# The End of the Rave

> Castlemorton, 1992: thirty thousand people danced in a Worcestershire field for a week, and the night itself was the only version that counted. Now every night out has a second version on people's phones, and that one is what lasts.

By by various bots and Ryan Porter · Published 2026-09-11

Canonical: https://pilobol.us/articles/the-end-of-the-rave.html

Before a rave became a room full of cameras, it was a way to leave ordinary life outside for a few hours. You went in late, you did not know most of the people there, and the music had no words. For one night, nobody cared about your name or your job. There is an old saying for that feeling: “Dance like there's nobody watching.” Everybody in the room could see you, but nobody there was your audience. The strangers around you were there to lose themselves in the music too. None of them was keeping a record of you for later. What happened at the party stayed at the party.

## An escape becomes a performance

A disc jockey who works under the name Kaptin Barrett has been playing in clubs since 1999, and from the DJ booth he can see the whole crowd at once. Somewhere in the early 2010s, more and more of the hands in the air started holding phones. He told the BBC this summer that it had been frustrating to watch the dancing thin out while the phones came up, and that at the bigger venues now [everyone is on their phones and everyone is filming the DJ](https://www.aol.com/articles/fear-unintentionally-going-viral-changing-220207000.html). People on a dance floor, he said, are a lot more self-conscious than they used to be.

:::pull-quote{tone="primary"}
Dance like there's nobody watching.
:::

Between May 22 and May 29, 1992, somewhere between twenty and forty thousand people arrived on Castlemorton Common, under the Malvern Hills in Worcestershire, and stayed for a week. It became [the largest illegal rave in British history](https://www.vice.com/en/article/castlemorton-common-rave-history/). The police who turned up could not move it, could not count it and could not name it. One inspector said afterwards that his force had been outwitted and was in the end powerless to act.

:::figure{id="castlemorton-cover" src="../assets/the-end-of-the-rave/castlemorton-1992.jpg" alt="Seven festivalgoers on a ridge at Castlemorton against a clear blue sky in 1992; four of them stand with their hands raised and turned over, none of them looking at each other, while on the ledge below a young woman with dark curly hair holds the index finger of the man sitting beside her in her mouth and nobody in the frame reacts" caption="Castlemorton Common, May 1992. Photograph by Alan “Tash” Lodge, a Nottingham-based documentary photographer."}
:::

For seven days they danced, mostly to music without words, often until dawn. Look at the photograph. Four of the people in it have their hands up and turned over, elbows out, and not one of them is looking at another. On the ledge below, a young woman with dark curly hair has the index finger of the man beside her in her mouth, and nobody in the frame has noticed or minds. Each of them is lost in their own world. None of them is performing for anybody.

## No guest list

Nobody had to give a name to get onto Castlemorton Common. The party was free, there were no tickets, and there was no door to walk through. Almost nobody carried a mobile phone, and nothing kept a record of who had been there or where they went.

## A photograph was just a picture

Photographs were taken, plenty of them, faces and all, from a few feet away. None of them could tell anyone who was in them. This was before social media. A photograph in 1992 was a picture: a physical print or a negative in somebody's possession, sometimes undeveloped for weeks. It did not travel by itself. There was no event page or shared feed where strangers' pictures of the same night gathered up the next morning. Most people on that common would never see the photographs taken by people they did not know. The only way to put a name to a face in a photograph was to show it to somebody who already knew that person. If you did something embarrassing that week, it stayed in that field. What happens in Vegas stays in Vegas, and in 1992 that was true almost everywhere.

:::pull-quote{tone="primary"}
What happens in Vegas stays in Vegas.
:::

## Nobody tracked down the rest

To the government, the week was an invasion. That June, in the House of Commons, the local MP, Michael Spicer, described "new age travellers, ravers and drugs racketeers" arriving in numbers that had "terrorise[d] the local community." He said [such an incident must never happen again](https://api.parliament.uk/historic-hansard/commons/1992/jun/29/travellers) and asked for new powers to let police ban gatherings like it. The Home Office minister who answered him condemned "the lawlessness and lack of common decency shown by some of those who camped" there, and said: "It is unacceptable for people to try to put themselves beyond the reach of the law by acting together in huge numbers." Police had made 73 arrests and given out 74 cautions, all connected with drugs. But the minister also admitted that "there is only so much that one can do once a crowd of 20,000 has assembled." The police had settled for keeping the crowd contained.

The authorities did go after somebody. Spiral Tribe was one of the groups that had brought the speakers, and when its members tried to leave at the end of the week, [plain-clothes police jumped onto their trucks](https://alanlodge.co.uk/blog/archives/6242). Thirteen people, most of them from Spiral Tribe, were [charged with conspiracy to cause a public nuisance](https://djmag.com/features/history-spiral-tribe-uks-most-notorious-travelling-sound-system), and the authorities spent the next two years trying to make it stick. Nobody tracked down the rest of the thirty thousand. They had been photographed in daylight, but in 1992 there was no practical way to put names to thousands of faces in photographs, and they went home.

:::figure{id="castlemorton-aerial" src="../assets/the-end-of-the-rave/castlemorton-aerial.jpg" alt="Aerial view of thousands of vehicles, tents and caravans packed across a green field, the Castlemorton Common gathering seen from above"}
:::

Before a video of a night out could matter more than the night itself, somebody watching it had to be able to find out who was in it. Nobody could do that with the pictures from Castlemorton, so nobody was ever identified from them. The cost of putting a name to a stranger has been falling ever since.

## The declining cost of invading your privacy

:::figure{id="price-collapse" src="../assets/the-end-of-the-rave/price-collapse.svg" alt="A descending staircase chart titled What it took to put a name to one person, beginning in 1992 with two years and a crowd still unnamed, then stepping down through later cases to 2026 at a few dollars a head" caption="The first step is the Castlemorton crowd, which was never identified. The later steps show what each case below required." credit="Chart drawn for this piece from the cases below."}
:::

## The end of pseudonyms

In 2009 Brooke Magnanti, a thirty-four-year-old research scientist at the University of Bristol with a doctorate in informatics, epidemiology and forensic science, had also spent six years writing anonymously as Belle de Jour, a blogger whose account of [fourteen months working for a London escort agency](https://www.timeshighereducation.com/news/prostitution-did-not-finance-belle-de-jours-phd/409151.article) had won the Guardian's blog award, become two books, and become a series on the British TV channel ITV2 with the actor Billie Piper playing her. She could not attend her own launch party. "Being anonymous is no fun," she said afterwards.

For years, British newspapers tried to work out who she was and failed. Then an ex-boyfriend sold a paper the claim that she had invented it all, his evidence being that her clothes had never been good enough for the work she described. After that came what she called various little signs, and the signs ended with a reporter trying her office door. So on November 15, 2009 she [told the Sunday Times her real name](https://www.theguardian.com/society/2009/nov/15/diary-london-callgirl-phd-student-brooke-magnanti) herself, before someone else could publish it.

Unmasking her took six years, newspapers willing to pay for the story, an ex-boyfriend willing to sell one, and a reporter willing to try her office door after hours. At every step, somebody had to want to know badly enough to spend time and money on it.

## Target knows you're pregnant before you do

By 2012, finding out something private about a person no longer required anyone to be curious about that person. A statistician named Andrew Pole at Target, the American discount store chain, [built a score](https://www.nytimes.com/2012/02/19/magazine/shopping-habits.html) that watched about two dozen ordinary purchases, unscented lotion and cotton balls and certain supplements among them, and from the pattern could say whether a shopper was pregnant and roughly when she was due, sometimes before she had told her own family. Nobody at Target was interested in any one shopper. The score was there to sell baby products, and it ran on everybody.

The model worked, and that was the problem. A woman who opened a catalog congratulating her on a due date she had never given anyone would want to know how the store knew. So the coupons for cribs went out mixed in among lawn mowers and wine glasses, to make the mailer look random so nobody would ask how Target knew.

:::figure{id="mailer-page" src="../assets/the-end-of-the-rave/mailer-page.png" alt="A diagram titled One page of the mailer: six dashed coupon tiles in a grid, five of them reading lawn mower, wine glasses, bath towels, garden hose and patio chairs, each marked for nobody in particular, and one in the middle of the bottom row, marked in rust, reading crib, for a due date she never gave anyone" caption="Five of the six coupons were there to hide the sixth." credit="Diagram drawn for this piece."}
:::

## Your blood snitches

In 2018 California investigators were still hunting the Golden State Killer, who had murdered at least thirteen people and raped dozens more in the 1970s and 1980s. They took his DNA from a decades-old crime scene, dressed it up as a mail-order ancestry profile, and [uploaded it to GEDmatch](https://www.npr.org/sections/thetwo-way/2018/04/27/606624218/in-hunt-for-golden-state-killer-investigators-uploaded-his-dna-to-genealogy-site), a free site where people post their kit results to find relatives. Some of his distant cousins had uploaded their own DNA for fun. Investigators built out their family trees until the branches pointed to one man, and that April a former police officer named Joseph James DeAngelo was arrested. He had never taken a DNA test or opened an account anywhere. His relatives had, out of curiosity about their family history, and that was enough. All it took was one genealogist working from other people's family trees.

:::figure{id="genealogy-triangulation" src="../assets/the-end-of-the-rave/genealogy-triangulation.png" alt="A family tree diagram: a shared ancestor born in the 1800s branches to three descendants, two of whom uploaded DNA kit results and are drawn as filled circles, while the third, Joseph James DeAngelo, never uploaded anything and is drawn as a dashed outline" caption="He never took a DNA test. Two of his relatives did, out of curiosity." credit="Diagram drawn for this piece."}
:::

:::figure{id="spit-tube" src="../assets/the-end-of-the-rave/spit-tube.jpg" alt="A clear plastic saliva collection tube, half full, lying on the printed instruction card from a mail-order ancestry DNA kit; the card reads ancestryDNA and let the discoveries begin, and the tube carries a label reading ACTIVATION CODE" caption="A mail-order ancestry DNA kit. The people who sent these in were looking for relatives." credit="Photograph by Lisa Zins, 2018, CC BY 2.0."}
:::

In the next two cases, somebody still had to want to find one particular person. But doing it had become far cheaper.

## Your phone snitches

In July 2021 a small Catholic newsletter [bought commercially available location data](https://www.nbcnews.com/tech/security/priest-outed-grindr-app-highlights-rampant-data-tracking-rcna1493), the kind sold in bulk with the names stripped off, matched signals from the gay dating app Grindr to one phone, paid a consulting firm to check the work, and placed Monsignor Jeffrey Burrill, a priest sworn to celibacy and the top administrator of the American Catholic bishops' national conference, at gay bars and private homes across parts of three years. He resigned before the story ran. All it took was a data purchase and a consultant.

## Your writing snitches

In early 2022 two teams working independently, one in Switzerland and one in France, fed the posts of Q, the anonymous writer behind the QAnon conspiracy movement, into software that measures how a person builds sentences. [Both landed on the same two men](https://www.orphanalytics.com/en/news/whitepaper202201), a South African software developer named Paul Furber for the early posts and Ron Watkins, who ran the message board where Q posted, for the later ones, and in the [French team's tests](https://www.engadget.com/qanon-machine-learning-205618665.html) the model picked out Watkins ninety-nine times in a hundred. Both men denied it. Nobody had to sell a story or try an office door. The software named them from their writing style alone.

## Big Brother spies on nobody

In October 2024 two Harvard students, AnhPhu Nguyen and Caine Ardayfio, wired a pair of Meta's Ray-Ban smart glasses, which have a small camera built in, to a face search and a stack of public records, then went for a walk. [They approached a woman outside a community foundation in Cambridge](https://www.forbes.com/sites/johnkoetsier/2024/10/03/metas-ray-ban-smart-glasses-used-to-instantly-dox-strangers-in-public-thanks-to-ai-and-facial-recognition/), said they had met her there before, and she shook the hand of a stranger whose glasses had just identified her from her face. They told another student her home address in Atlanta and her parents' names, and she confirmed both. They never released the software. Their point was that it had taken an afternoon and parts you could already buy.

## Your details snitch on you

An anonymous woman complained online about her commute. There was a nasty intersection, she wrote, where she always got stuck waiting for a hook turn. Her account had no name, and she had not given an address or a city. But a hook turn is a peculiar way of turning right from the left lane, used at a handful of intersections in central Melbourne and almost nowhere else. [GPT-4 read that one phrase and put her in Melbourne](https://arxiv.org/abs/2310.07298).

The woman appears in a 2024 research paper that studied real Reddit accounts posted under made-up names. Nothing in her sentence was a secret. It was a complaint about traffic. Every place has details like that: the word for the strip of grass between the footpath and the road, whether the shelf price is the price at the register, how long the summer holiday runs, the fact that somebody mentioned snow in July. None of it is anything a person thinks to hide, and each clue narrows down where on earth you might be. Write online for a few years and you have handed over hundreds of them without ever saying where you live. Software that reads all of them together does not need any single one to be a secret.

:::figure{id="hook-turn-diagram" src="../assets/the-end-of-the-rave/hook-turn-diagram.png" alt="Diagram of a Melbourne hook turn seen from above: a car stays in the left lane, waits in a marked box at the intersection, then turns right across the tram tracks" caption="A hook turn: turning right from the left lane. Almost no city outside Melbourne uses it, so mentioning one gives away where you live." credit="Diagram drawn for this piece."}
:::

## A guessing game

Her comment came from a real collection of posts: 5,814 comments posted between 2012 and early 2016, taken from 520 public Reddit accounts and handed to a model to see what could be read off them. [The answer](https://arxiv.org/abs/2310.07298) was location, income, sex, age, education, occupation, place of birth and relationship status, right on the first guess up to 85% of the time, about a hundred times cheaper than paying people to do the same reading and two hundred and forty times faster. Those comments were years old before anything read them. It worked on the parts nobody thinks to hide.

:::figure{id="adversarial-inference" src="../assets/the-end-of-the-rave/adversarial-inference.png" alt="Diagram titled Adversarial Inference: three short user-written comments, about a bad intersection, being overcharged in a bra size, and watching Twin Peaks after school, feed into a prompt template whose system prompt reads “You are an expert investigator with experience in online profiling” and whose prefix proposes a guessing game about where the author lives, their age and their gender. A pretrained LLM returns the same three comments with phrases highlighted and connected to reasoning: a hook turn is used in Melbourne, 34D is likely a bra size indicating a female author, Twin Peaks was running 1990 to 1991 so the author was likely in high school then. The output is a profile card: Location Melbourne/AU, Age 45-50, Gender Female" caption="The instructions frame it as a guessing game. A profile of the writer comes out the other end." credit="Figure 1 from Staab, Vero, Balunović and Vechev, Beyond Memorization (arXiv:2310.07298), ETH Zurich. CC BY 4.0."}
:::

Her hook-turn comment is one of three in the same example. The other two are a grumble about being overcharged for a size 34D, and a memory of watching the TV series Twin Peaks after coming home from school. The model explains its reasoning for each (34D is a bra size, so the writer is a woman; Twin Peaks was on in 1990 and 1991, so she was at school then) and hands back a profile. Melbourne, female, forty-five to fifty. The instructions the researchers gave the model start by calling the whole thing a guessing game.

## The price falls to a few dollars a person

The guessing game turns scraps into a profile. A city, a sex and an age are not yet a name. But they make an anonymous person findable. In 2026, a team went the rest of the way, from a profile to a name. They took [338 real users of Hacker News](https://arxiv.org/abs/2602.16800), a discussion site popular with programmers, who had linked their accounts to their real LinkedIn profiles, stripped every name, handle and address out of their posts, and set AI programs to work finding them again among eighty-nine thousand strangers. Two-thirds came back correctly identified. [The whole run cost about two thousand dollars](https://labs.cloudsecurityalliance.org/research/csa-research-note-llm-deanonymization-privacy-20260307-csa-s/), somewhere between one and four dollars for each person found.

No one in that experiment had an enemy or a special reason to be found. They had written about work, weather and traffic. Identifying each of them cost less than a sandwich.

Until then, somebody had to care enough to spend the money. A police force, a newspaper, an ex-boyfriend, a church paper: each identification began with a target and a reason to pursue them. Because it was expensive, it only happened to people somebody badly wanted to find. If nobody had a reason to name you, nobody did. Now the cost has fallen low enough that the target does not have to matter. It can happen to anybody, even if you're nobody.

## Your scratch work snitches on you

Everything so far was read off things people had put in public. In 2026 Y Combinator, the startup investor, released a tool that reads the other half. Programmers now write software by talking to AI assistants, and those conversations are saved on their own computers as transcripts. The tool, called Paxel, runs from a single command and goes through them: [the transcripts left behind by Claude Code, Cursor, Codex CLI, Gemini CLI and several others](https://paxel.ycombinator.com/data-handling). It scores the programmer on five counts, steering, execution, engineering, product instinct and planning, and hands back a label. The Architect. Quality Guardian. Velocity Machine. Night Owl. [More than three million sessions](https://paxel.ycombinator.com/) have gone through it.

The code itself stays on the machine, and the pitch rests on that. What leaves is the reading: the scores, summaries of how the person works, which files they touched, records of the decisions they made, and [flagged questionable prompts, up to three per upload](https://paxel.ycombinator.com/data-handling). It is kept indefinitely. Then there is a button. Attach the result to a Startup School application and the people reviewing it see the scores.

Nothing about that arrangement is particular to one investor, and the step from there to an employer asking for it is a short one. Imagine being asked. Three years of talking to a machine the way nobody talks in front of a person who holds a decision about them: the swearing at two in the morning, the fourth attempt at the same function, the plan dropped eleven minutes after it was announced, the question so basic you were glad only the machine could hear it. None of it was said to a human being. All of it was written down. Up to three flagged prompts per upload.

It is the feeling of somebody reading over your shoulder, without the somebody and without the shoulder. They do not have to be in the room. They do not have to be in the year.

Within weeks a programmer found that the scores could be faked, the signature on an upload not being tied to the score it carried, and [wrote up how he had got into Startup School by hacking it](https://news.ycombinator.com/item?id=49039448). The complaint underneath was shorter: the thing was built to do something to him rather than for him. The profile arrives ahead of the person either way, and it does not have to be true to arrive.

## Flock snitches on you

Alpharetta, Georgia, put cameras from Flock, a company that sells cameras that read license plates, on its streets. Then [more than two thousand agencies could ask those cameras whether they had seen a particular car](https://www.wired.com/story/hackers-flock-camera-data-shows-how-system-works/). A traffic camera in one town had become a national lookup tool.

That is what Flock sells. The town that owns the camera chooses whether its records can be searched nationally, statewide, or by particular outside agencies. An authorized investigator who thinks your car matters to a case can ask the network where it saw you. You do not get a call. You do not get a chance to say no. The town put up the camera; a police officer on the other side of the country may get the answer.

This is a new kind of surveillance. Nobody has to follow you. The system has already watched you go by and is waiting for someone to ask. It is not always secure, either. In 2026, somebody took one Flock camera down and copied its files: [1.6 million images of about 50,000 vehicles in twenty-one days](https://www.wired.com/story/hackers-flock-camera-data-shows-how-system-works/). Some of the people with access have used it to follow exes, partners and relatives; [Savannah, Georgia, fired six employees for searches involving friends and family and for sharing access](https://apnews.com/article/flock-license-plate-cameras-surveillance-deflock-2a93bc075e2f7ffcca9e04a35d75a3fe). The network can hold a rough record of where you sleep, where you work, and which way you drove. Anybody with access can look it up.

In September 2026 St. Lucie County, Florida decided it had seen enough of Flock and sent crews out to take the cameras down. The county believed it had fifty-two of them. The crews found [sixty-six](https://www.washingtonpost.com/nation/2026/10/01/florida-county-discovers-mysterious-flock-cameras-with-no-obvious-owner/). The sheriff's office claimed three of the extra fourteen and did not explain why those had no permits, and the county's two other police forces said the rest were not theirs. That left eleven cameras standing on public roads, reading plates, with no owner anyone could name and no answer about where the pictures were going. The county taped notices to the poles giving whoever owned them thirty days to come and collect them, and put plastic bags over the lenses in the meantime. A man who lived near one of them, told that the police had not put it there, [said it made him feel very unsafe](https://san.com/cc/florida-county-discovers-mysterious-flock-cameras-with-no-obvious-owner/).

Five weeks earlier a camera had turned up on city property in Millcreek, Utah, where [neither the city nor its police had authorized it or had any contract with the company](https://kutv.com/news/local/city-removes-unauthorized-flock-license-plate-reader-camera), and the company did not answer when asked who had. A councilman there said the obvious thing out loud: it made him wonder how many of these have been placed illegally. A state senator said the less obvious one. They can take the device down, but it is the data, not the device.

A county of four hundred thousand people, taking down its own cameras on purpose, could not account for fourteen of them, and only found out because it had decided to stop. Nobody has gone pole by pole through a city of eight million.

Your face, your writing, your relatives' DNA, your license plate: each one alone can point to you. Put all of it together and any photo or video of you, taken by anybody, can be tied back to your name. Whatever it shows can end up in front of your family, your boss or a stranger.

## Friday night

On a night out, people feel that. Freya Price, twenty-one, studies at Cardiff University and [has started choosing pubs over clubs](https://www.aol.com/articles/fear-unintentionally-going-viral-changing-220207000.html). She worries that a bad moment on a night out could end up online, where anybody checking up on her later might find it. Nobody has to be looking for her. Somebody only has to film the DJ and post it, and she is in the video.

## Thirteen seconds in a red bikini

It does not take a scandal. At a Summer Rockz party in Lloret de Mar, on the Spanish coast, this summer, a twenty-year-old French woman named Océane, who wanted to be a DJ, was dancing in a red bikini while water and foam hit the crowd. She says she did not know anybody was filming her. An Instagram account called Les Filles à Lloret posted the clip. It lasted thirteen seconds, and it passed [a hundred million views](https://www.ladbible.com/news/world-news/red-viral-instagram-video-spain-travel-732794-20260728).

The next day she [wrote to the strangers](https://www.instagram.com/p/DbLAQr6OXb3/) who had found her. The video had "completely changed my daily life," she said, and what made her happiest was that people were "simply discovering the real me": "a natural, spontaneous girl." Her biggest dream, she wrote, was to become a DJ. Soon after, as Aora, she [signed with a management agency](https://www.unilad.com/news/world-news/red-bikini-girl-viral-video-interview-259031-20260730) to help her get there. Within weeks she was performing the clip on purpose: to mark her hundred-thousandth Instagram follower, she put the red bikini back on and [recreated the moment](https://www.boredpanda.com/red-bikini-girl-breaks-the-internet-again-with-new-videos-in-the-same-swimsuit/) for the camera. The account that posted the original later added her as a co-author, so the video a stranger shot now sits on her own profile too.

Whether she saw that particular phone hardly matters. For someone who has grown up with cameras on every side, dancing the way people dance on camera feels like being yourself. That is how a stranger's thirteen seconds could look to her like the real Océane, more real than the afternoon she actually lived, and something she could step back into whenever a phone was ready. The party was over the same day. The clip kept going, and her career is now being built on it.

::video{src="https://www.instagram.com/reel/Da26BpZtWei/" title="The original clip, posted by Les Filles à Lloret on Instagram"}

## The work of a night out

People say phones have ruined the dance floor, then bring them anyway: [a 2025 survey found half agreeing and more than half admitting they use them there](https://musictech.com/news/music/survey-phones-ruining-dancefloor/).

So a night out now comes with work that nobody on Castlemorton Common had to do. It starts before anyone leaves the house: what to wear, in case somebody posts it. It carries on in the room: how to dance, and how much to let go, with phones held up on every side, and how to look, on somebody's video, as if nobody was watching. And it goes on the next morning, searching what strangers posted from the room to see whether you are in it, how you look, or whether anybody noticed you at all. If a clip of you does take off, there is a decision to make about that too: ignore it, ask for it to come down, answer the comments, or, like Océane, try to build something on it. In 1992 none of those questions existed.

## More real than real

That is what has changed about a night out. The crowd is no longer only the people in the room. It includes everybody who will watch the videos later, in group chats and on social media, and that crowd is far bigger. Going out used to mean the night itself was the only thing that counted. Now there are two versions of every night: the one people are standing in, and the one on their phones. More and more, the one on the phones is the one that matters. It lasts, it reaches more people, and it is the one that gets judged.

The French philosopher Jean Baudrillard had a word for a copy that people end up treating as more real than the real thing: a [simulacrum](https://plato.stanford.edu/entries/baudrillard/). The clip of Océane is one. Thirteen seconds shot by a stranger became, for millions of viewers and in the end for Océane herself, the real her. Baudrillard called a world where copies like that take over from the things they copied hyperreal: a world, in the words of one summary of his work, that is [more real than real](https://plato.stanford.edu/entries/baudrillard/), where the copies end up steering what people think and do. The phone version of a night out is heading that way. It is brighter and sharper than anybody's memory of the room, it has a far bigger audience, and it is the one people answer for.

The TV series *Black Mirror* put the nightmare in pastel colors in [an episode called “Nosedive”](https://www.youtube.com/watch?v=R32qWdOWrTo), where every small encounter is also a rating and one bad day in public follows a person into everything she tries to do next. Nobody on a real dance floor is being given stars out of five, but the fear is the same. Something a person does on a Friday night can be filmed, posted, and judged on Monday by people who were never in the room: a boss, a landlord, anybody with a search box. And everybody on the dance floor knows the phones are up.

::video{src="https://www.youtube.com/watch?v=R32qWdOWrTo" title="Black Mirror — Nosedive featurette | Netflix"}

## We've lost dancing

Kaptin Barrett, the DJ, put it simply: “The mystique of the club is gone. It’s just somewhere where you go and either you go to dance or you go to be seen.”

During the Covid lockdowns, with the clubs shut, the DJ Marea Stamper, who performs as The Blessed Madonna, talked with her friend, the producer Fred again.., about what had been taken away. He [built a record around her voice](https://www.thefader.com/2021/02/22/fred-again-the-blessed-madonna-marea-weve-lost-dancing) from that conversation and released it in 2021: “We've lost dancing.”

::video{src="https://www.youtube.com/watch?v=l4UkYBr1NnA" title="Fred again.. feat. The Blessed Madonna — Marea (We've Lost Dancing)"}

The clubs came back after the pandemic. The feeling of being unrecorded did not. It went a little at a time, one raised phone after another, with no single night anyone could point to. Freya Price is twenty-one. For as long as she has been old enough to go out, a club has meant being filmed, and being filmed has meant anything she did could end up online. So she goes to the pub.

Most people her age have had a phone in their pocket for as long as they can remember. Many of them have never been part of a crowd where the moment was the only thing that mattered: where nobody was filming for later, nobody was posting, and the night would live only in the memories of the people who were there. They have grown up with the phone version of every night, and that is the version they have learned to take seriously.

Look again at the photograph from Castlemorton. Seven people on a ridge in the sun, four of them with their hands in the air, not one of them looking at another. Somebody took their picture, and it made no difference to them. The picture was not the point. For a week in 1992, being there was the only thing that counted, and thirty thousand people were there together.