Mike Backes writes screenplays and makes Rocket Science videogames, so he should know what he's talking about when he tells Ron Martinez that Hollywood is going to eat Silicon Valley's lunch.
If anyone has the creative resources to swim the convergence currents, it's Michael Backes, both Hollywood screenwriter (Congo, Rising Sun) and Silicon Valley game developer (co-founder of games start-up Rocket Science). But don't expect the hackneyed hype about Siliwood from Backes (at left below). As far as he's concerned, the valley has some growing up to do. The outspoken digital savant sat down with Ron Martinez, former VP of business and creative development at Spectrum HoloByte, to ponder questions like "Can a synthetic character ever get past the Turing test limitation?" and "Where are the Eisensteins of the interactive medium?"
Martinez: Former Apple evangelist and technology executive Guy Kawasaki recently wrote a column in Macworld about doing business with Hollywood. Listen to this: "The biggest barrier to closing a deal with these folks is understanding their culture and values. I use the terms culture and values loosely." Here's more: "Be greedy. There are two reasons. First, whomever you're dealing with will be. And second, Hollywood may have distribution marketing, but you have technology. Let's face it. Their idea of user interface is VCR+." As a representative of the Hollywood crossover community, what do you think of this?
Backes:
The same thing can be said about a lot of people in Mountain View, Cupertino, or Sunnyvale. You meet these marketing bozos and these 32-year-old Presidents of All Media. Guy's a really smart guy, but he doesn't have a clue about Hollywood. Just like Hollywood doesn't have a clue about Silicon Valley. They speak two different languages, and there are few translators. Sometimes I think it's easier to deal with a priest on the Pentecost Islands than with some of these people who want to shuttle back and forth between Hollywood and Silicon Valley. They're clueless. They stare at each other across the table, and they don't even know what to trade. At least Peter Minuit, the first director of New Netherlands, could show some trinkets and get Manhattan.
Well, there's no common vocabulary. From the fundamental imagining of the entertainment software product - at the point of conception - there are radically different mental maps. Is there really a convergence between Hollywood and Silicon Valley? Or is that bogus? Is there instead merely cross appropriation, as opposed to convergence?
As long as hardware weenies continue to think they have any control over the content business, there'll be people like us here ready to exploit them at every opportunity.
But isn't there something they know that you don't? About the way the software works, or what constitutes a good game, or what constitutes interactive entertainment?
The bottom line is they have no clue about user interface. I don't think there's any user interface genius in Silicon Valley. If there were, you'd see it in the software. OK? And you don't. They're still stealing ideas out of mid-1970 Xerox PARC. Show me some spectacular interface breakthrough that's been made in the '80s or '90s by anybody. It's a new industry and a growing industry, and a lot of people have been spectacularly successful and made a lot of money in this industry, so they think they bring something spectacular to the table. I think, as a rule, they don't. Computers are still way too hard to use, way too complex, and videogame machines don't have any front end to 'em.
But what about the numbers? Computer games, or videogames, out-gross the Hollywood box office.
Except look at all the ancillary stuff. The bottom line is that Jurassic Park toys sold a billion bucks worth. The movie business creates content that bleeds all over the place. We create industries. Now, I think movies can cross over to games, provided that the environment of the movie is compelling enough to make an interesting environment for a game. However, because most games don't have any characters worth a damn, you got a real tough time bringing any kind of dramatic interaction over from the games, because - guess what?
There's no dramatic content to begin with.
No, there isn't. Super Mario World is just a place.
It's a set, but there's no play. But the notion of intelligent characters requires a different aesthetic; think of nonlinear, user-controlled or user-defined characters. The whole range of aesthetic potentials is really interesting.
Absolutely. Profound. The real future for some of this stuff lies in having people trying on different faces. Masquerading electronically and trying on different characters. But you've got to define a journey through the electronic environment that's dramatic, and most people take a kind of clichéd approach.
In computer games, yes; but that happens in movies, too.
Absolutely. But you could take Street Fighter and make a really kick-ass movie out of it.
How would you do that?
I'd do it just like they did. What was the movie with Charles Bronson and James Coburn? Hard Times? Where they did bare-knuckle boxing? That's not a bad picture. The idea of a world competition to find out who's the most badass. It's like a hopped-up Kung Fu episode. It's hard to make good movies. That's why there are so few.
Right. That raises a couple of issues. When I showed up in Northern California from New York, I visited a lot of games companies, and I noticed that a product-manufacturing mentality had pervaded the industry. It was about getting a group of people in a room and making products with features that you could bullet. But it's also pretty clear that players now have increased expectations that games not only look great, but are also emotionally evocative, cool, compelling, and addictive. And the talents needed to make that happen are very rare. We are moving to a talent-driven industry model, and if you want to know what these business mechanisms are, look at the movie business.
The game business is still like early film: the guys who operated cameras were the movie directors. That isn't the case anymore. The Great Train Robbery of the interactive medium has yet to be created.
I think you're right. And so you move to a production model because that suits the rhythms of capturing talent when it's available and packaging it together to make something happen.
In the past, if you were the game designer, you were the boss. As dramatic interaction gets more complex, is the game designer necessarily the correct person to be the creative head of the ship?
Those were the first principles. What constitutes game design? What is a game?
I still feel that a game is an environment. That's really what I feel first. It's a piece of the world in which a player interacts. That's why I'm heartened by these 3-D environments you can move around in.
Real-time 3-D?
Absolutely. It's really neat. Even if it's not real 3-D, it's doable. Doom at least makes you feel a sense of impending dread, because you don't know when something more powerful than you is going to step around a corner and blast you to smithereens.
This gets you into story. You're an accomplished screenwriter. It's amazing to people not in the business that a screenplay can sell for US$3 million. Up north, I've been in situations working with a game designer and I say, Well, we really need a story to frame this, so we can inform and organize all the elements. Who the characters are and why we care about this world, and they say, No! Spec out the game first. The story you put in a manual. So the question is, When you go about creating a story, how can you create something that may have emotional power and economic power in the marketplace? And how is it structured? And how could those structures be mapped onto or integrated into games and interactive entertainment?
Good screenwriting is really about a single directed point of view. You start messing with that point of view when you allow it to be diffused by offering choices of how that point of view can be ignored, or subjugated, or diluted, by giving it to a player. A good example is the T-rex sequence of Jurassic Park. It's the high point of that film: it's supremely visualized, but from a specific, highly manipulative point of view that just takes you where one person wants to take you through this scene. In other words, they are grabbing you by the head and saying, Look here, look there, look here, look there, and the emotional response you have is one of unmitigated terror and utter fascination. The problem with a user-defined experience is that you can stare at your shoelaces rather than at the T-rex, and if you're worried about loading your AK-47 or whatever, the emotional content is diluted because the narrative drive that comes from the artist's point of view is lost! How do you re-create that?
Well, one thing I've noticed is that the enjoyment of a game is the direct experience of structure. You try to understand the gaming structure, the mechanisms, the trade-offs. If I do this, it costs me that.
That's where the point of view comes in. From the game designer's standpoint, the point of view is expressed precisely the way you just said.
Whereas in film, you don't want the audience to be so conscious of structure. If it is conscious of structure, you've probably lost some of the emotional grip.
If you're sitting in the audience going, "They're trying to tug my heartstrings," you're out of the movie.
So, are games, are interactive entertainments, a fundamentally different experience? Is any apparent overlap between movies and games just artifact?
In a game environment, the creation of a sense of place is extraordinarily meaningful. It's still so early. What if you could drop somebody in a spectacularly realistic simulation of anything? Let's take this one more step. Jim Cameron talks about the effectiveness of what he calls movie stories in a can. What he means by "in a can" is you're stuck in a colony on an alien world. You're stuck in a spaceship. You're stuck in an underwater base. What you have is a very confined, defined, environment. You're not going to go to Tokyo all of a sudden.
It's almost like he sweats the drama out of the characters in a situation like that.
Exactly. Some of the most effective gaming experiences I've had, going back to the Infocom games, are those that have a really cool world where the players can go just where you, the designer, want them to go.
One of the pleasures in the Infocom and adventure games was that frame-breaking you could do. You're stuck, you're looking at the string and the lump of clay, and you've got a match. Suddenly, you realize that you've got plastic explosives and a fuse and a match and that you can blow your way out of there. So there's a little satori, I mean it's a cheesy little satori, but it's a satori!
It's done in movies all the time.
Indy's stuck in there with the snakes.
But it works. So it happens in movies, it happens in games. But, the difference is, everybody's fallen in love with the movies. Very few people have fallen in love with videogames. I was in London a couple of months ago, and I finally tracked down this American performer, Dennis Seavers, who moved to London in the '60s. For a while, he drove hansom cabs around London, and gave verbal tours of the city. But he would make up a story that integrated the tour. So he'd spin a novel set in this world of Old London. And eventually, he decided to do it in a house. Three nights a week he takes eight people through about 350 years of history by taking them room to room through this house. Every time you go to another room, you jump 50 years. But the conceit he uses that's profoundly cool is to have people still living in the house. He presents it as-if. So you walk in a room and there's still a meal there, steam still coming off the coffee. Somebody's just extinguished a tobacco pipe. You can hear people talking and moving through the other rooms. And he does it all by himself with tapes timed so that when he leaves one room, the next tape starts in the next room. It's a multimedia experience. CD-ROMs are going to have to go a long way to catch this. My wife described it as Pirates in the Caribbean by Stanley Kubrick. It's really amazing! He throws so much detail at you, about how the world worked, and what people thought, and how they looked at the universe - you're transported.
You were talking earlier about how people have fallen in love with movies, but they haven't fallen in love with games. They might have developed a really killing jones for a game, but that's not the same as love. With movies, or drama, at least in the best experiences, you have the sense that you now know something about the world, or about yourself, that you didn't know before you walked through the door. Why don't games do that? Is it that the experience is so narrowly imagined?
You need a lot of real estate. Dennis Seavers packs an extremely high-resolution, information-dense experience into three hours, so much so that you're absolutely immersed. It's like a Berlitz class in English history.
You're allowed to add a lot to the printed experience. When I was a kid, I had a clam-digging buddy who had a gift for simple speech. He held up Moby Dick and said, "There's a big whale in there." It's your job as the reader to generate this experience, detonating all sorts of meaning between your eyes and the printed page.
People who read a book and then go watch the movie say, Gee, it's not how I visualized it. A book can evoke pictures in your head. But your pictures are different from my pictures. The second a visual artist comes in and starts to say this is how the world looks, well, they'd better be a damn good visual artist. I mean, like Roman Polanski in Chinatown, he's a damn good visual artist, and you're really transported.
When you write a screenplay, you need to create a character that people suspend their disbelief for.
Sure. But when you're writing a screenplay, you create a subset of that character, you create these puppets, and you hope their motivation is sufficient enough that they appear realistic. That's all you have to do. The bottom line is that you can create a teeny tiny territory in a movie and have characters appear spectacularly realistic.
And why is that?
Because people fill in the blanks.
So, maybe one of the fundamental problems with software, or entertainment software-
Is that the second you examine it for a while, the blanks become extremely apparent.
That's right. You can take the thing apart.
It's like the Turing test. It really is. If you can look at something long enough to figure out it's fake, you've got a problem. That's the nice thing about Doom. In Doom, you've got to be aggressive. And aggressive environments work really well. If you've got some stuff shooting at you, you can't stop and smell the fake roses, because if you try to, somebody's going to put a blaster up your ass.
Right. All an enemy really has to do is show up and shoot you. But if the enemy tries to convince you to believe in a mission, that's a different story. So, is Rocket Science working on this? What are your next titles?
We're doing a World War I fighter game, and we're developing a spiritual adventure game called The Presidium. The nice thing about Rocket Science is that we developed all these nifty graphics and great production resources, and now we're starting to look at how storytelling can change.
I have this concept of "first waffle software": the first waffle on Sunday morning is the one you throw away.
Exactly.
And usually it applies to the first wave of software shipped on a new machine. But it can also apply to the first game from a new production group. I was talking earlier about studio model versus production model versus product manufacturing. Rocket Science is very clearly a production model - it packages talents with a variety of disciplines to create the end product. How did that come about?
The big score at Rocket Science was [co-founder] Ron Cobb. Everybody's talented. But Cobb brings to the table this kind of polymath ability, to not only think of it, but draw it and design it. There are two or three in my world, the movie business, who can do that, but Ron Cobb is pretty much on top of the food chain. Let's say you needed a time machine. For years, Ron was the only guy in Hollywood who could give that to you. Ron would say, Well, how about a DeLorean? How about a DeLorean with a phase conjugator in it, or a flux capacitor? He knows them both, and he can explain why they work. What drives people nuts about working with Ron is that he knows why every bolt is on his spaceship. They're there for a reason.
We work in a medium in which you can design things that work. That must be pretty appealing to him as well, to create something that is rendered as a 2-D image, but if you build a mechanism underneath, you can have this whole machine working.
Yeah! Taking a 3-D model and adding physics to it is a really cool idea. I'm always fascinated by the physics of interactive entertainment. Most games don't mess with physics too much. They're trying to re-create the world in the magic box. But the magic box is capable of doing many different kinds of physics, where time doesn't work like it does in our world. All this obsession for real time misses the point. People are so concerned about realism they haven't thought about naturalism, or romanticism, or surrealism.
Or magic realism.
All these things are completely possible, but the literature of interactive is so immature at this point, it can't accept surrealism or dadaism or any of these other things.
The Poetics of interactivity has yet to be written.
Or the Pattern Language of interactivity. Eisenstein became this great theorist of film because he didn't have any film. For years, the Soviet Union didn't have any hard currency to buy any movie stock. So they sat around and asked, What if we had movie stock? What kind of movie would we make? And they developed an aesthetics of filmmaking. There aren't that many people, right now, who are asking, How do we push this medium? Ron Cobb can do that.
In any medium, there's an entrenched priesthood that has control of the technology, control of the medium, control of the economics. People who have the money finance productions. You place two things in front of them, one is new, and all the artists believe in it -
And the other one is safe.
Or a redefinition of the existing marketplace. In our industry right now, to get shelf space, it has-
To be like Doom.
A Doom-alike. In a more mature, talent-driven medium, like the movie business, there's enough to go around so you do have an indie, or independents' channel. There is a Sundance festival. But there is no Sundance of interactivity, even though there've been a couple of tries at it: the indie channel is still evolving. Things tend to get locked up in the leading-edge technology and in the budgets you need to create an experience that gets shelf space.
We're still waiting for The Great Train Robbery. We're like a bunch of guys who just invented movable type and we're saying, Now where's Thomas Pynchon?
Where's Finnegan's Wake?
It's going to take some time to get there.
You've got film grammar, which has changed over time, but is essentially sequencing graphics to impart narrative. It's very well known, very formal.
That's right. It's a language.
In our medium, the interactive grammar's form is dependent on structural imperatives of the machine. The machine does this thing so you can do this other thing. If your medium is a shape shifter that changes regularly, do you ever get to the point where you've got a masterpiece? Part of what you do is develop a new medium as you create your work in that medium. Chris Crawford, who I guess you could say is the self-appointed dean of computer game design-
He's the John the Baptist of interactive. Out in the wilderness, screaming.
Right, and very smart. He gave a very compelling presentation of first principles of interactivity, which I'm going to paraphrase. He says it's like a conversation between two people. I say something in a conversation and the other person hears me. Before this person responds, he or she processes what I've said, determines what the result is going to be, and sends it back to me. And then I hear it, and I think about that, and I send it back, and then we've got this feedback loop going - both sides are processing, and we're collaboratively generating this experience. Interactive media should take one of those people and replace him or her with a machine. Now this is very profound and has great implications for the nature of interactivity. And it really defines the difference be-tween selectivity, which is the Coke machine - hitting buttons and getting a predetermined response - and true interactivity, in which the machine is processing and deciding what it's going to do as a result of your actions.
Well, the bottom line is that the computer eventually has to have some model of the world. That's what we as interactive artists have to provide. It has to be able to really manipulate. In most games, nothing happens when you walk out of the room. It's like My Living Doll, the old TV show in which the guy was dating a robot. When he left the room, the robot would shut off.
Right. C3PO says: "If you won't be needing me, I'll shut down now." If you're trying to model a world with potential destinies built into it, and you as the author are writing stories at the systems level, then you have to have some sense of how the world works, and you have to be able to express those in a codable form. You have to be able to express the algorithms of tragedy and its data structures, or the algorithms of comedy, or whatever.
That really is the Poetics of interactivity. And guess what? Aristotle ain't shown up yet. Those fundamentals are really crucial, but they have yet to be defined, or demonstrated. And one of the two has to happen. Somebody's either got to be Aeschylus and do it, or somebody's got to be Aristotle and talk about it.
Well, you talk about the Greeks. There was a line in a great biography of James Joyce by Richard Ellman that really struck me. Oscar Wilde, as you know, took this spectacular fall. Ellman said that Wilde was "Greek enough to know that over-reaching attracts nemesis." And this over-reaching attracts nemesis notion struck me in a major way, because that is a mechanism of tragedy. Wilde, in his The Ballad of Reading Gaol period, spent years deconstructing the elements of his own downfall, his own tragedy. He was able to name all the pieces and describe the mechanisms of how this thing happened to him. If you as a world designer understand those mechanisms, you can build them into your system, so characters who are wandering around in a potential set of destinies can know what to do when they encounter a situation that requires this knowledge. Your characters will know how the world works. So I started looking more closely at clichés: "misery loves company," "everybody loves a winner." Why are these things clichés? Because they've been true. Now, if you could build up a database of these -
The clichés are an encapsulation of how the world works. In the program.
The ones you choose to put in and the way you weight them constitutes a form of authorship: it's your subjective vision about how the world works. I could do one and you could do one, and they'd be entirely different. We turn our characters loose: they dip into different worlds.
I'm in this weird position, because often I'm asked, Who in the movie business is going to do good games? Who's going to do good interactivity? I think it's the people who are able to play with the narrative structure, because they're so fluid. If you're Stanley Kubrick or Oliver Stone or Quentin Tarantino or Martin Scorsese, or you're comfortable enough that you can play around in this medium fluently, then I want to talk to you. It's inevitable that one or two of these guys who are really fluent with the nature of film, the grammar of film, will jump into the interactive pond and potentially make a big splash. But in Hollywood, everybody thinks that because you know computer graphics, you're the guy.
It's so visually oriented. People don't understand that it's not just about what you see. That's part of it - it's also about what you do. The notion that the audience is doing something is fundamentally jarring to a lot of people.
How many computer programmers are out there who could get the idea of building a data structure for tragedy?
A lot of times in the games business, people say, Ah - Stories! The first demand I have is, Give me a definition of story before you even tell me that you're going to make a story. The best one I've heard wasn't from a game designer, it was a terse one from Raymond Chandler: "Story is something that happens to characters you care about." That has many implications. People who create software don't necessarily think that way. Some do, but it's rare.
Conscious writing or creation is very rare.
I had a friend who was a featured actor in Basic Instinct, and she invited me and another friend onto the set. We were in the room where the murder victim would be discovered in his bed while Paul Verhoeven, the director, set up the shot. And this process was really fascinating, because of the amount of control and the amount of mindful creation of sequenced graphics taking place. They were working consciously with things like collision. They wanted the police boss and Michael Douglas to come face to face at some point, to signify their conflict. And that's why, when people begin to get cameras and make QuickTime movies, they've got a long way to go before they can work consciously with collision graphics in a discovery-of-a-body scene.
Absolutely. Ninety-nine percent of so-called digital movies on CD-ROM have the visual grammar you'd find in an Edison One reel. The acting's just as bad, too. It's just awful. For people who can't get a job in television, it's a tossup between the infomercial and doing some asshole's CD-ROM project.
And the asshole's CD-ROM project is cooler because it's interactive.
Exactly. And then we spend all our money on those guys. The budgets on these suckers are so small, you either have to call in a lot of favors, or you're going to get a lot of crappy talent. Big-budget movies like Speed and Alien have a contained environment in which you can have an experiment. You're looking at a microcosm of the human condition that is quite representative of the macrocosm of the human condition. And what's really slick about that is, by taking this pertinent subset, you get originality. Look at these two-character plays like My Dinner with Andre. They're a rich environment. Human interaction is by nature so complex emotionally, an examined subset fills up the beaker. Two people in a room is incredibly powerful.
So you throw ten people in a bus, of which three or four really have foreground roles. And you do get a sense of how the world works and how people work. It's got this great kinetic action going.
Look at the other side of it, like Pulp Fiction. There are no big crowd scenes in the whole picture. It's really small. And it's really about putting two characters, like Uma Thurman and John Travolta, in a can, and seeing how they react. There are scenes of such surprise. God! I'd love to see some of that in interactive! Can a synthetic character ever really get past the Turing test limitation?
I think so. It's not going to happen with procedural mechanisms, all well defined, where all you really have is a Rube Goldberg machine and no matter how many times you drop the ball in, it's going to come out that same bottom slot every time. What's going to make intelligent characters happen are artificial-life mechanisms, in which emergent behavior arises from communities of interacting entities that are operating with simple rules. This gets to what Marvin Minsky talks about in his Society of Mind, or Cindy Baron imagines with software Jungian archetypes, in which there is a community of interacting entities that together produce a behavior transcending the individual behavior of any of the constituent entities.
It looks like real human behavior because what's looking back at you looks like human motivation. It has to be a system that allows for emergent behavior. Like the real world.
The real world is nothing but emergent behavior, for Christ's sake. And this really points up the notion of getting a programmer who can also construct an algorithm of tragedy. We need to be working with people who can construct artificial-life mechanisms or distributed communities that model personalities capable of emergent behavior. It's not just creating an entity that's capable of certain behaviors, it's creating that entity and then siccing shit on it, so it gets scared, so it has to dream up something new, to survive.
Sure. Environmental pressures.
Yeah. That's what I think a lot of drama is all about: putting characters in situations, getting them in trouble, keeping your protagonist in trouble. But get him in trouble and then let him work his way out of it. Like your movie-in-a-can.
Absolutely. Peter Bogdanovich, who directed The Last Picture Show, has this great line: the key to great movie comedies is making sure that the next person through the door is the person your character least wants to see. It's so true. It works in a horror movie, too. It works in a lot of movies. You have to put the character in a pressure cooker, otherwise you don't have any dramatic action. Intentional filmmaking is really important. Every single Stanley Kubrick shot is there for a reason.
There's an interesting question about the relationship of intentional writing to commercial success. Somebody asked William S. Burroughs, Did you ever decide to get up in the morning and write a big, splashy, commercially successful novel? He said, "Well, I'll tell you, that's what I'm always trying to do. But this is what comes out!"
Often, people will go out in the science fiction community and try to rope in a novelist to write a script. It doesn't work very often.
Why is that?
Just because somebody can write a book doesn't mean he or she can write a movie. And just because somebody can write a movie, it doesn't mean he or she can write a videogame. A lot of people have a hard time figuring that out. The people I like to work with are the people who are really trying to push this thing. I've worked with Rocket Science for a while, and I've been working with Steven Spielberg, and that's been great, but one of the most fascinating groups I've worked with is this little company called The Electric Community. They're doing a thing called Reno, for Fujitsu. They wrote Habitat at Lucasfilm. They're trying to create the infrastructure for a networked world. They are Randy Farmer, Doug Crawford, and Chip Morningstar. They've thought it out down to the tiniest detail on how a networked universe has to work.
About the physics and behavior of the creatures that'll live in it?
Absolutely. They're really into muses and MUDs. They're the Eisensteins of this medium. It's profound to find any Eisensteins in this medium. The idea that there are theorists of network communication, outside of an academic ivory-tower mutual masturbation society, is amazing. They want to build cyberspace. It is going to change the world. The one thing I want to do more than anything else is not just write movies or make videogames. I really want to see a new medium emerge. Networked computer environments provide that medium. It's community. We live in a world that everybody tells us is a scary place. And whether it's true or not has nothing to do with reality. People believe it. The Net is the answer. You bring back the agora, the Greek marketplace, and the idea of a place where people can go and hang out. That's what these guys want to make. When you have a new medium emerging, it sooner or later has to find its natural distribution method. Thomas Edison invents motion pictures, but then he's got to figure out how to get it out there. Gutenberg invents moveable type, and then he's got to figure out how to get it out there. Somebody invents interactive computer media, and then you've got to figure out how to get it out there. And let me tell you, how to get it out there is staring us right in the face. n
