이 포스팅은 쿠팡 파트너스 활동의 일환으로, 이에 따른 일정액의 수수료를 제공받습니다.
It was Seokho who called, which was unusual.
Not that they didn’t talk — they talked often, the easy, unscheduled contact of two people who had known each other long enough that communication required no occasion. Texts about code. Occasional calls about the foundation, about the industry, about the specific landscape of Korean technology that they had both spent their careers shaping and that neither of them could entirely step back from even when they wanted to. But Seokho calling on a Thursday afternoon in November 2024 with no context except “naengmyeon, Saturday, the Myeongdong place” — that was unusual.
“Is something wrong?” Dojun asked.
“Nothing is wrong. I want naengmyeon and I want to eat it with you. Those are two separate facts that I’ve decided to combine into a Saturday.”
“That’s all?”
“That’s all. Not everything needs an agenda, Park.”
This was demonstrably untrue of Jang Seokho, who had never in his life done anything without an underlying logic, but Dojun had learned over nineteen years not to press before the meal. The meal was always the context. You couldn’t understand Seokho’s conversations without first eating with him.
“Saturday,” Dojun said. “Noon.”
“11:30. I have a board call at 3.”
“You always have a board call at 3.”
“I always have things to do. That’s not a complaint. It’s just what a life looks like when it’s full.” A pause. “Bring nothing. No laptops. No foundation business. No Aria updates.”
“Now I’m worried.”
“Don’t be. I’ll see you Saturday.”
The Myeongdong restaurant was the same restaurant.
Not “same” in the way that chains reproduced across the city — identical furniture, standardized menus, the corporate homogeneity of a franchise protecting its brand. Same in the specific, irreplaceable way of an institution that had survived long enough to become a landmark: the same family, three generations now, the grandmother who had run it when Dojun and Seokho first came here in 2009 having passed the kitchen to her daughter and the daughter to her son, the recipes intact, the method unchanged, the naengmyeon as cold and perfect as it had been when two university students sat across from each other and discovered they were worth paying attention to.
Seokho was already there when Dojun arrived at 11:28. He was sitting at the corner table — the same corner table, because Seokho had been coming here long enough that the table was implicitly his — and he was reading something on his phone with the concentrated expression he wore when reading technically complex material. He looked up when Dojun sat down.
“You’re early,” Dojun said.
“I’m always early. You’re usually on time. The combination produces harmony.” He set down the phone, screen-down, the gesture of a person declaring the phone portion of the day concluded. “You look well.”
“You look tired.”
“I am tired. Tired is my resting state.” He waved at the server — a young man who recognized them both and who had already, Dojun noticed, placed two water glasses on the table. “The naengmyeon will fix it. It always does.”
They ordered. The menu had not changed in fifteen years, and they had not changed their orders: mul-naengmyeon for Dojun, bibim-naengmyeon for Seokho. The preference inversion had been a point of mild argument in 2009 — Seokho believed that the cold broth version was inferior, Dojun believed the spicy mixed version was too assertive — and had by now solidified into one of the harmless, permanent disagreements that long friendships collected the way old buildings collected the marks of weather.
“Nova’s Q4 results,” Dojun said.
“I said no business.”
“You said no laptops and no foundation business. Nova is your business.”
“Fair.” Seokho straightened slightly — the CEO adjustment, the postural shift that happened when Nova entered a conversation. “Q4 was the best in company history. Revenue up twenty-three percent year over year. Enterprise clients up eighteen percent. The Southeast Asia expansion is running ahead of projection.” He paused. “Keiko says I don’t celebrate enough. She says I move from one milestone to the next without stopping to feel anything between them.”
“Is she right?”
“She’s always right. That’s why I married her.” He looked at the table. “I’m stepping back.”
Dojun was quiet.
“Not immediately,” Seokho said. “Eighteen months. We’re in succession planning now. The board has three internal candidates. I’m working with all three. By the end of next year, Nova will have a new CEO who is not me.”
“Why?”
The question was genuine — not skeptical, not concerned. Seokho had built Nova from a Daejeon apartment to a publicly listed company with a three-trillion-won market cap and twelve hundred employees across six countries. The decision to leave that was not nothing.
“Because,” Seokho said slowly, “I’ve been doing this for nineteen years and I’ve become the thing I was trying not to become.” He looked at Dojun directly — the eye contact that Seokho reserved for statements that mattered, that he had developed over years of presenting to investors and boards and that he now deployed only with people who could handle the unedited version. “You remember what you said to me in 2009? The first time we ate here?”
“I said a lot of things in 2009.”
“You asked me what I was building for. And I said ‘myself.’ And you said that was a fine start but a terrible finish.” Seokho picked up his water glass. “I’ve been thinking about that sentence for fifteen years. And I think I finally have an answer.”
The naengmyeon arrived. The server placed the bowls with the practiced efficiency of a restaurant that had been doing this for decades — the cold broth shimmering, the buckwheat noodles in their careful arrangement, the half-egg and the thin slices of radish placed with the slight precision of a kitchen that still cared about the visual. Seokho’s was a darker red, the gochujang working through the noodles.
Neither of them picked up their chopsticks immediately. This was a thing they had developed over the years — not a ritual, not a ceremony, but the accumulated habit of two people who had eaten together often enough to synchronize without trying.
“What’s the answer?” Dojun asked.
“I’ve been building for proof,” Seokho said. “Proof that I was right. Proof that the KAIST kid who everyone said was brilliant but impossible could build something that lasted. Proof that I was worth something without my parents here to see it.” He looked at the bowl. “Nova is the proof. It’s complete. And now I keep running the company past the point where I’m running it for proof, which means I’m running it for something else, and I’m not entirely sure what that is.”
“What do you want it to be?”
“That’s what I’m figuring out. Keiko has ideas. She wants to travel — not the investor roadshow travel, the actual kind, where you go places because they’re interesting and not because there’s a meeting at the end. Sora is six. She’s never had a father who was actually present rather than technically present.” He cut his noodles with the scissors — the same motion, nineteen years of the same motion, the naengmyeon scissors that he wielded with the automatic precision of habit. “I want to know what I like. Not what I’m good at. I’ve spent twenty years being good at things. I don’t actually know what I like.”
Dojun ate. The mul-naengmyeon was exactly as it had always been — the cold broth with its clean, slightly sweet depth, the noodles with their elastic resistance, the combination of temperatures that was the defining sensory experience of Korean summer even though it was November and the restaurant’s heating was running. Some foods were the wrong season and still right.
“I know what you like,” Dojun said.
“Do you.”
“You like hard problems. Not corporate hard problems — engineering hard problems. Problems where the constraint is physics and mathematics rather than budget and stakeholder management. You’ve been doing the second kind for twenty years because Nova needed it. But in 2009, in your KAIST lab, surrounded by the distributed systems papers you’d covered with annotations—” He paused. “That was the Seokho that the CEO version costs energy to contain.”
Seokho was quiet for a moment. Eating. The specific, Seokho-level focus of a person who was processing something that required the full machinery.
“I annotated a paper last week,” he said finally. “A paper on consensus algorithms. Byzantine fault tolerance in heterogeneous networks. I read it at 6 AM before my first call, and I annotated it, and when my EA came in and saw the printout covered in notes she looked at me like I’d started speaking in tongues.” He looked at Dojun. “I felt more like myself in those forty minutes than I have in months of board meetings.”
“Then that’s your answer.”
“Research? At forty?”
“Why not at forty? Tanenbaum published his best work in his fifties. Lamport was doing Paxos in his late forties. The age at which you do the best work is when you have enough experience to ask the right questions and enough stubbornness to ignore the people who tell you the questions are solved.”
“I’d need a university affiliation. A lab. Funding.”
“Kim Taesik will give you an affiliation in the same phone call it takes him to say ‘obviously.’ The foundation has a research arm now — Somin set it up last year. Funding is not your problem.”
Seokho cut more noodles. The motion was slower this time — thinking-pace, not eating-pace.
“You’ve thought about this,” he said.
“I’ve thought about you for nineteen years. This is not new.” Dojun set down his chopsticks for a moment. “Seokho. You built one of the most important technology companies in Korea. You listed it, you ran it well, you leave it in better shape than you found every piece of it. That’s complete. Now you get to do the thing you wanted to do before you had to prove you could.”
The restaurant was warm. Outside the window, Myeongdong moved in the November afternoon — the shoppers, the tourists, the particular Myeongdong energy of a street that never decided whether it was a market or a boulevard and had chosen to be both. The same street where they had eaten their first naengmyeon together, when they were twenty-two and twenty-three and carrying the specific, enormous ambition of people who had not yet had to deliver on it.
“We’re old,” Seokho said.
“We’re thirty-seven and thirty-eight.”
“That’s old for someone who peaked at twenty-five.”
“Neither of us peaked at twenty-five.”
“No.” He almost smiled — the Seokho almost-smile, the expression that on anyone else would have been a full smile but that on him was calibrated by the same precision he applied to everything. “You know what I’ve been thinking about, lately? The conversation in 2009. When you said ‘what are you building for’ and I said ‘myself.’ You already knew the answer wasn’t good enough. You’d already been through the version of success that doesn’t satisfy. You already knew what I was going to learn.”
“I’d lived it.”
“You’d lived both versions. The one where it works and the one where it doesn’t. And you came back to the second chance and you—” He stopped. Looked at Dojun the way he had looked at the Byzantine fault tolerance paper: with the full attention of a mind that was no longer performing but genuinely trying to understand. “Did it work? The second chance. Did it turn out the way you hoped?”
Dojun thought about Room 302. About Hajun and the concurrency bug. About his mother’s recipe notebook on the kitchen table. About Hana, who knew his full story and whose first response had been you’ve been carrying this alone for five years, you idiot. About the foundation, and Somin, and Lee Minseok who had wanted to build something he wasn’t embarrassed by.
“Not the way I hoped,” he said. “Better. The way I hoped was specific. The way it turned out was larger. I hoped to fix the things I’d broken. What I actually got was something I hadn’t known to want.”
“Which was?”
“The things I built the second time were better because I built them with people instead of for them. Aria isn’t mine — it’s Hana’s, and Minjae’s, and Jihye’s, and every engineer who came after. The foundation isn’t mine — it belongs to Somin and Kim Taesik and my mother’s board meetings and the three thousand students who are now out there building things. The code in Lighthouse—” He paused. “Even the code. The architecture I designed, but thirty-two engineers refined it into something better than what I had in my head. That’s the only kind of success that doesn’t feel like failure after the fact. The kind that happens in rooms you’re not the center of.”
Seokho was quiet for a long time. The naengmyeon was almost finished. The restaurant moved around them — the lunch crowd beginning to thin, the server refilling water without being asked.
“I want to be in rooms I’m not the center of,” Seokho said. “For a while. Just to see what that feels like.”
“It feels good. Strange at first. Then good.”
“You would know.”
“I would know.”
They ate the last of the noodles. The broth went cold as the meal concluded — the naengmyeon paradox, the dish that was cold by design but that became colder over the course of eating as the restaurant heat and the body warmth failed to compete with the bowl’s insistence on its own temperature.
“I want to ask you something,” Seokho said. “And I want you to answer honestly, not helpfully.”
“When have I ever answered helpfully?”
“Never. That’s why I trust you.” He set down his chopsticks. “Are you happy? Not ‘is the work good’ and not ‘are the people well.’ Are you — you, Park Dojun, whatever version of yourself is currently operating — happy?”
The question landed the way Seokho’s questions always landed: without warning, after being prepared for much longer than it took to ask.
Dojun considered it honestly.
He was thirty-seven years old in this body, ninety-seven in accumulated memory — the first sixty years of one life and thirty-seven of a second, the extraordinary mathematics of a man who had lived twice and who carried both lives in the same mind. He was the CTO of a company he’d helped build from a classroom project. He was a father, a husband, a son, a teacher. He was sitting in a restaurant in Myeongdong eating cold noodles with the person who had been his rival and was now the closest thing he had to a brother.
He thought about the Tuesday morning market visits and Hajun’s semaphore solution and the SNU students asking what they wanted to build and his mother’s recipe notebook and the room where it had started, empty and renovated and still somehow the same room.
“Yes,” he said. “I’m happy.”
“Good.” Seokho nodded. Once, the specific Seokho nod that meant: assessed, confirmed, filed. “I needed to know.”
“Why?”
“Because if you weren’t, everything I’ve been working toward — the thing you modeled for me, the second kind of success, the version that’s about the people rather than the proof — it would mean it doesn’t work. And I need it to work.” He picked up the water glass. “I’m about to bet the rest of my life on it working.”
Dojun understood. Seokho, who had spent nineteen years running his life like a proof, was about to abandon the proof structure and live experimentally. And he needed to know that the experiment had yielded results before he committed to running it himself.
“It works,” Dojun said.
“Even with the—” Seokho gestured, the gesture that meant the time travel, the impossible knowledge, the weight of living twice — the gesture they’d developed over years as a shorthand for the thing that couldn’t be said in a restaurant.
“Even with that. Maybe because of it. The first life showed me what not to do. The second life showed me what to do instead. But the doing was real — it wasn’t scripted, it wasn’t guaranteed, it failed in ways I didn’t predict and succeeded in ways I didn’t plan. The second chance wasn’t a cheat code. It was just a different starting position.” He paused. “You don’t need a different starting position. You started well. You just need to run the next phase differently.”
“The Byzantine fault tolerance paper,” Seokho said.
“Start there.”
“It’s an obscure area. Not commercially valuable. Nobody in the investment community will care.”
“Tanenbaum didn’t write for the investment community.”
“Tanenbaum is Tanenbaum.”
“And you’re you. Which is enough.” Dojun signaled for the bill — the motion they had traded back and forth over nineteen years, an unspoken accounting of who had paid last and who owed next, a ledger that had long since stopped mattering because neither of them kept score. “Call Kim Taesik. This week. Tell him you want a lab.”
“He’ll ask why.”
“Tell him you want to find out what you like.”
“He’ll think that’s vague.”
“He’ll think it’s the best possible reason. He’s been waiting for you to get here.”
The bill came. They split it — they had always split it, from the first meal, the specific refusal of either of them to be the one who paid, which had started as pride and had stayed as principle.
Outside, Myeongdong was doing its November thing — the Christmas lights already installed and running, the street a river of shoppers moving in both directions, the particular sensory density of a street that had been commercially alive for a century and intended to continue.
“Same time next year?” Seokho said.
“Same table.”
“Same noodles.”
“Different conversations.”
Seokho almost smiled again. “We’re always having different conversations. That’s why this works.” He put on his coat — the good wool coat, the one Keiko had bought him for their first anniversary and that he had worn every winter since because he saw no reason to replace things that functioned. “Dojun.”
“Mm.”
“Thank you for the second question.”
“I only asked one question today.”
“I meant the first one. Fifteen years ago. ‘What are you building for.'” He turned up his collar against the November air. “I have an answer now. It took me long enough.”
“What is it?”
Seokho looked at him. The full look — the one that the Nova Systems CEO rarely showed, that belonged to the KAIST student with the annotated papers and the distributed systems problems and the mind that was always three conversations ahead. The look of a person who had, after nineteen years, found the sentence that fit.
“I’m building for the people who’ll do it better than me,” he said. “Sora. The engineers at Nova who are smarter than I was at their age. The students in your foundation. The researchers who’ll read whatever I write about fault tolerance and use it as a platform for something I can’t imagine.” He paused. “You told me in 2006 that code that fails honestly is more valuable than code that succeeds silently. I think builders are the same. The ones who build in the open, who document their failures, who leave clear commit histories — they matter longer than the ones who build privately and protect the source.”
He turned and walked into Myeongdong — into the Christmas lights and the shoppers and the November Saturday that Seoul was fully committed to being. He didn’t look back, because Seokho never looked back, but he raised one hand briefly — the gesture that between them meant goodbye and see you and we’re still talking even when we’re not talking.
Dojun stood at the restaurant entrance for a moment.
The street moved. The lights were warm and excessive. Somewhere in the city, Hana was at the Aria office finishing the Series E paperwork, and somewhere in a school classroom, Hajun was in his last class of the day thinking about concurrency or distributed systems or whatever engineering problem had colonized his mind this week, and somewhere in a lab at SNU, three thousand foundation students were building things that would replace the things Dojun and Seokho had built and that would be replaced in turn by what came after.
He thought about the sentence Seokho had just said. Building for the people who’ll do it better than me. The thing that Dojun had been trying to explain, in various forms and metaphors, for nineteen years — and Seokho had arrived at it in his own words, in his own time, the way real understanding always arrived: not transferred but grown.
He pulled out his phone. Texted Hana.
Lunch was good. Seokho is stepping back from Nova. He’s going to do research.
The reply came in the time it took to put the phone away and take it out again: Of course he is. He’s been an engineer pretending to be a CEO for years. How does he seem?
Happy. Almost.
Same thing for Seokho.
I know.
He walked back through Myeongdong toward the subway. The Christmas lights reflected off the wet pavement — it had rained earlier, the November rain that was the city’s announcement that winter was arriving whether you were ready or not. The reflections turned the street into something doubled, the actual lights and their mirror images, both real and neither quite the same as the original.
He thought about fault tolerance. About systems designed to continue functioning when individual components failed. About the distributed architecture of a life — the nodes that held their own state, that communicated when they could, that together produced something more robust than any single component could sustain alone.
He and Seokho had been nodes in each other’s system for nineteen years. Not always in communication. Not always in agreement. But fault-tolerant — the way real systems were fault-tolerant, not by preventing failure but by designing for it. By building the relationships strong enough to survive the periods when they weren’t.
That was the architecture. That was what he’d been trying to build, in both lifetimes — not the code, not the company, not even the foundation. The architecture of a life that held together when parts of it failed. That distributed its load across people and relationships and institutions rather than concentrating everything in the single point of failure of one brilliant, isolated man.
He had the diagram in his head — the nodes and edges, the dependencies and redundancies, the graceful degradation that kept the whole system alive when individual components went down. He’d drawn versions of it on whiteboards in the Aria engineering lab, and the students in Room 302 were beginning to understand it, and somewhere in Myeongdong, Jang Seokho was walking toward a Byzantine fault tolerance paper and the beginning of what came next.
The subway entrance appeared. He descended.
Underground, in the particular Seoul subway silence of people in transit — the headphones, the phones, the collective agreement to be present and absent at once — he stood in the car and thought about cold noodles and warm restaurants and the specific, irreplaceable comfort of eating the same meal with the same person in the same place for nineteen years while becoming, continuously and without permission, someone new.
The train moved. The city moved above it.
He was going home.