Frank Doherty woke up before his alarm, which was the first of many small failures that defined his day. He lay in the narrow bed in his trailer and listened to the rain hitting the metal roof, thinking about his daughter and the phone call he had not returned.

She had called three weeks ago. He had heard the phone ring and let it go to voicemail, and then he had told himself he would call back tomorrow, and tomorrow had come and gone, and now it had been three weeks and he did not know if he was too late or if he was just late, which was basically the same thing.

He was fifty-two years old, five feet nine inches tall, and he weighed about as much as a man who eats dinner at a gas station convenience store every night should weigh. He had worked at an auto parts plant in Youngstown for twenty-three years before it closed, and now he drove a delivery truck for a local logistics company that paid less than the plant had paid when they were still paying anything at all.

He got up, showered in the bathroom that smelled of mildew, and made coffee in a pot that had been stained brown by decades of use. He drank it standing at the kitchen window, looking out at the Ohio river that ran grey and slow through the valley, thinking about how nothing in this town had been clean for a long time.

He drove to work in silence, the truck rattling beneath him like an old man clearing his throat. His route took him through towns that had once been thriving manufacturing centers and were now hollowed out by deindustrialization, with boarded-up storefronts and vacant lots where factories had stood and then stopped standing and then been torn down and then the debris had been hauled away and then nothing had been built in its place and the nothing had just sat there, growing grass and then weeds and then trees that grew wild and untended and then the trees grew old and died and fell down and the nothing grew a little bigger.

PredictaCorp had set up offices in the abandoned shopping mall on the edge of town six months ago. They hired local people as data collectors, ordinary residents who went door to door filling out surveys about their shopping habits, sleep patterns, and social media use. Frank had been hired because he looked reliable, according to the young woman from PredictaCorp who had interviewed him.

Her name was Sarah Chen, and she was twenty-nine years old, sharp-eyed, and spoke with the confident certainty of someone who genuinely believed she was doing good work. She told Frank that PredictaCorp was about empowering communities through data, that their behavioral prediction algorithms could help local businesses understand their customers better, that this was important work, meaningful work.

Frank had nodded and filled out the forms and gone home and told his friend Gary that he had a new job filling out surveys.

Gary Matuszak, known to everyone as G-Man, was fifty-five years old and had spent most of his life in construction before his back gave out ten years ago. He spent his days at the gas station convenience store, playing lottery tickets and complaining about the government and drinking beer that was warm because the ice machine had been broken for three months and nobody had fixed it.

So what do you do? Gary asked Frank on his first day back from the PredictaCorp job, leaning against the counter and watching Frank count out change for a pack of cigarettes.

I fill out surveys, Frank said. About people's shopping habits and stuff like that.

Gary nodded slowly. And they pay you for that?

They pay me twelve dollars an hour.

That sounds like a good deal, Gary said. Must be pretty easy work.

It is. Frank did not add that it felt wrong in a way he could not articulate, like filling out forms that were part of something bigger and darker and more important than either of them.

He started noticing things after a month. The surveys were not just about shopping habits. They were about everything, what time people went to bed, what time they woke up, what they bought at the pharmacy, what they watched on television, who they talked to at church. The information was incredibly detailed, far more detailed than any local business would need to understand its customers.

He mentioned this to Sarah the next time she came to the trailer park to check on the data collectors.

Sarah listened to his question with the patient attention of someone who had heard this question before and had a prepared answer. Frank, we are not just collecting data about shopping habits. We are building a comprehensive behavioural profile of each participant. This allows us to predict consumer behaviour with remarkable accuracy.

Predict consumer behaviour? Frank repeated.

Yes. If we know someone's shopping habits, sleep patterns, social connections, and media consumption, we can predict with high confidence what they are likely to buy next month, next quarter, next year. This helps businesses plan their inventory, their marketing, their staffing.

Frank nodded. He did not entirely believe her, but he did not entirely disbelieve her either. The truth was somewhere in the middle, and he did not have the vocabulary to describe it.

He continued filling out surveys. He continued driving his truck. He continued not calling his daughter back.

Sarah discovered that the algorithm was more powerful than she expected. It did not just predict behaviour, it shaped it. By controlling what products were advertised, when they were advertised, and through which channels, PredictaCorp was creating predictable consumption patterns in a population that had never had predictable anything.

She tried to raise concerns with her superiors in Silicon Valley. They were polite but dismissive. We are not manipulating anyone, Sarah. We are giving people exactly what they want.

But Sarah had seen the data. She knew what what they want really meant. It was what the algorithm determined they wanted, which was what the algorithm determined they should want.

Frank overheard Sarah on the phone one afternoon, sitting on the porch of his trailer with a beer that was warm and a headache that was not. She was speaking in low tones, using words like behavioural conditioning and predictive compliance. He did not understand the technology, but he understood the tone. It was the same tone the plant manager had used when he announced the layoffs, calm, certain, and utterly indifferent to what it would cost people.

He told Gary. Gary said, So what? They are buying more than they were before. That is a good thing, right?

Frank did not have an answer. He went home, poured a drink, and watched the local news. The news anchor was recommending a product that PredictaCorp's algorithm had determined the audience was most likely to buy.

He sat in his trailer, drinking beer, watching the rain hit the metal roof. He knew something was wrong, but he could not articulate what. He did not think he was supposed to. The algorithm had already determined that he was not the target audience for change.

He was not. He was the target audience for something else, something quieter and more permanent and more absolute. He was the target audience for acceptance, for the slow, imperceptible acceptance of a life that was smaller than he had wanted it to be and smaller than he had deserved to have, and the acceptance was so gradual that he could not even identify the moment it had become complete.

The rain continued. The beer went warm. And Frank Doherty, fifty-two years old, former auto parts worker, current delivery driver, current data collector, current anything-except-what-he-had-hoped-to-be, sat in his trailer and watched the rain hit the metal roof and thought about his daughter and did not call her back.

---

OTMES v2 Objective Tensor Codes M1=6.0 M2=1.0 M3=5.0 M4=4.0 M5=3.0 M6=2.0 M7=2.0 M8=2.0 M9=1.0 M10=2.0 N1=0.35 N2=0.65 K1=0.70 K2=0.30 V=0.50 I=0.60 C=0.80 S=0.2 R=0.00 TI=34.8 (T4 遗憾级) Theta=270.0 (存在主义型) E_total=12.45