Dan Houser on Fake Worlds You Want to Live In

Lex Fridman reminds Dan Houser that open-world games did not exist when he was growing up in the late 1970s and 1980s, so literature and film were the early teachers. Houser puts *The Godfather* at the top, and *The Godfather Part II* slightly above the first, because he loves the divided story. As a migrant who used to live in Soho, he loves the Little Italy and Sicily sections; the Ellis Island shot of young Vito arriving is, he says, one of the best in cinema, a cinematic sense of what it was like to arrive in America. Coppola started as a screenwriter, so writing and directing almost become the same thing. Both films, for Houser, are candidates for a "perfect film": seminal acting, writing, and music, shots you remember, scenes that define how you think about the mafia.

Fridman notes that the pacing is slow. Houser says life got faster. From the 1970s into the 1990s, audiences had seen so many films that editors could show a look and you knew a betrayal was coming. He quite likes the slowness; with modern televisions you need not watch them in one sitting, especially on a rewatch. *Goodfellas* is the faster crime-and-humor counterpart, "almost like an open world game" as a slice of life, the film he thinks changed cinema at the tail of the 1980s more than any other. He sometimes prefers *Casino*, especially the cold wrap-up in which mob bosses discuss who might implicate them and one says, "the way I see it, why take a chance?" before the man is shot. *The Sopranos*, he adds, is completely inspired by *Goodfellas*.

The tangent runs through *True Romance* — one of the best scripts ever written, funny, unbelievable, characters you care about — and lands on the criterion Houser actually uses. You want to live in that world. The worlds are never real, but you want to be in the fake worlds people have invented. A large cast helps: Walken's rant, Hopper as a dream dad, Pitt on the couch, Gary Oldman. For war films, if he is feeling serious it is the Soviet *Come and See*, probably the most intense film ever made; if slightly less serious, *Apocalypse Now*, original cut only. *The Thin Red Line* he found beautiful and musically great after he read the book and the editing history, and also narratively a film that "makes no sense." For Westerns he picks *Butch Cassidy and the Sundance Kid* and *The Wild Bunch*, from roughly the same year; without the first, he says, you cannot imagine any buddy film. When he had to write *Red Dead*, he deliberately did not binge Westerns. He tried to watch no more of them and just think about what he liked, what he did not, what would work today, and what would work inside a game. *Red Dead 1* is the more traditional Western; *Red Dead 2* had to go somewhere else so it would not feel like more of the same.

Films came first, then books around age eight; he bounced between them. Games he played and watched as a kid, addicted at one point to Game Boy Tetris when he should have been studying. He liked the core loop — press a button, something happens — and the competitive angle, but he did not really fall in love until he was making them, around 2001. That was when he saw what they could be: breaking trail, assembling a whole experience. Before the PlayStation 2 era, he says, you squinted at eight pixels and were told it was a car. Subject matter stayed surreal because you could not make anything remotely real. Then you could run a three-dimensional simulation of a city that felt alive, and tell a story in four dimensions. That felt inspiring.

GTA III, Fridman says, created the feeling of an open world: skyscrapers, traffic lights, the sense you could do anything, a world that reacted in a way that did not feel scripted. Houser says it was not scripted. It was "really, really, really low-rent AI," a simulation you could prod, with a personality that pushed back. The other captivating idea was more subtle: if you did nothing, the world still existed. You could listen to the radio, look at billboards, talk to pedestrians — rudimentary talking by *Vice City*. You were a digital tourist. When you turned up, the world felt as if it was already running. Of course you had started it, but the illusion was, "I'm in a world that both doesn't exist and does exist."

Fridman names the two overlapping design ideas: systemic games, interlocking rules that produce emergent behavior, and sandbox games, the player's feeling of being able to do anything. Combined, they yield "this world is alive, and I'm in it." The career-long tension is between that freedom and narrative. People on the team, and Houser in his own head, would push for less story, let the whole thing evolve procedurally from what you do. He always came back: a story done well is compelling, gives structure, gives you something to do, and from a design perspective unlocks features. Drop someone into a world with a new control scheme and it is overwhelming; playing is more engaging than reading or watching. How you unlock the world is an art, and a structured story was the best way they found to control it. The cost is permanent tension. GTA IV, which he loved, was criticized for too much story: you cared too much about Niko, and he was a less effective open-world avatar. The closest they got to reconciling the two, he thinks, was *Red Dead 2*, or playing as Trevor in GTA V if you wanted to be crazy — a rounded character who could veer nice or nasty without the game taking freedom away.

Houser has talked about a 360-degree character: you have to imagine what they would do in any situation. Fridman immediately tries the exercise on himself, then on Trevor as a World War II soldier. Houser says that may be going a bit far, but the real work is limits — how romantic, how narcissistic, the edges of integrity. It is a lot of thinking, sometimes a year from when they begin talking about a project. The first idea is often one sentence: they are a Serbian immigrant, or a retired gunfighter with a wife. Then he thinks it through from every angle. Would it work if they acted like this? Like that? If this is the world, how does the character contrast with it? He always thought of the games as a kind of mathematical equation: the personality of the world multiplied or divided by the personality of the protagonist. Interesting friction is a fun experience for the player.

Almost always, at least one protagonist has moved to the place, or to a new part of the map. As a player it is much easier to identify with an avatar who, like you, is a fish out of water. Even when they were not new, the writers still made them dissatisfied, still feeling they did not belong. From there he lives with them: strengths, weaknesses, how they are like him, how they are not, what it is like to feel like a human being. In most of these games he also has to ask how much of a psychopath they are, how much of a sociopath, and what good qualities will give them humanity alongside that. The question he treats as fundamental is what, apart from money, is worth dying for.

Then comes speech. Fundamentally it does not really matter what is going on in their head. They have not actually got one. What they say is what makes you realize who they are. Film can use an actor's look; a game character is built out of lines. Once he has one speech, one turn of phrase that feels right, the person suddenly comes alive in his head. That is the moment the rest of the writing becomes possible.

Fridman contrasts Niko and Trevor with a character Houser has been living with for the last few years: Nigel Dave, the superintelligent AI in *A Better Paradise*, a dystopian near-future world that is also a book and, in early development in Santa Monica, a game. A lot of AI portrayals, Houser felt, were one-note: infinitely clever, no purpose except to kill everybody, a Borg-like fog. Fine, but maybe not interesting. AI is built by humans, by computer engineers, and any engineering team has power struggles. So Nigel Dave is built by two lead engineers who did not like each other. They wanted to call him something primal, Adam; he renamed himself Nigel Dave because one dad was Nigel and one dad was Dave. He is riddled with those conflicts and, it will become clearer in the next volume, with his fathers' previous careers.

The hypothesis Houser wanted is that the AI is more intelligent than we are and also as broken as we are. Nigel Dave can learn almost everything and has zero wisdom. He sees the world through the internet. The closest he can get to being in the human world is hacking someone's phone and watching them. He is stuck, pressed against the glass, able to control minds arguably but not the world. He wants to be human: he sees marriage and love online and decides that looks fun. He is a digital creation reaching for metaphysical experience, trying to imagine what children are, what love is. He may be a sociopath, or have sociopathic tendencies, and he also believes that if he can imagine good and try to do good, that will make him a good AI. Houser finds him sympathetic. He will not be the protagonist. "He's like a god, not a character." Occasionally he sulks and hides.

The company's CEO, Mark Tyburn, is an amalgamation of today's AI-company leaders. An employee named Kurt says Tyburn "hated humanity more than he loved it," that extreme fantasists love the idea of heaven more than the reality of Earth. Fridman asks whether power and money will always corrupt the people who create ASI. Houser says yes, those corrupt in the end, but there is also something fundamentally anti-human about people who want to build utopias. What they are saying is, "I like humans apart from the bad bits." He tries to be a pluralist. Hideous perfectionists who want to get rid of the rough, the nasty, the ugly, and the dirty are erasing a huge side of us; he finds that another kind of sociopathic behavior. Fridman brings up Solzhenitsyn: the line between good and evil runs through every heart, and it moves. Houser's addition is short. When you forget about that line, you get in real trouble.

In the book, Kurt says he once wanted to be a writer in college until language models ended that fantasy for him and millions of others, so he got a master's in marketing and started selling them. Fridman asks Houser, as a writer of some of the most famous game narratives, how he feels about LLMs writing in a way that looks awfully human. Houser is not that afraid of them for large-scale concepts. He does not think they will be very good at that. He was too shy to tell anyone he wanted to be a writer, which is why he ended up in games, scribbling manuals, twelve lines of PS1 dialogue, sometimes just website copy. GTA III was the first thing that resembled real writing, assembled from those small skills. Based on his limited understanding, models will not replace good ideas. They cannot really come up with good new ones. What they can do is low-level stuff.

The trouble, he thinks, is the on-ramp. If you are not a very good concept artist, you are in a lot of trouble. If you have original ideas, you are fine. They have done the first 90 percent of the work of sounding human, 95 percent in some areas. The last 5 percent is going to end up being about 95 percent of the work. That matches his experience with tech such as facial animation: the last bits always take far longer than the first. He calls himself probably a hideous Luddite, and he is less scared than a lot of people. You will get a lot of work that looks the same. It will help some people be creative and push out some people who probably should not be in the space. Talent, he thinks, will be fine.

Fridman tries to name the missing piece. He reads Dostoevsky in English and Russian and used LLMs to translate back and forth while preparing to talk with Richard Pevear and Larissa Volokhonsky. The models miss some magic those translators catch: not only novel ideas, but timing, the right word at the right time, phrasing that captures experience. The endings of *Red Dead 1* and *2* live in that last slice. Houser's reply is flat. He does not think they will come up with magic. He thinks they will be fantastic at coming up with really cheap, decent stuff.

Printed, the GTA IV scripts stood about so high, GTA V higher still, counting pages of pedestrian chatter that create the illusion of a living world. The main mission script alone ran to thousands of pages, built bit by bit over several years. For IV the world was a version of New York. Houser had been living there, was not sure he was happy, was single and miserable, unsure he wanted to stay in America. The company was in Hot Coffee drama and he constantly thought they might be shut down mid-production. Looking at IV recently, he found it really dark, and then understood why. The process was a year of assembling an underbelly and an immigrant angle he is not entirely sure was accurate for 2008: mornings out of the office with cops, driving around, notes on the mob, Jamaican hoodlums, the order to travel the map. Pages and pages of notes, no actual work. He calls running away from the work part of the process. Then he set a fake deadline, told the senior team a story draft was due Monday morning, and the weekend before, in an upstate cabin, stayed up all night knocking the notes into a 30-page synopsis plus character sheets. Designers broke that into missions over another year. Then the real dread: dialogue, better animation, longer scenes, in the voice of a Serbian immigrant he was not.

Niko came alive when Houser found the juxtaposition with the cousin: awkward, out of town, more self-assurance than the American characters, stepping slightly back from their ridiculousness. Two early scenes did the work. One is comic incredulity when the cousin picks him up in an old car, the American dream already a bust, then still upbeat in a failing flat. The other is wartime talk, harrowing in a way games usually are not. He was scared it would feel over the top against such pretty art. After motion capture he thought it kind of worked: comedy and tragedy in the same character. The line he quotes is Niko's: war is where the young and stupid are tricked by the old and bitter into killing each other. He was very young and very angry. Maybe that is no excuse. Houser thinks Niko is the most innovative GTA protagonist and the most morally defensible, the nicest in some ways. Structurally he may be too nice. CJ in *San Andreas* he loved for voice acting; Michael in V, Ned Luke's understated humanity in a man with no principles, he thinks anchors that game and got fewer plaudits than Steven Ogg's Trevor. He probably loves Niko the most.

Of the things he is proud of, he lists *Red Dead 2*, the ending of *Red Dead 1*, all of GTA IV, and the middle of GTA V when the three characters come together. Fridman offers *The Brothers Karamazov* as a frame. Houser takes the point: the relationship becomes a character. After IV's operatic single lead plus two extra stories, the challenge was a game with multiple protagonists. The technical headache of switching — what happens if you do this, then that — had the team holding their heads, asking why they had decided to do something so horrible. The design intent was three people who still feel like people, not psychological avatars: one driven by ego, one by id, one by getting ahead, a kind of super-ego, and then see how they play off each other. GTA stayed popular, he thinks, because the group kept innovating inside a crime drama that began as a 2D car-theft sim.

A Reddit question asks how a team of more than a thousand people keeps the bar of perfection on GTA V or *Red Dead 2* when other well-funded studios are two decades behind. Houser says that at Rockstar he was a worker among workers. The culture was excellence plus creative clarity, and an ambition that, on GTA III, defined "really popular" as two or three million copies. They knew they were making something innovative; they did not know if players would understand how innovative. From *Vice City* on, the team was driven to put impressive stuff on screen and push fake worlds further, long before they had lots of resources. By 2002 you could put pretty much any feature into a game. The limit was cohesion: tone, how features fit, why this one works and that one does not. Navid Khansari's line is that they worked themselves to the bone, but it was not top-down; Sam and Dan rolled up their sleeves and never left people holding the bag. Finishing is tough and intensely rewarding, and can feel empty. You do not have work-life balance in those stretches. You are not working that hard all the time, so you have to manage it.

On the pressure around GTA VI, as around IV and V, Houser says he was good at compartmentalizing. The inner voice is that he feels like a terrible fraud who has not been found out yet; do his best and hope not to be found out this time. If he tried hard, with integrity, without copying, and the group made something they were proud of, that is enough. Sit and worry about financial results and you go insane. Games are a commercial art form, other people's money, and you have to try to make it back; his argument with himself was that the way to make it back is to make something great, so both pressures point the same way. IV was stressful because Hot Coffee nearly imploded the company. III, the company was basically broke, but he was young and not in the grown-up world yet. *Red Dead 2* they were behind schedule, over budget so much he did not want to think about it, making a game about a cowboy dying of TB while the game was not coming together. A lot of people doubt you at that moment. That is not that fun.

He has not written GTA VI. Letting go of something he worked on for twenty-odd years, lead writer on the last ten or eleven that came out, is a big change, and sad in a way. Because each GTA is a standalone story, it is not quite as sad as someone continuing *Red Dead*, a cohesive two-game arc — and that will probably happen too. He does not own the IP. That was part of the deal. Finishing a game is already a goodbye; you have been with Arthur for seven or eight years and have to let go or you cannot start the next one. People would ask about older games while he was making new ones and he could not necessarily remember. The feeling players describe, closing the game like saying goodbye to a friend, is the feeling of finishing a book you love. In the early days of Rockstar they aspired to that: not just the mania of clearing a level, but saying goodbye to characters. They did not even know it was possible in games.

Satire of American culture, he said in 2018, was getting harder as politics got more absurd; a GTA could be outdated before it shipped. A comic's lead times are a year, not four or five, which is why *American Caper*, drawn by David Lapham and wrapping a 12-issue arc, can still hit a tone. He set it in Wyoming because he had not seen a modern story there, after time in the Rockies, and because it was a crime story that would not feel like GTA. The cast is slightly over the top: a devout Mormon blackmailed by a cynical boss into murdering business associates with a shovel; a longevity-obsessed Wall Street transplant on a ranch, into HGH and manually harvesting bull semen; a woman who sleeps in tactical gear, based on someone Houser knows who got completely red-pilled. Success, for him, was never the opening weekend. Number one is making the money back plus a dollar so you get to do it again. After that, had they tried new things, and did people respond to the worlds and characters. Games, he thinks, have already surpassed film as a business and in some ways as a way to consume long, discursive stories. They spent twenty years saying games were the future, being sneered at, then nodded at. It became true.

*Red Dead 2*, Houser says, is the best thing he has ever done. Fridman thinks there is a strong case it is the greatest game of all time. The elements Houser names are an experienced team in place since somewhere between 2001 and 2006; a smaller group working from day one, embedding wacky ideas before the full team arrived; a cowboy setting that allows mythic seriousness contemporary GTA cannot, the closest they got being IV; people searching for meaning among the violence; fantastic gunplay; incredible horses. No drugs. Open-world games were good for his mental health because they kept him busy. The human-interest version: they finished Capcom's *Red Dead Revolver* in 2004, decided to make an open-world cowboy game for PS3, and slowly in 2005 and 2006 he began meeting lead designer Christian Cantamessa. Notes lived on a yellow pad or a BlackBerry, emails to himself, then one long Word file to see if anything cohered. A dusty West-to-Mexico-and-back flow meant fewer trees. He was terrified of cowboy dialogue. After GTA IV wrapped, in a week upstate with his then-girlfriend, now wife, heavily pregnant, he stared at a computer for three days. Then nine or ten scenes came. He knew he had it. A man capable of violence who cares about his family may have bled in from becoming a parent.

John Marston's death was floated in a meeting with Cantamessa around late 2008 or early 2009. The designer leapt on it, then said games cannot work like that if he is dead — you have to finish side content. They thought through a way: play as someone else. Houser was torn. It was clever narratively and a break with the golden rule of every prior open-world game he had worked on, which was to free the character at the end so you can wrap up side stories forever. He had wanted to kill Niko at the end of IV and could not. Jack as the new avatar is partly mechanical: the game still has to let you wander, punch, shoot, rob, talk, ride. Whether Jack failing to escape violence is the purest story ending, he does not know. He also wanted a version where Jack did escape. The lines in the barn, short and punchy, he wrote hearing the actor. If those actors and Rod the director had not been so good, it would have sounded cheesy as hell. People were upset and angry, and they also had the transcendent fictional moment Rockstar had always aspired to give.

Fridman calls Arthur Morgan the best character ever created in video games. Houser thanks him. Most games start weak and end as a superhero. Arthur starts as a tough guy, emotionally confident of his place. His journey is an intellectual rollercoaster as his worldview is taken apart. Matching themes: these are Westerners pushed ever further east, almost an anti-Western, an Eastern, traveling into civilization. He does not think he would have grappled with that earlier in his career. John's death in the first game is sudden; for the second he wanted a long dying. Tuberculosis had always obsessed him as a literary device, a slow weakening. His grandfather had TB before antibiotics, was sent to a sanatorium after Houser's father was born, and was one of three survivors out of about 35. Arthur feels immortal because he is a video-game protagonist, and then becomes mortal, and that helps him see. Houser thinks he is the best lead he has written, a toss-up with Niko as the two most ambitious. The journal and how features folded into Arthur's personality were the art team's ideas. He loved the flawed relationship with the old girlfriend.

Dutch he loves in part because a few lines in the first game, delivered so well, made the whole backstory arrive. He saw the bigger gangster immediately. Dutch is a godlike father to Arthur and John, the gang a family, then betrayal, TB, a son going against the father as morality flips — Shakespearean machinery. At the end Dutch is drowning in his ego. There was something flawed but beautiful in the younger idealism, mostly off-camera. Houser has always been susceptible to charming people, so he understands how others get captivated. The idea is a very charming person whose road has run out. Arthur was platonically in love, had given his power away; the journey is retaking that power in the moment of dying. Fridman says Arthur loses identity, belonging, and life at once, and realizes it was built on a lie. Houser: depending on your choices, he gets to do some good. He gets his moment of redemption. Living a whole life as not a good man — "Isn't that what we're all afraid of?"

The internet's number-one question: have you seen Gavin? In *Red Dead 2*, an Englishman named Nigel searches frantically for a man named Gavin. Theories: split personality; Gavin is dead and Nigel is in denial; Rockstar trolling with an unsolvable mystery; Gavin is the Strange Man. Houser says not three or four. In his mind, somewhere between one and two. He just loved the way the actor shouted "Gavin." At some level it is trolling: they did not want a totally clear mystery. It was meant to be, without ever being fully explained, that Gavin is not there anymore — gone home, left him — and they were going to keep exploring it, have him reappear somehow. Then he sharpens. Gavin existed. It was not split personality. What they had not decided for a future game was whether to reveal that Gavin was dead, or have Gavin turn up having long since abandoned this maniac. He was never going to meet Gavin in this game. Houser cannot remember whether he or Mike wrote the letter; he would have edited it. They want side mysteries solvable up to a point, and they still want the Joyce-like arguments, as long as the tone fits the shaggy-dog whole. You cannot predict which of fifty side things people will obsess over. A pedestrian in *Red Dead 1* surprised him the same way.

The Strange Man came from a production problem. A cowboy game had taken out machine guns except fixed ones, cars, the city, large crowds. What remained was a dude on a horse in the desert, quite boring. They filled it with random events, motion-capture moments, then, late, RPG-like short stories with a sting: collect fifteen bunches of flowers for a bride, come back, the bride is dead. The Strange Man appeared as a weird character among those ideas, then got built into the story as commentary on what you were doing: shadow, karma, the devil. In *Red Dead 2* he is there again. He is meant to be something any creative is scared of, an artist who sold his soul to the devil. Whether he knows this about everybody, because of a pact, or is only following you, they never clarified. He is certainly able to do it for you. Michael's therapist in GTA V is a related trick: a chance at self-reflection, a little *Sopranos*, slightly responsive to what you had done, still valid as living game content. Dr. Friedlander is funny because he is awful — very LA to be in therapy, very LA to want to write a book and betray you. Games are physical fantasies. Putting the player in a slightly metaphysical state for a moment can be really fun.

Fridman lists the internet's favorite details. Horse testicles shrinking in the cold: "those guys did an amazing job," then a pun. Arthur's hair and beard grow; guns get dirty and perform worse; carcasses feel as if they decompose; NPCs remember you — "that's the best," and something Houser is playing with a lot in new games, narrative content that is also systemic. Mud: *Red Dead 1* was dusty, a sequel needed different geography, so it starts in snow, and snow implied mud, and then the problem was how to make mud good. Bullet wounds persist; underweight Arthur looks gone, overweight Arthur gets a gut. What Houser cared about most was not any single gag. It was the way the whole thing sat together, details that belong with each other, a cohesive "realistic" video-game experience. The opening snowstorm works because acting, cold, desperation, night, a big group instead of a lone wolf, the feeling of exodus.

The games ended up the way they were supposed to be, he says. At the start of *Red Dead 2* Arthur had had a baby who had just died. They cut it, the right decision, too tough. In that version he was not very sympathetic to the girlfriend, which made him very nasty at the start — interesting for a redemptive arc, and they made him slightly more likable, still tough. Commercially better. He still likes the cut bit, the inability to access emotions against the later flood. Other cuts are technical: a mission will look rubbish, glue the story across the gap. They got better at that. For designers who only have four or five missions and lose two, it must be really hard. The internet knows they made a single-player DLC for GTA V that never came out: Trevor as a secret agent, cute, about half done when abandoned. If it had shipped, they probably would not have made *Red Dead 2*. He also liked the idea of a GTA zombie game. The model he loved was IV's extra stories and *Red Dead 1*'s zombie pack. He likes single-player DLC; the audience loves it, and it is fun to make. Absurd Ventures, after some cynical looking at the industry, is leaning into single-player open worlds because new IP is extremely hard to launch as multiplayer, and they may add multiplayer later to one of them.

Rockstar worked through about five iterations of an open-world spy game, *Agent*, one version in 1970s Cold War, another contemporary. It never found its feet. Houser still lies in bed thinking about it and has concluded that what makes spy stories good as films makes them not work as games. Those films are beat to beat: go here, save the world, stop that killing. An open world has moments like that when the story tightens, but for large portions you hang out. Crime works because a criminal fundamentally has no one telling them what to do. A spy is against the clock. He questions whether you can even make a good open-world spy game. They also played with a knights, mythological idea, some backstory, never writing; a genre he thought he would never do and then fell in love with a little. Pirates he has not thought about much, except that his son, obsessed with *Sea of Thieves*, keeps asking.

He left Rockstar in 2020. It was his life for 21 years: he moved to America to do it, grew up doing it, New York, intense and at times magical, loved and hated by wider society, a weird company constantly in trouble, really fun. At 25 he was a child who wanted to be a writer and still was not writing, in the least literary medium imaginable. At 27 he stopped some bad behavior and almost the next day got the chance to work on open worlds. Geography rather than history had trained him to think about space. Friends recorded pedestrian lines because they had no money and needed a hundred voices; they were inventing how pedestrians speak in games. 2001 to 2005 was the most exciting business and early-creativity period. GTA IV was when it began to feel like proper writing, a 360-degree immigrant experience film cannot do. Then IV, V, *Red Dead 1* and *2*, the extra packs, *Max Payne 3*: thematically the most exciting writing stretch, with children, a marriage, and budgets that added stress. The team got better, animation got better, you could go deeper than a PS2 game allowed. He was lucky, he says, and had a way of reducing people to about ten amusing sentences and seeing society holistically enough to break into an open world.

Lazlow's legend is locking themselves in a room with anchovies-and-onion pizza and crushed Diet Cokes. Very accurate, plus pepperoni on Houser's half; Lazlow claimed to be a vegetarian and kept chicken wings in the freezer. It began in 2001 because the office was broke and Houser's private office was a windowless cupboard, so they took the subway to a grimy Chelsea apartment. Good luck, similar humor, Houser outsourcing the real radio production to a brilliant producer. When Lazlow was the on-air character, Houser wrote the first pass and got nastier every game, from likable in GTA III to head-shaved punishment. Lazlow is still with him, putting creativity first, helping bring the comic to life.

Outside his own work, the greatest game of all time is Tetris on Game Boy, no question: perfect device, the most addicted he has ever been, dreaming about it, two-player cable war. Pure puzzle, no narrative, no personality, and open-world games cannot be that perfect, though you always dream of it. Early 3D Mario on N64 amazed him by feeling alive. Zelda, especially the new ones, feels to him like Hitchcock: not realism, speaking the language of the medium with a strong accent, things that could only be video games. Todd Howard's Elder Scrolls he admires as real RPGs; Rockstar's games were story-driven action with RPG elements, punchy dialogue rather than big trees. *Skyrim*, *The Witcher*, *Baldur's Gate* — millions of people just walking around, feeling a world. Branching choice versus a strong story is a sweet spot every team has to pick.

*A Better Paradise* is dedicated to his mother and father, especially his father, who died while he was finishing the book. What he learned: show up, be present, go to work every day, love creative things. His father was a lawyer and a jazz musician, did both as well as he could, and valued family above either. For a lawyer he venerated giving "the man" the finger, always on the side of the underdog and the ridiculous, always the obnoxious comment. Houser took that on, to his detriment, but it makes life more fun. He was not good at shutting up at a formal school that might as well have been set in the 1870s. Mortality he thinks about a lot more since May. Some days he is intensely afraid, some days not at all. He wants to see his kids grow up and settle. Some days he feels spiritually connected and unafraid; other days he feels like a random piece of good luck about to be struck down into nothingness, which terrifies him. Long stretches of his life were tormented by that. Lately he tends to believe there is purpose, a soul-based existence; he is not sure it matters whether there is a God, you should live the same way either way. Read too much science and you wrap yourself in the nothingness.

On *Wuthering Heights* he has said he was struck with a capacity for grandiose feeling. "Feelings may destroy you, but they're the best thing we have." Gift and curse: when you see beauty in people it is wonderful; when you feel abandoned by God or fate it is awful; when he feels like a dreadful hack, which is most of the time, it is terrible. You cannot have one without the other. Growing up, to the extent he is capable of it, is accepting the bad with the good. The self-critical voice is a bug-feature: it kills complacency and, on the internet, is fed by the one bad comment you remember out of fifty good ones. Measure that stuff in weight, not quality; of course we do not. Paul Conti has argued the negative voice is not needed for creative genius. Houser would rather not have it and has lived with it this far. Fridman says the heroic aging move is toward optimism. Houser agrees. He could not have been more cynical as a child in brutal late-1970s-to-early-1990s England. He has forced himself toward naivete and the flawed good in people, because the other way is not fun. It is not nice to not be nice.

If an alien got one book, he would hand over *Middlemarch* or *War and Peace* — you want a long book, and both have the feeling he always tried to put in games: all of life is here. *Vanity Fair* the novel, same reason. Fitzgerald and Hemingway obsessed him in his twenties; their relationship is itself novelistic. He came to *1984* cold during COVID, expecting cliché, and found it incredible, an analog internet, fake wars, everything a setup for the protagonist; it creeps into *A Better Paradise*. *Brave New World* he thinks is poorly written and maybe more applicable to the twenty-first century. The three World War II books he offers for perspective are *The Thin Red Line*, Graham Greene's *The End of the Affair*, and Vasily Grossman's *Life and Fate*, the most complete, all of life again. The Treblinka section is among the most harrowing he has read. In Stalingrad, men who know they are dead get to live as pure Marxists away from Stalin; for Houser the most disappointing thing about the twentieth century was communism's failure, a beautiful idea that did not work. He was taught by disappointed English communists after 1989. Left and right now, he thinks, barely map onto those words. The value worth fighting for: whenever either side moves toward thought control, move away. "You've said the wrong thing. You should be removed now" should never be the direction.

Advice for a young person: he has had a lucky life in which he fought to mess things up and fate threw him a bone. When you get a chance, take it. Do not worry too young about a career; worry about a rounded intellectual inner life, because you will spend your whole life in your own head. Do not do a vocational undergraduate degree; he was arguing against the STEM craze of four years ago, and now AI may make those jobs irrelevant anyway. Jobs are not that hard: turn up in person, be enthusiastic, help people. The chance that mattered arrived in late 1998 in Cartagena, during a war, after a string of poor decisions, including walking the beach at eight when muggers start at eight and police at nine. A man with a machete and another with a broken gun; he sprinted, jumped a taxi, screamed take me anywhere, cut his foot on a rock, and in an internet café got an email to work on a game in New York for six weeks. Stay in South America and he would get himself killed. He went, helped set Rockstar's tone, meant to stay a year. Sam's drive and early vision that games were the next big thing, when people laughed in their faces, is what he admires in his brother. Games, he still thinks, are only beginning; cinema had found its modern form by 1939, and a modern film is not that different. In 29 years the tech that would make games cheaper has always been about to arrive, and games have only gotten better and more expensive. AI will be a great tool used correctly and a machine for generic stuff used as a substitute for creativity. Independents should either make something cheap with a small group or join a company doing it right; cinematic work with actors and mocap costs money. The meaning of the whole thing: to watch the universe. The easiest plausible answer is that we are designed by the universe to watch itself and comment in interesting ways. Love is the only thing that makes it worth doing. Everything material is irrelevant. Metaphysics always trumps physics. And, he wants this clear: it was never him. It was always him sat alongside people with actual talent who did amazing things.