay-to-win) and bets on honest progression: people enjoy themselves and leave whole, even if that retains and monetises less.