Story Time With Valve's Erik Wolpaw, Pt 1

It all began one sunny, seemingly inauspicious afternoon in a Starbucks. It also ended there - but, you know, later. Ragged and bone-weary from three days of wading through PAX's diseased hordes, Valve's Erik Wolpaw, Double Fine's Anna Kipnis, and I huddled around one last vestige of civilized humanity: a table. Then we spent nearly an hour talking about this year's sudden upsurge in crazy-interesting videogame stories, because it seemed like the thing to do at the time. It isn't anymore, but - if you'll believe it - it was considered cool back then. Those were the days. Anyway, here's part one. If you behave yourself, you might get part two tomorrow. And maybe a cookie. But probably not.

RPS: People spent ages during your PAX panel bothering you about entirely non-scripted stories in games like Day Z, and you said you never, ever, ever want to talk about it again. Sooooo... mind talking about it again?

Erik Wolpaw: It weighs on my mind heavily, but I don’t really have an answer for it. Not at the moment. I don’t know what to do about it. If I had my way, every game would be full player agency. It’d put me out of a job, but that’s actually what I like as a game player. I think about it every day. Literally every day I go to work and I think about it. It’s part of my job to think about it. How to do that. I don’t have the solution yet. They didn’t seem to have the solution. Anna doesn’t have the solution. Tim [Schafer] doesn’t have the solution. Tim’s fallback, which I respect, is "I’m gonna make the games I make - and that’s it. That’s what I’m gonna do, this is what I like to do." Which is respectable.

Anna Kipnis: It’s actually a lot of work to write a good game story -  to craft a really good story. But to have player agency and have a good story, that’s extremely difficult.

Erik Wolpaw: Yeah. They are at odds. The one answer I did give is that, even with decision-tree sorts of stories, it’s crazy hard to come up with one good ending. Much less five good endings. Then the problem is, if you come up with five endings, hopefully one of them is great. The chance of all five of them being great, let’s just call it zero. One of them’s gonna be okay, and three of them will probably be crap. At the point where you’re trying to have that, you’re like "I don’t want to give people the crappy endings." I want to give them a good ending.

This is, again, in an authored experience. If it was full player agency, which is totally theoretical, but in some full player agency procedurally generated thing, I guess it wouldn’t make any difference. It really would feel like it was on you as the player. Like, I got the ending I deserved because it wasn’t preordained. But in the case where it’s even a little bit authored, it’s preordained. If they get the mildly crummy ending, even if the decisions they made led to that, it’s still gonna be like, "Somebody wrote this crummy ending for me, and I’m not having a good time." It’s just tough. In some ways it’s laziness on my part, I guess, but just knowing how hard it is to come up with a good ending…

Anna Kipnis: People take for granted that writing is easy. It’s not.

Erik Wolpaw: Yeah. Everybody can write. I mean, you’re a writer. But everybody thinks that they can, because everybody can to a certain extent.

RPS: I think also, it's a modern-day society thing. We’re all tuned into social networking stuff and texting and whatnot, so we’re always writing. Everyone’s always writing.

Erik Wolpaw: Yeah. Writing all the time. Even with comedy stuff, everybody gets a decent zinger off in an e-mail every once in a while. Everybody’s capable of it. The hard part is delivering it consistently and reliably. But yeah, it’s a weird skill to have. I wouldn’t even presume to be like, "Well, I could probably just sit down and write the physics system for our game." I wouldn’t even know where to start. But people are so tantalizingly close to being able to do our jobs that it’s difficult sometimes. I mean, it’s also rewarding and all that.

RPS: During one PAX panel, someone asked if writers are valued in the gaming industry. The impression I've always gotten - at least, from a perspective of what's absolutely crucial and what's not - is big, fat "no." I mean, for instance, even BioShock's plot was still up in the air six months before release. But at Valve, development and writing are parallel processes, right?

Erik Wolpaw: I know all this, obviously, just secondhand, so it could be a disgruntled writer thing. I don’t know. Luckily, I’ve worked in two places. Double Fine, where writing is obviously really valued, and Valve, where it’s also really valued. So my two experiences have been at places where it’s important to the studio. I think most studios that are going to put a story in their game do feel like it’s important, and the team probably feels like it’s important, but they may just not know how to go about getting a good writer or integrating a good writer.

Again, part of that’s because everyone can kind of write. It’s hard to judge whether this person’s going to do a good job writing your game. Again, with the physics system, there’s at least some bar you can test them with, to see if they’re going to get close. But with writing, everyone can write. Unless they have a track record of doing it, [there's no point of reference].

Anna Kipnis: This is something that Erik probably wouldn’t say about himself, but he can actually program. He might not be able to write a physics engine, but what I’m saying is… The way he writes dialogue, and I’ve worked with him on a lot of stuff, you know how it’s going to be put together. Somebody might be able to do comedy and they’re good at writing zingers, but they still might not be good at writing video game dialogue.

Erik Wolpaw: That is a thing. And Tim’s a programmer originally, too. It’s a useful thing, to know… Well, to be able to integrate the dialogue into the game yourself, but even if you’re not doing that, to have a really clear understanding of what it means to put it into the game. Pretty much every position working on the game - and Anna I think would agree - benefits from knowing some amount of programming. Having some idea of what goes into it.

You want to know all the systems, too, because part of it is you’re working in a group for a long time on a project. The decisions you make impact a lot of different people. There’s this artistic integrity where everybody should obviously be working as hard as they can to finish my vision. But there’s also the reality of, these are people that I care about and I work with. I don’t want to make them do a ton of work unless I’m sure that it’s at least got some chance of success. I want to know, the things that I’m asking them to do when I’m making a request, what that actually means. Tim knows and just doesn’t care. You will just get this done. Maybe he’s not like that as much anymore. I remember on Psychonauts… well, everything’s fine now. We get along. Anna’s just not saying anything. [laughter]

RPS: I think there's definitely something to that idea of a singular creative vision, though. I mean, some of my favorite games this year were developed by one-person teams. What I felt like that helped them do is realize one vision really well. Because it wasn’t just a matter of them knowing program and also how to write a story. It was them knowing where the game was going to go. They even made the soundtrack.

Erik Wolpaw: What games are we talking about?

RPS: Have you played Lone Survivor?

Erik Wolpaw: I did, a little bit. I didn’t get too far into it. But I know what it is. I didn’t know the guy made the soundtrack to that too.

RPS: Yeah, Jasper wrote the whole soundtrack. So there are points where the music will come in at this perfect moment, and you can tell it’s just because we had this vision for the game. He knew exactly where he wanted these things to happen. It’s affecting to the point where it’s like… "I feel sad, but I’m not really sure why I feel sad. I just know I feel it really strongly right now."

Erik Wolpaw: There’s definitely that. That singular vision sort of game. That’s obviously tough to pull off with a team, especially when you have a work environment where everybody gets to contribute. Personally that’s the environment I want to work in. You can argue with the results, but there’s no way that you could have made Psychonauts or Portal with [a one-person structure]. They simply aren’t one-person games. There’s a lot of interesting stuff. This isn’t really story-based, but some of a Brendon Chung’s stuff - like Atom Zombie Smasher and Gravity Bone…

RPS: Yeah, I was actually just about to bring up Thirty Flights of Loving.

Erik Wolpaw: On Thirty Flights of Loving, I guess he had [current Double Fine community manager] Chris Remo do the music. Even he had to branch out a little bit to realize his vision. And, of course, there's Tim. He actually is a genius and has a vision and could push people to do it. I’m not really confident enough. I want the crutch of having people who are smarter than I am around me to do this. I do not want the responsibility of having to do it all by myself. I mean, because I couldn’t. Fully honest with myself. I’m not going to do that. But yeah, the polymath guys, who are artists, musicians, programmers, writers - it makes sense for them.

RPS: On the topic of stuff like that, you were talking about how in an ideal world, what you want is full player freedom. But how about things like Thirty Flights of Loving? I mean, it's ultra-scripted, but still brilliant. Plus, I felt like simply being able to look around the environment and interact a little really added a lot to my attachment to the story.

Erik Wolpaw: Thirty Flights of Loving and Gravity Bone -  they’re amazing. I just don’t know where they exactly fit in the world. I love that they exist, but I wouldn’t want every game to be like that. At some point I want my game to be a performance test. I want that as well. Gravity Bone actually has a couple little puzzles in it, some platforming. Thirty Flights of Loving, I think he wisely took it out, because the actual gameplay of Gravity Bone wasn’t that great.

As much as I love them, it makes me a little bit sick to my stomach. There’s no game in it, but I like it. Maybe I shouldn’t like it as much as I do. But I do feel like that’s definitely one viable branch of the interactive narrative. Maybe you don’t even call it a game. It’s an interactive narrative. It seems like this demo reel of narrative tricks that Brendon Chung has thought up. Here’s all the things I’ve thought up about game narrative in the last year that I’m gonna package into a small package that holds together, but it’s almost like a demo reel.

Every time I play either one of them, I’m like, "I’d love to steal this, this is so ingenious what he did here." Like the jump cuts in Thirty Flights of Loving. Why did I never think of that? I can see why other people didn’t, but it’s my job to think about things like that, and it never occurred to me. The chase scene in Gravity Bone is amazing. Chases are really hard to pull off, because games train you to run forward. Like, if we give you some danger you should probably go fight it, right? Whereas there’s that scene in Gravity Bone where you crash through something and you land on that table and everybody looks at you… I’m babbling about it now. But it’s great. It’s such an incredible narrative moment that had no dialogue. Totally interactive. It was amazing.

But Brendon Chung… Part of the thing that sticks in my mind are the seamless things that he does. He’s clearly someone who understands game design as well. Atom Zombie Smasher is almost pure game, right? It’s literally a game about organizing dots in a maze.

RPS: What I liked about it was that even though it was very narrative-based, it didn’t spoonfeed you the plot. It just gave you these little snippets, like you said. It’s more up to the player to piece together the whole picture. To the point where there are some bits where even once you’ve finished it and you sort of know the general arc of it, you don’t quite know everything. I guess to me that was the game. Putting everything together.

Erik Wolpaw: My only thing about that is that… Again, I want to preface any criticism, and I don’t even know that this is criticism, with the fact that I sent Brendon Chung fan mail after Thirty Flights. I don’t do that a lot. But I felt like the structure that Thirty Flights of Loving had - piecing it all together - you could have pulled that off in a movie as well. I didn’t think it was specifically a game thing, where there was so much exploration involved.

I felt like the puzzle pieces were there. Maybe there were a couple of parts where you had to poke around a bit to get them. But you were basically moving from one point to another, in a linear way. If you gave me a movie of Thirty Flights of Loving, the puzzle would have been effective. Anyway, that’s a challenge. When Brendon Chung manages to mix Gravity Bone and Atom Zombie Smasher together, it’ll be the greatest game ever made.

RPS: So what about something like Dear Esther? Again, it’s that sort of mental puzzle, except that the game part is that it rearranges everything each time.

Erik Wolpaw: Dear Esther is interesting. It’s just that the tone of Dear Esther isn’t necessarily the tone that I go for. I thought it was great. I guess it just didn’t speak to me as much as Gravity Bone does, because that’s more in my pace. The story that he’s telling. But yeah. Also along that same line of this new genre - I don’t even know if it’s new, but it seems to be making a resurgence - of the interactive short story. Both of them, I’d like it if you could do that, still make a short story, it doesn’t even have to have a lot of game, but that involved a little more exploration. Dear Esther was pretty linear. I know that in the beginning you go down the beach the one way and get to a dead end and come back. But they’re both pretty linear. And with Thirty Flights of Loving, it’s linear game-wise even though the story it tells is totally not.

All this stuff - whatever criticisms I have - it’s not based on why they do this. It’s a criticism of me thinking about it because this is part of my job and being like, how can we make this even better? This branch is exciting. This narrative direction where it’s, again, like an exploratory, potentially even combat-less narrative experience. There must be some less arty way to put that.

RPS: I talked with Notch yesterday, and it was interesting, because he’s actually very fascinated now by stuff like Proteus where it’s completely stripped out. It's kind of the total opposite of Minecraft, where anything can be interacted with. You just wander around this island. Have you played Proteus at all?

Erik Wolpaw: No. What’s that one?

RPS: It’s first-person, and the graphical style is really simple. Almost painter-like, but not quite. You just walk around this island. It’s very naturally beautiful. Then things happen sometimes - really ethereal, crazy stuff. At one point I walked into this little circle of rocks and things, and suddenly - it was daytime - the sky just zoomed past, and it was night almost immediately. I basically wept at that point. Except doves came out. I wept doves.

Erik Wolpaw: This seems like more of a traditional game-ey game, but the procedural generation stuff is really interesting to me. Like Jim's robot game [Sir, You Are Being Hunted] looks really neat. I don’t know what he’s doing narratively with it, but as something I want to play, it looks pretty cool. It’s such a different branch entirely. And Minecraft is procedurally generated exploration, basically.

RPS: Whenever you’re doing story stuff, whenever you’re coming up with what you write, what's most important to you? For some people, it’s very much a conceptual thing - like, "Here’s this high-minded idea about what I can do with a story in a game or a book or whatever." I’m going to write around this. Other people are just like, "I have this cool idea. I want to tell this story. I’m going to execute it." Narrative techniques come of that. So where do you start? Is it a technique, or is it more like "This is the story I want to tell"?

Erik Wolpaw: The process tends to start with some gameplay, a gameplay mockup, and I try to figure out a story that will work with this very basic gameplay. Not a finished game, but you start poking at "OK, what story could complement this?" And not just what story, but what storytelling, what method, would complement this gameplay that we have here? Again, Left 4 Dead, very different kind of story structure from Portal, which is very linear. And then you just kind of go back and forth from there.

I think my co-writer on Portal, Jay [Pinkerton] thinks the same way. We tend not to start with any themes, because we like to let the story develop and then look for themes afterwards. If you push a particular theme, you might get sort of frantic about it. It’s easier and more fruitful to go back and see what themes have evolved as you’ve been writing the story, and then you find places to maybe push those a little bit.

RPS: Your results always end up being incredibly unique from a structural standpoint, though. I mean, Half-Life 2 and Left 4 Dead were brilliant at telling story via little environmental cues, and Portal's a textbook example of linear story done right. Is there pressure to continually top yourself in that regard?

Erik Wolpaw: Yeah. There’s always pressure. But Valve tends to hire kind of highly motivated people who would probably feel that pressure [regardless of whether or not they were at Valve]. But yeah. It makes everybody better, too. You understand that you’re doing this thing and it’s going to have the Valve name on it. People expect it to be really good.

The upside to working at Valve, though, is that they will not put something out that’s not good. If it takes an extra six months to get it right, Valve isn’t going to go out of business. You do have that going for you - which takes some of the pressure off. The company isn’t riding on every single decision that you make.

But yeah, there’s a lot of pressure involved. As there should be. Someone at Valve said recently, "If you’re not doing something that scares you, you may not be doing the right thing." At some fundamental level, you should be doing something that feels kinda scary. Like, I could fail at this.

Check back tomorrow for part two, wherein we discuss scrapped Portal ideas, virtual reality, Valve's newfound love of community-driven content, how the Portal song came about, and why Wolpaw still hasn't played Psychonauts.