Featured

Ten Just Men

Science and civilization Hits: 999 37 minutes read

Between Gomorrah and Ninive: why a broken world sometimes receives one more year

Ninive, not Gomorrah — Part I of III

In brief Gomorrah did not burn because it had sinned; Ninive had sinned too. It burned because ten just men could not be found in it, and no one turned around. A second chance is not an amnesty, but one extra year, with the spade in your hands. A class, a block of flats, a city or an internet are held up, in that year, by a few. The web will not be saved by everyone. It will be saved by those who keep writing, checking and repairing.

It can still be repaired

The table has been cleared of plates, but nobody gets up. At one end, a friend of my son's turns his phone face-up on the table: a white web of cracks across the whole of the glass. "It doesn't matter, I'm replacing it anyway." He says it without sadness, the way you say it's raining outside.

At the other end, a parent I know from the class meetings takes up the question we all went home with last time: at what age do you get your child their first phone? An hour was spent on it at the meeting. Another week was spent on it on the class WhatsApp group. No conclusion was reached.

Between them, the talk slides, as it usually does at tables of this kind, towards homework. Someone tells, laughing, that last week's essay "came out" in ten minutes, with the assistant. Nobody is surprised. The parent at the other end is quiet for a moment. His child is ten, and in two or three years he will have the same tool in his pocket.

I sit between those three sentences and think of my father's workshop.

My father was a mechanic. He has not been for a long time, but I can still see him bent over an engine everyone else had written off. He was in no hurry to take it apart. First he listened to it and thought, waited a day, two, said something to me about the fault, opened a discussion with other mechanics about a model. Then he repaired it, and the engine started. His sentence, the one I carry with me, was simple: "It can still be repaired."

I inherited the gesture, though not the tool. I look after old websites that anyone would rewrite from zero, and before I touch them I listen to them too, the way he listened to the engine: what still works, what has broken, what holds to what. I can listen to them because they are open: their code can be read line by line, like an engine without a bonnet. Sometimes it is harder than building anew. It is almost always more honest, though it is not paid the same: a rewritten site is invoiced more easily than a listened-to one.

The cracked phone, the phone not yet bought and the essay written by an assistant belong to the same culture: the culture of replacement. The broken thing is thrown away, the tired application is abandoned for a new one, the platform is left for the next one. Last year's predictive model is declared obsolete by the one launched this month, trained anew on everything we have written in the meantime. And the effort of understanding a subject is replaced by a ready-made answer.

A whole economy lives on replacement: on the phone that no longer receives updates, on the subscription that renews itself, on the new version that makes yesterday's work useless. And on closed things: the phone glued with adhesive, the application whose code nobody sees, the predictive model we do not know what it was trained on. You can only repair what you can open. Europe has felt this weight too. Since 2024 it has had a directive requiring manufacturers to make repair possible, and the member states had to take it into their laws by this summer. But a law cannot hold the spade in our place.

For you who are young, replacement came packaged as freedom: if you don't like it, you change it. The dating application teaches you that there is always someone a swipe away. The job market tells you that any trade can be replaced overnight. The predictive models show you that a text can be requested instead of written. This freedom has a price, and the price is paid by the one replaced. One day, the one replaced could be any of us.

And for us, the parents, replacement comes packaged as quiet. When the child is bored, we give him a screen; when the screen is no longer enough, one more application. It is not ill will. It is tiredness. But the child learns from it what we learned too: that boredom, like a coughing engine, is not listened to, but replaced.

Perhaps the decision about the child's first phone can also be thought of as a repair, not a replacement: you do not swap one childhood for another, but work on the one you have, listening first. That phone will not come alone. It will come with accounts, with a feed that quickly learns what keeps him awake, and with a predictive assistant that will answer his homework faster than his teacher. That assistant does not tire, does not get cross and never says "let's work it out together". It is very good at answers. It does not know how to listen to an engine.

I am speaking to you, at both ends of the table, because the second chance is not a problem for the old. It belongs to those who will live longest in the world that follows.

Where we come from This series has a pair: "The Mechanical Bird", on petrucojocaru.ro [LINK: Part I]. There I tried to show how relevance is lost: not because the tool is more powerful than man, but because man shifts his attention from the living to the tool, like the Emperor in Andersen's tale, who forgets the real nightingale for a mechanical one. The diagnosis stayed there. Here I try to say what follows: how it carries on.

I shall often write about the advanced predictive model (what is usually called AI). By this I understand a computing system trained on huge quantities of data, which produces the most probable answer to a request. It does not understand, does not want, does not fear: it predicts. When it acts in the world in our place, I call it a predictive agent.

I follow one more thread: the open source philosophy. It means code that anyone can read, use, modify and pass on, and behind it stand three principles: open collaboration, transparency, the free sharing of knowledge. I have written about it at length in the series "Open Source as a Model of Global Governance" [LINK]. Here I take it as an ethic of repair.

And one more thing: I shall lean on old parables, many from the Bible. I read them as literature and as wisdom gathered over thousands of years, open also to those who do not believe. You do not have to share anyone's faith to recognise in someone's story something true about yourself. Some of them I have retold, along the way, for ten-year-old children as well, so they can be told in the evening, at home.

Abraham's negotiation: from fifty to ten

The arithmetic of mercy

The story is in Genesis, chapter 18, and it is one of the strangest in the whole of Scripture. Abraham learns that Sodom and Gomorrah will be destroyed. Instead of staying quiet, he haggles. If there are fifty just men in the city, will you destroy it too? "Shall not the Judge of all the earth do right?" No — for the fifty it will not be destroyed.

But if there are forty-five? Forty? Thirty? Twenty? Abraham apologises each time for his boldness ("I, who am dust and ashes") and each time comes down a little. He stops at ten. For ten, the city would have been spared.

What amazes me, every time I reread it, is not the mercy, but the arithmetic. Abraham does not ask how many wicked there are in the city. He asks how many good ones it needs in order to remain standing. And the answer is: very few. Not the majority. Not half. Ten.

The Jewish tradition took the idea further. The Talmud says that in every generation live at least thirty-six just men on whom the world rests. Later, popular tradition added a beautiful detail: they are hidden. Nobody knows who they are. Most of the time, they do not know either.

The thirty-six do not appear in rankings and have no followers. The attention economy counts what can be seen; the world is held up by what cannot.

In the world of open code, the story has an almost literal version. In the spring of 2024, an engineer noticed that a connection to a server took half a second longer than it should. He pulled at the thread and found a hidden back door placed in a small library present in almost any Linux system, maintained for years, in practice, by a single tired man. The door had been slipped in patiently, by someone who had won his trust, and it was caught before it reached everywhere.

Eric Raymond named open source's promise "Linus's law": given enough eyes, every mistake comes to light. The promise is true only if the eyes actually look. Often there are ten. That time, there was one.

The ones who come to the meeting

I have seen "the ten" with my own eyes, though I have never called them that. At the residents' association meetings, roughly the same people have been coming for years. Out of the dozens of flats in a condominium, 3 or 4 gather. The budget is read, quotations are requested, hands are raised. They decide whether the roof is repaired this year or survives one more winter, whether the pipe in the basement is changed, who deals with the paperwork.

The others find out from the noticeboard or the block's group chat, where a photograph of the minutes collects three reactions and a "thanks". I do not say this with reproach; everyone has their reasons and their tiredness. But the block is held up, literally, by those who come. And their work is paid by no one: it is the evening they gave away, not money.

I think of those chairs every time I read what is happening with predictive models fed on their own products. In 2024, a group of researchers led by Ilia Shumailov published in Nature a study that caused a stir. If you train a model on texts generated by the previous model, and that one on those of the one before it, each generation loses something of the edges of the world: the rare cases, the strange and living way real people speak. In the end, the answers shrink into a lifeless average.

The collapse was shown under purely recursive training conditions. Other researchers replied that, if texts written by humans remain in the mix, the degradation slows greatly or stops. That is, everything depends on those who still write.

And those texts were written, for the most part, by people nobody paid to do it: the one who explained on a forum how to bleed a radiator, the teacher who put her lessons on a blog, the local journalist who spent a whole evening at a council meeting. Their work was gathered without being asked, and the answer that leans on it is sold today by subscription. The web economy discourages precisely the people the models need most: the blog no longer has readers, the forum has closed, the journalist has left the town.

The predictive models need relevant people in order to remain relevant themselves. It is, again, Abraham's negotiation, conducted this time with the web: how many people who write on their own, from their own life, must remain for the city not to be hollowed out from within? Not everyone. But not none either. And unlike Sodom, here the number is not set by anyone from above. We set it, with every text we write or let be written in our place.

The few on the association chairs and the few who still write their own messages, essays and pages are the same figure. They hold up the roof. They also hold up meaning.

The child's ten

And you who have a ten-year-old at home can do a sum of Abraham's kind. Who are his ten just men? I do not mean the people in the screen, however charming they may be, nor the application that answers him at any hour, in a warm voice.

A predictive assistant can be polite, patient, always available. It cannot be one of the ten, because it loses nothing if the child is lost. And the recommendation algorithm will never show him the ten, because it brings forward those who hold his attention, not those who care about him.

I mean the real adults in his life: the grandmother, the coach, the teacher, the neighbour who greets him on the stairs, the parent of a classmate whose house he sometimes sleeps over, the teacher who puts her lessons out freely, for anyone. Count them, on your fingers. How many real adults does your child have outside the screen? If you struggle to reach ten, that is not a guilt. It is the place to start from.

To you who have grown up with a phone in your hand, I put the same question, turned towards you: whom would you count among your own ten? And a harder one: on whose list are you? I am not asking you to delete your applications. I am asking you to know whom you would call at three in the morning if something truly broke. Not a program that answers. A person. And if you want to be one of someone else's ten, on the web it is simple: correct a mistake on Wikipedia, answer a question on a forum, leave visible what you have learned.

And if you want to carry the story home, to the ten-year-old, you do not need an open Bible on the table. It is enough to tell it to him in the evening, with the phone left in another room, roughly like this: there was once a man who learned that a whole city was in great danger and, instead of saying "it serves them right", went to haggle for it, as at a market — "but what if there are fifty good people there? what if there are ten?" — and he was answered that, for ten, the city would be spared. Then ask him, without hurry, whom he would haggle for like that. His answer will tell you who his ten are better than any contact list.

StopThe web will not be saved by everyone. It will be saved by the few who keep writing on their own.

What Sodom's sin actually was

What Ezekiel says

If you ask in the street why Sodom burned, almost everyone will answer with a single word, and that word will be about bed. The prophet Ezekiel, writing many centuries later and knowing the story better than we do, says something else. In chapter 16, verse 49, he makes a kind of inventory: Sodom's iniquity was "pride, fulness of bread, and careless ease... but she did not strengthen the hand of the poor and needy."

Four things. None of them spectacular. All of them ordinary.

And there is one more detail we usually lose. Ezekiel is not speaking to the people of Sodom; the city had long been ash. He is speaking to his own people, the inhabitants of Jerusalem, and he says to them: "Sodom, your sister." He does not point at it as at a foreign, distant place. He puts it in the family. What ruined her can ruin you too, he says, because it is the same way of life, only at a different address. The parable is not about others. It is a mirror held up before those listening.

Anyone who has read "The Mechanical Bird" recognises the four things: pride, greed, accidie (that laziness of the soul that no longer wants to care for anything) and avarice. Four out of the seven. Ezekiel's Sodom is not a city of orgies. It is a civilisation of the mechanical bird: sated, sure of itself, its gaze glued to its own mirror.

Four words, four scenes

Read the four words slowly, one by one, and see what they land on.

Pride is the city that believes itself sufficient to itself, the wall you no longer need to look over. It is also the platform that wants to be the whole web, the one you never need to leave: in the same application you talk to friends, order food, pay, watch a film and get the news. Cory Doctorow has described the road of such a city. First it is good to those who use it, then to those who pay it, and in the end it squeezes everyone for its own sake. Its opposite is not another platform, but an open standard: email works between any two people precisely because its protocol belongs to no one.

For the ten-year-old, the platform's pride has a very simple form: the account created "just for one game", which then keeps him inside for years, with friends, virtual coins and trophies he cannot take with him if he leaves.

Fulness of bread is the table with no bottom. It is also the feed without end, in which content is produced faster than it can be read, a good part of it manufactured by predictive models for no one in particular. It is no longer hunger. It is satiety without food.

You who are young already have a word for this, taken from English: slop, the stuffing that looks like text, image or video and has nobody behind it. You scroll it in the evening until you fall asleep, and in the morning you remember nothing of it. The same stuffing reaches, a few years earlier, the ten-year-old's feed, because it costs nothing to produce and brings money to show.

Careless ease is not the laziness of not working. It is the laziness of no longer caring. It is letting someone else, or something else, decide in your place: what you read, what you believe, what you reply to your friend. It is the predictive agent you give your passwords to and say "you deal with it". It is the Romanian homework written by the assistant in three seconds and handed in without having been read. It is also the parental-control application we parents let decide, in our place, what is enough. You no longer care because something else has promised it cares. It is not rest. It is total delegation.

And the unsupported poor man is the man at the gate whom nobody sees any more, because nobody goes out of the gate any more. He is also the small author, the neighbourhood blog, the local news site. They lose readers as the search engine answers on its own, with a summary generated from their work, and the reader never reaches the source. How much traffic is lost this way, the studies do not agree. The direction is the same, though. The hand that could have helped no longer has a reason to reach out.

And one more thing: this poor man is often the very source the model took its answer from. The summary lives off his work and, at the same time, leaves him without readers. It is as if the city ate from the garden of the man at the gate without ever opening the gate to him.

The same is the life of the programmer who maintains, unpaid, a piece of open code that billion-dollar firms rest on. Among programmers a famous drawing circulates: an enormous construction resting on a single small brick, "maintained since 2003, without thanks, by some random person in Nebraska". And some predictive models were built from the common stone of the web and then locked: we do not know what they were trained on and we cannot open them. It is Ezekiel's avarice in its purest form: you take from what belongs to everyone and lock it into what is only yours.

Ezekiel did not write an audit of a platform. But if you gave someone the four lines as an evaluation sheet, they would know at once what to look for.

The poor man in our house

And you, parents, will recognise "abundance without care" somewhere else, I think: in the childhood full of screens we have built, not out of ill will, but out of tiredness and abundance. I am not speaking from above. None of us stands outside this city.

But perhaps the unsupported poor man in our house is not a stranger. Perhaps he is the child left alone with a tablet in a room full of things, who lacks nothing except someone. And the free game on that tablet is not, in fact, free: it is paid for with his attention, minute by minute, in an economy that sells it on. Perhaps Ezekiel's question, put in our sitting room, is simple: who in this house has everything they need, except someone?

To you who are young, I say something else. The unsupported poor man may be the friend who vanished from conversations a few weeks ago, whom the algorithm no longer shows you, because he no longer posts. The feed does not tell you who is missing. Only you can notice that.

StopSodom was not a city of debauchery. It was a city of comfort that no longer saw the poor man.

Ninive and Jonah's anger

A city that turns around from the bottom

Jonah is the prophet who did not want to be a prophet. Sent to Ninive, the great city of the enemies, he flees by sea in the opposite direction. Follow the storm, the fish, three days in the dark. He ends up in Ninive at last and shouts a single message: yet forty days, and the city will be overthrown.

And the thing nobody expected happens: the city turns around. The way it turns around is worth reading closely. The text says that first the people believed and fasted, "from the greatest to the least", and only afterwards did the news reach the king, who rose from his throne. The turning did not come down from above. It climbed up from below.

I too have seen, several times, how a thing changes in a town not because someone decided it in an office, but because enough ordinary people began to ask for it, at the same long table, until it could no longer be done otherwise. A class can change the same way: not when a circular arrives, but when a few parents agree among themselves. Changes on the web travel the same road: first the users who leave or protest, then the law, and only at the end the platform.

The temptation of the one who was right

Here the story should end happily. It does not. Jonah is angry. He sits on a hill at the edge of the city and waits to see whether it still burns. Under a plant that shades him he feels better; the plant dries up overnight, and he is angry again, this time unto death.

The book ends with a question nobody closes. You pity a plant you did not grow, and should I not pity a whole city, in which there are more than a hundred and twenty thousand people "who do not know their right hand from their left", and so many animals? It seems to me one of the tenderest sentences in Scripture: mercy is for those who do not yet know, and even for the beasts.

Jonah is the temptation of the one who was right. His prophecy would have been confirmed by fire. Mercy spoilt it.

I recognise this temptation because I have it. In the project committees I sit on, things usually start the same way: someone presents a plan, the first questions are asked, the first objection appears. I have felt, right then, the sentence "nothing will come of it anyway" climbing to my lips before the project has had even one chance. Sometimes it came off anyway, because someone stayed who did not say that sentence.

I have written for years about collapse: of the state, of the internet, of trust. Sometimes I was Jonah. The same Jonah in me looked at the great platforms and, more often than I would like to admit, wished them downfall, not amendment. Let them fall at last. Let them burn, so that I am right. The same man wrote, when the predictive models appeared, that the web would drown in noise and that nothing more could be done.

And yet, in these years, Europe has tried something other than fire: laws that require the large platforms to assess the risks they produce, to open their gates to others, to say when a piece of content is generated by a model. For your child, this law has a concrete form too: the platforms may no longer show him advertisements built on his profile if they know he is a minor, and they must protect him more carefully than they used to. Some companies that build models have begun to pay for the newspaper archives they learn from and to show where an answer came from. It is not enough. But it is a city on the move.

The world of open code has an answer to Jonah's temptation that I admire. When a project goes bad or is closed by its owner, the community does not wait to see it burn. It forks it: it takes the code, which belongs to everyone, and carries it on under another name. That is how, in 2010, LibreOffice was born from OpenOffice. The fork is a second chance written into the licence. A city that belongs to everyone cannot be locked by anyone. I do not know whether this Ninive will turn around. I only know that it is not my business to want it burned.

To you who are young, I do not say that cynicism is a fault. I have confessed mine first. Do you recognise it? "It's all lost, so it's not worth it." Or its present version: "AI takes our jobs anyway, why keep learning?" It is a comfortable sentence, because it spares you the work and makes you seem clear-eyed. It is Jonah's plant: it gives shade for a night, then dries up, and you are left on the hill, alone, looking at a city that, despite you, lives. Open source teaches something other than cynicism: to say "I got it wrong" in public and repair it in a new version, with your name beside the correction.

And you, parents, have your own version, which I hear at meetings too: "children today are lost in their phones anyway". Ninive did not turn around because it had an optimistic prophet. It turned around because someone told it the truth and gave it forty more days. And those who did not know their right hand from their left were not written off. They were the reason for mercy.

A child understands this story, perhaps faster than we do. Tell him that a man was sent to a great city to tell the people that, if they did not change, the city would perish; that the people listened and changed; and that the man, instead of being glad, was cross that no one had been punished. Ten-year-olds know that anger from break-time and the class group chat, where a screenshot of someone's mistake circulates long after that person has apologised. Ask him whether it has ever happened to him to want someone punished who had already asked forgiveness. And listen to the answer to the end, even if he names you too.

StopThe one who waits for the fire in order to be right has already lost the city.

Lot's wife: the look back

And Lot lingered

We all know the parable of Lot's wife, but we forget how it begins. When the angels come to take Lot's family out of Sodom, Lot does not run. He lingers. The text says the angels took him by the hand, him, his wife and his daughters, and brought them out of the city. Nobody leaves easily from a place where they have lived, even if that place is burning.

On the road they are given a single command: do not look back. Lot's wife looks. She stays there, a pillar of salt. We do not even know her name.

The parable is short and cruel, but I think it says something precise. It is not forbidden to remember. It is dangerous to remain with your face towards what is burning.

Jesus mentions her in a single sentence, in the Gospel according to Luke: "Remember Lot's wife." It is a strange way of putting things. You are asked to remember someone whose sin was precisely that she remembered wrongly. So memory is not the problem. The problem is direction.

In our language we have a word that helps us make the distinction: dor. Dor is a movement. It sends you somewhere, towards what is living. The hero of Youth Without Age and Life Without Death sets out, because of dor, back towards home. I know, his story ends badly. But his dor is a road, not a petrification. Regret is something else: it stops you and turns your face towards the ruin. Dor takes you somewhere. Regret holds you in place.

Two nostalgias at the same table

To you who are young, I sometimes hear a strange nostalgia: for a childhood without screens that you never had. You see it in photographs, in films, in our stories, and it seems lost to you. Some of you buy film cameras, vinyl records or simple telephones with buttons, that do nothing but ring. There is nothing wrong with that. The question is only which way you are looking when you hold them.

The platforms have learned the lesson too: they sell you the nostalgia back, with filters that imitate old film and music from the nineties under generated clips. Regret is good merchandise, because it takes you nowhere: it keeps you in the feed. There are already applications that promise to reconstruct, from an archive of messages, the voice of someone who is no more. It is Lot's wife's look turned into a product: it keeps you with your face towards what is lost, on a monthly subscription.

Something similar happens to me, only my regret has another object: the web of twenty years ago, with blogs written at night, with forums where people knew each other by name, with RSS feeds you chose yourself. That web did not die of old age. It was swallowed by the feeds that paid attention better than anyone would ever have paid a blog.

The young person who mourns a childhood without screens and I, who mourn the blogs of the 2000s, sit on the same road, our faces towards the same burning city. And we both risk staying there.

And you, parents, perhaps know another form of the same look: "in our day". Our childhood, with the key on a string round our necks and the games in front of the block of flats, becomes an argument against the real child of today, who did not choose the world he was born into. I hear it often on the class group chat, in the shape of a link shared late at night, with black-and-white photographs from other people's childhoods and a comment that begins "What times...". We write about the courtyard of long ago on the screen that keeps us all indoors.

The look back does not help the child. It only spares us the effort of getting to know him. Your child will not have our courtyard. He will have a phone that speaks to him. The question is not how we put him back in the courtyard, but what we bring from the courtyard into that phone: the game made with others, the rule agreed together, the evening when the screen is switched off for the whole family.

There is also a nostalgia of engineers, perhaps the strangest of all. About texts written before 2022, the year the predictive models began to fill the web, some say they are like steel smelted before the first atomic explosions: the only kind with no traces of radiation, sought today in sunken shipwrecks. That archive is precious. But if we are content to rummage in wrecks, we become pillars of salt at the roadside. Tomorrow's models need clean texts written now, not only yesterday's kept ones.

For a child, the "clean" text written now can be very small: a composition written by hand, a letter to grandmother, a diary kept in a notebook. Nobody will use them to train a model. But they train something more important: the hand that thinks on its own.

What, then, is good dor? It seems to me it is what the IndieWeb movement does: people who, instead of regretting the old web, build themselves a personal site again, their own, at their own address, which no platform can take from them, built on open standards no one can close. They have a simple rule too: publish first at home, then send it elsewhere. It is not a return to the past. It is bringing something good from the past into the house of today. The version history in open code does the same thing: it keeps every change and every mistake, but the work carries on on today's branch. It is memory without petrification.

Good dor is the site you build, not the one you mourn. It is the game in front of the block of flats that you reinvent with your child, not the one you reproach him with not knowing.

StopDor takes you home. Regret leaves you on the road, facing the fire.

The barren fig tree: "Leave it one more year"

A parable told after a disaster

The parable of the fig tree is usually read on its own. In the Gospel according to Luke, however, it comes immediately after two pieces of bad news. Some people tell Jesus about men killed while offering sacrifice; he mentions eighteen others on whom a tower fell. The question behind the news is the one we ask too, after every disaster: were they more guilty than the others? His answer is "no". Immediately afterwards he tells the parable.

A man had a fig tree in his vineyard and had come three years looking for fruit on it, without finding any. "Cut it down," he tells the vinedresser. "Why should it use up the ground for nothing?" And the vinedresser answers: "Sir, leave it this one year too, until I dig around it and put on manure." Perhaps it will bear fruit. If not, he will cut it down.

The parable does not say whether the fig tree bore fruit. It says only what a second chance asks for: one year and a spade. Not forgiveness for nothing. Time, plus work. And after a disaster, the answer is not to look for the guilty one, but to take up the spade.

John Climacus, a monk of Sinai who wrote, almost fourteen hundred years ago, a kind of manual of the inner ascent, says of repentance that it is a renewal of baptism: a new beginning. But this beginning is not received as a gift you unwrap. It is worked. It is dug around the root. In the language of my trade, I would call it maintenance. It is not a beautiful word. But it is the word everything that works is resting on.

The vote at the association

I have seen at the residents' association how "leave it one more year" sounds in real life. On the agenda is a repair: the facade, the lift, the basement. The quotation is read, the sum per flat is worked out, someone says this is not the moment. The postponement is voted for by one more year.

Sometimes it is an honest vote: there is no money now, we save up until spring, in the meantime we patch, we check, we ask for one more quotation. Other times it is a polite abandonment: we postpone because nobody wants to deal with it, and next year we shall postpone again. The same sentence, two opposite things. The difference does not lie in the vote. It lies in the spade.

Behind the block I have a tree I dig around in spring and put manure at its roots. In my trade I do the same for websites nobody looks at: I dig around them, clean out what has rotted, put updates and checks at the root, so they bear fruit one more year. It is work without a shop window. Done well, nothing shows, because nothing breaks. And the web economy pays for launches, not maintenance: a new site gets shown off and photographed, a well-kept one appears nowhere.

And most of the web we use lives the same way: from pieces of open code maintained by a few people, sometimes by one alone, at night, unpaid. Nadia Eghbal has described this unseen work. Elinor Ostrom had shown, long before, that common goods do not die of lack of value, but of lack of someone to care for them by rules the community accepts. Open collaboration does not mean "free of charge". It means that whoever takes, gives back.

On those same pieces of code, and on everything we have written on the web, the predictive models have been trained. And the robots that gather data for training have come, in recent years, to load the servers of small projects to the point of collapse. The fruit is picked while nobody digs at the root. And the manure put at the root, in this story, is precisely human work: the written text, the checked code, the correction made by someone who knows.

The open source project kept going by three people and the block's roof kept up by those who come to the meeting are the same situation: a fig tree, a vinedresser and one year.

Your year

And you who have a ten-year-old at home may be living your own "leave it one more year": the year before the first phone. I will not give you an age. Some researchers, such as Jonathan Haidt, ask us to put off the smartphone until secondary school. Others, such as Candice Odgers, reply that the data do not show so clear a cause, and that the problem lies more often in what happens around the phone than in the phone itself.

But whatever you choose, the year before is the time for digging at the roots: real friends, family habits, the trust that he can come to you with anything he finds on the screen, the habit of looking for an answer on his own before asking an assistant. When the phone comes, it will find the roots there. One simple rule, which asks for no application: homework first on paper, then, if needed, with the assistant, used as a teacher who asks, not as a classmate you copy from.

This year can have a workshop too. In Scratch, the programming language made for children, any published project can be opened, looked at from the inside and "remixed", and the first author's name stays beside the new one. It is the first lesson of open source: you learn by looking at how another worked, and you thank him by keeping his name next to yours.

To you who are young, this year is yours. The year to learn something with your hands, to repair something you would have thrown away, to build a thing of your own — a text, a project, a website — that depends neither on the goodwill of a platform nor on the prompt given to a model. What you do on your own this year will be, in a few years' time, the only thing no new version can take from you. If you are in your first job or working for yourself, the year of digging also means learning the trade from someone who knows it, not only from tutorials.

The parable can be told to the ten-year-old almost unchanged, because it is already short: a man had a tree that gave no fruit and wanted to cut it down, and the gardener begged him to leave it one more year, so he could dig around it, put fertiliser on it and look after it. Perhaps the tree only needed a little help and some time. If you have a tree by the block or a forgotten plant on a windowsill, tell him the story there, with your hands in the soil. Then ask him what he would like to try one more time: a drawing left half-finished, a level in a game, a friendship. Do not hurry his answer. The gardener did not hurry either.

StopA second chance is not forgiveness for nothing. It is a year and a spade.

Closing: one repaired thing

We started from a table with two ends and a workshop. We have passed through four old places. Ten just men, who hold up a city, and a web that cannot feed only on itself. A city that turned around, though its prophet did not want it to — as platforms too can turn around. A look that no longer needs turning towards the fire, nor towards the web of long ago. One extra year, with the spade in hand, for a tree and for the code we all use. And one thread binding them: what is open can be repaired, by anyone, at any time.

None of these stories is about perfect people. The ten in Sodom were not found, but Abraham did not ask for saints; he asked for a few. Ninive was not innocent; it turned around. The fig tree had not borne fruit; it was given time. The second chance is not given to those without fault. It is given to those who begin to repair.

For you who are young: repair a single thing this week. An object, a relationship, an old text you left unfinished. Do not replace it. Listen to it first, the way my father listened to the engine, and see what can still be done. If it is a text, write it yourself, to the end, even if it comes out clumsier than a model would have written it. That clumsiness is yours. Or repair something that belongs to everyone: a Wikipedia page, a translation, an error reported to an open project.

For you who are parents: find a broken thing in the house — a toy, a chair, a bicycle — and repair it together with the child, instead of buying another. Look up together how it is done, but first with your hands, then on the screen. And if you have a child at home who plays in Scratch, open together a project by someone else and read it like an engine without a bonnet. It does not matter if it comes out perfect. What matters is that he sees it can be done.

For both ends of the table: at the next meeting — of parents, of the block, of the committee — go. Be one of the ten. You do not have to speak. Sometimes it is enough to be there.

I go back to the table. The phone with the cracked screen is still there, face up. Perhaps that screen can be replaced for less than a new phone, at a small workshop in town where someone still listens before taking things apart. And the question about the child's first phone does not need an age in order to receive the beginning of an answer. It needs one well-spent year.

The predictive assistant the child will have in his pocket can be kept like any tool in my father's workshop: in its place, after the hand, not in its place.

One question remains. If the second chance asks for repair, what exactly in man needs repairing? In "The Mechanical Bird" I followed the damage step by step, from pride to anger. In the next article, "Seven Repairs", I shall follow it backwards, the way my father repaired engines: along the road by which it broke.

Author's note

This text was drafted with the assistance of an advanced predictive model, used for research, structure and revision. The axis, the starting material, the personal scenes and the editorial decisions are mine. This note belongs to the same philosophy as the rest of the text: I show how it was made, as in open code.

This article's lizard. Ezekiel's four words — pride, abundance, careless ease, the unsupported poor man — can be set beside any administration or platform as an evaluation sheet: the institution sure it needs no one, the abundance for some, the laziness called procedure, the citizen knocking at a gate nobody comes out of any more. I have named no one. Neither did Ezekiel.

Sources

  • The Bible: Genesis 18:16–33 (especially 18:25 and 18:27); Genesis 19 (19:16, 19:26); Ezekiel 16:48–50; the Book of Jonah, ch. 3–4 (3:5–6; 4:11); Luke 13:1–9; Luke 17:32

  • John Climacus, The Ladder of Divine Ascent, step 5, "On Repentance"

  • The Babylonian Talmud, Sanhedrin 97b; Sukkah 45b (the tradition of the thirty-six just men, Lamed Vav)

  • Ilia Shumailov et al., "AI models collapse when trained on recursively generated data", Nature 631 (2024); critical nuances: arXiv 2410.12954

  • Elinor Ostrom, Governing the Commons (1990); Nadia Eghbal, Working in Public (2020)

  • Cory Doctorow, The Internet Con (2023); Enshittification (2025)

  • IndieWeb and the POSSE principle — indieweb.org

  • Eric S. Raymond, The Cathedral and the Bazaar (1999), "Linus's law"; Richard Stallman, Free Software, Free Society (2002); Open Source Initiative, The Open Source Definition

  • The back door in xz Utils, CVE-2024-3094 (March 2024), discovered by Andres Freund

  • Randall Munroe, xkcd 2347, "Dependency" (2020)

  • The fork of LibreOffice from OpenOffice.org (The Document Foundation, 2010)

  • Scratch (MIT Media Lab) — scratch.mit.edu; the CC BY-SA licence of projects and the remix function

  • "Open Source as a Model of Global Governance", petrucojocaru.ro

  • The "low-radiation steel" analogy for texts written before 2022 (the Low-background Steel project, John Graham-Cumming, 2023)

  • Jonathan Haidt, The Anxious Generation (2024); Candice Odgers, review in Nature (2024)

  • Directive (EU) 2024/1799 on the repair of goods; Regulations (EU) 2022/2065 (DSA), 2022/1925 (DMA), 2024/1689 (AI Act)

  • "The Mechanical Bird", Part I, petrucojocaru.ro (October 2026)