I Say No at Night, I Use the Same Tools in the Morning

Eight in the Evening

Eight in the evening, more or less. Dinner just ended, the dishes in the sink, the table still to be cleared. My son is three years old and is trying to convince me to let him keep the tablet for a little longer, which in his grammar of time means a span between twenty minutes and infinity. I say no. Not in anger, not with a parenting lecture. I say no with the precise tiredness of someone who repeats the same scene every evening, knowing the next one will be the same.

So far it is a scene any parent recognises. It could happen in any home.

I Know How That Tablet Works

The difference is the reason I say no. It isn’t because I read an alarming article on screens, or because I follow a paediatrician on Instagram who recommends the digital detox, or because in my day we played outside. The reason is that I know how that tablet works. Not in the generic sense, not “technology is bad for children.” I know in the technical sense. I know how the software running on it is built. I know which design decisions the people who built it made, and I know why they made them. I know what happens when my son touches the screen, what logic decides what to show him next, and what objective that logic is trying to reach. That objective is not his wellbeing.

I know this because it is my craft. I have been building software for nearly twenty years. I don’t work for the big social platforms, I have never had an office in Menlo Park or Mountain View. I do far more boring things: ERP systems, training platforms, applications that help companies organise data. The kind of stuff that would never end up in a journalistic investigation, because it isn’t interesting enough.

Someone might object that the problem doesn’t concern me, that attention-capture techniques belong to TikTok, to Meta, to ByteDance, and that between a social platform and an ERP system there is the difference between a Formula 1 track and a country road. It would be a sensible objection if it weren’t false. The techniques for keeping people glued to a screen are not an industrial secret. They are a shared, documented, taught discipline, discussed at the conferences I attend, debated in the Slack channels where I spend my working days. They are common heritage to anyone who builds digital products. The levers TikTok uses to keep a fifteen-year-old glued to the screen are close cousins of the levers I would use to help a client raise the average session time on a corporate training platform. The context changes, the intensity changes, the user type changes. The basic tools go by the same names, are taught at the same conferences, are found in the same design books.

In the evening, at home, I read the floorplan of my son’s tablet the way you read the blueprint of a building: this serves that, that serves this, here the design pushes you in this direction, there it stops you from going in that one. In the morning I work with the same tools as the people who drew that floorplan. From this fracture comes a book I wrote and decided to make freely available. It is called I tuoi figli non sono i tuoi utenti (Your children are not your users), and it tries to do one very specific thing: transfer a piece of professional knowledge to the people who don’t have it, because the difference between having it and not having it, when it concerns your own children, is enormous.

Skinner, the Pigeon, the Feed

To give a sense of what kind of knowledge we are talking about, I’ll take a single example among many, the one I think is the most important and the least known. Between the 1940s and the 1950s the psychologist B. F. Skinner ran a series of experiments with pigeons in a cage. If the pigeon pecked a disc with its beak, it got food. Skinner discovered something interesting: if food arrived every time the disc was pecked, the pigeon pecked when it was hungry, and that was it. If instead food arrived unpredictably, every three pecks, every five, every two, the pigeon began to peck the disc continuously, compulsively, even when it wasn’t hungry. Variable reward, that mechanism, is the psychological engine of slot machines. It is also the psychological engine of a social network feed, not by poetic analogy but by explicit design.

When your child scrolls Instagram for twenty minutes, most of the content seen is trivial, irrelevant, forgettable. Every now and then something good arrives, something that makes him laugh, a piece of news that interests him, a photo that concerns him. He doesn’t know when it will arrive. He only knows that if he keeps scrolling, sooner or later it will. His brain is doing, in an elegant digital version, exactly what Skinner’s pigeon was doing.

Personalised Random

There is an important difference that makes things worse, and that people outside the industry usually don’t consider. Arcade slot machines distribute wins randomly, because they know nothing about the player. A social network’s algorithm knows what you like, what you pause on for a second longer, what outrages you, what moves you. It uses this information to calibrate content distribution so the variable-reinforcement effect is maximised on you, specifically. It is no longer random, it is personalised random, paradoxically more effective than pure chance because it keeps the unpredictability (you don’t know when the good content will arrive) and adds the relevance (when it does, it is calibrated on your interests). On a teenage brain, whose prefrontal cortex is still under construction, this mechanism is particularly potent.

The Other Mechanisms

Variable reward is one of the mechanisms the book describes. There are others: infinite scroll, which removes decision points; autoplay, which makes “keep watching” the default; notifications calibrated to produce an involuntary physical response; dark patterns that make signing up easy and leaving difficult. They are all documented, all taught, all applied without distinction of age, even to the products your child has on his phone. There is nothing clandestine about any of this. It is open literature, accessible to anyone who wants to read it. The problem is that the people who read it are almost always the people who build those products, rarely the people who use them.

What the Builders Know, and What They Do With Their Own Children

From here comes the question that gives the book its title, and which is the simplest one I know about the whole matter. The Silicon Valley private schools that ban screens in class are not an urban legend. Tim Cook, Apple’s CEO, stated publicly that he would not allow his nephew to use social media. Sean Parker, the first president of Facebook, described the platform’s design as consciously oriented to exploit a vulnerability of human psychology, and added that only God knows what it is doing to our children’s brains. Chamath Palihapitiya, former vice-president of user growth at Facebook, said publicly that his children were not allowed to use that stuff. The testimonies of insiders who admit to protecting their own children from the products they helped build are not hidden, not ambiguous, not scarce. You only need to look.

The observation I make in the book, and that I make here too, is smaller and more ordinary than these famous declarations. Among the colleagues working in software I talk to regularly, the ones with children are almost unanimous on one point: stricter rules on phone use than the average non-tech parent, delayed access to social media, careful control over which apps get installed, conversations with the children about how notifications work and why. It is not a scientific datum, it is an anecdotal observation. But it is consistent, it is recurrent, and its very existence proves the point: people who know the mechanism behave differently.

If the people who build these things keep their own children away from them, what is it that they know and we don’t?

Informational Asymmetry, and the Book

It is not a rhetorical question. There is no secret room where tech executives plan how to harm the world’s children. There is something far more ordinary and far harder to fight, which economists call informational asymmetry. On one side, the people who understand how digital products work, how they are designed, which psychological levers they use and why. On the other side, the people who use them, and let their children use them, without that information. This asymmetry has existed as long as industries have. Food-company chemists know things about food that consumers don’t. Automotive engineers know things about cars that drivers don’t. The difference is that in those sectors, over time, a system of labels, safety standards, and transparency obligations was built. For digital products that system is still largely to be built, and in the meantime our children spend hours inside them every day.

The book tries to close that gap. Not all of it, that would be an excessive claim. A part. It explains how the main mechanisms work in language that doesn’t require technical skills, recounts what we know and what we are still trying to understand about the effects on our children, and tries to outline what each of us can do in our own role: as parents, as citizens, as users, and also, for those of us who work in the industry, as people who build this stuff. It is not a parenting survival manual, and it is not a book against technology. It is an attempt to transfer a small part of professional knowledge to the people who normally don’t get it.

The book is free, in epub and pdf, at ebook.margiovanni.it. If you read it and find it useful, what I ask is that you pass it on to another parent. That handoff between two people who know each other is worth more than any recommendation algorithm, and it is exactly the kind of thing the mechanisms the book describes cannot do.

The author

Andrea MargiovanniAttention

I care about attention as civic raw material. I design products and systems knowing they compete with people's life-time, and I think this is a moral fact before it is a design one.

See the guide →