Creating A Glitch In the Industry

Stewart Butterfield is best known as one of the founders of successful photo-sharing service Flickr, but he's long held the inspiration to make an online game. Prior to launching that service, he worked with a team on Game Neverending, a title that never launched, but was a shared social and creative space.

Post-Flickr, Butterfield has moved forward with plans to launch Glitch, which he hopes will become a successful social online world in a way different from traditional MMOs.

The game, which is built on a sophisticated and flexible web-based toolset which allows quick iteration, is a colorful and appealing, with a variety of aesthetics and snarky humor. It also is designed to allow players a voice in the direction the game itself takes.

Butterfield hopes the mix will attract a savvy new audience. This might be a difficult proposition, as new-style gameplay is tough to market, and creative-focused social spaces are still unproven. But with investment and artistic vision behind it, Glitch stands a chance.

Where'd you get the idea for Glitch?

SB: So, I've been kind of thinking about this for a long time -- I've been thinking about it forever, basically since Sim City. When I used to play Sim City, I got obsessed of the idea of playing the game from the perspective of the little ants, like when you see them on the cars and freeway. Rather than top-down, it would be emergent from the collective decisions of all the people playing the game.

I feel like MUDs became Ultima, EverQuest, WoW, and MOOs and MUSHes just never went anywhere. There's that path that just really... I don't know, I think it would be really interesting. There's obviously things that are kind of similar, kind of inspired by... Metaplace was like that a little bit.

Was.

SB: Second Life... Yeah, "was."

Second Life is also an incipient "was", I think.

SB: Yeah. Yeah, well... In both cases, I think, there wasn't enough game context. Well, there wasn't any game context to take off. I remember the first time I ever installed Second Life and sat down, I was like... First of all, it was super fucked up then. I mean, it was really buggy... That was probably 2003.

And actually at that time, there was kind of a buzz. There was Second Life, then there was There, and then the Sims Online was about to come out. We felt like that was like, not "social games" in today's sense, but there was going to be this era of social games, and all of them busted basically.

I think There [did] just because they spent too much money, because otherwise a lot of it had really nice polish and nice feel. When you were talking to someone, they had a great way of doing eye contact and spreading people out in a group, so it was a good social experience.

But again there wasn't any game there, and it was all about these brands. I don't want to go into a virtual world and look at Gap shit, American Eagle T-shirts... It's just... I don't know, it seems kind of gross.

I'm sure you could write a psychology thesis on it or something like that, but you can't really role-play in that context. If you have real world brands in front of you, you can't... You can't invent a persona because you can make yourself look different and you can fly and stuff like that. I don't know, it definitely breaks the magic circle. There's no real opportunity for playing.

There's never been something that I feel like has been the right balance of social hang out, social experience, and enough of a game context. I mean, Metaplace was fucking awesome, technically, the stuff that they did. The tools were really cool and stuff like that, but again, just like Second Life, you get there and like, "What am I going to do?"

So, I think that the idea of a collaborative simulation is probably the core bit of it. That and bringing something like the MOO or MUSH experience to a bigger group of people, people who never got to experience it because they didn't... You know, if you got online on '95 or later, they were pretty much dead.

So, what do you see as sort of the important aspects of that experience that you want to preserve and move forward to people?

SB: That it's open-ended. It makes it tough for us to market, right? Or tough for us to have a headline. It's a real disadvantage, I think, because... like FishVille -- or I don't even know if that's a real one -- or any of those subject matter-plus-Town, Ville, or City, it's totally obvious what you're going to do.

Or if it's an existing genre of any other kind of game. People know what an RTS is. People know what a first-person shooter [is]... It could be a variation. You could like the style. Maybe they have some nice mechanics. Maybe they have better physics or whatever. It doesn't really matter; you know whether you like those kinds of games or not.

For this, what we want to do is build something that's a lot more open-ended, where people end up building the world out the way they want but still in the context of a game. So, it's not just, "Go and make a world." You get building blocks. There's a physics -- I don't mean in the Newtonian sense -- to the world. There's a ruleset and dynamics that are built into it that people can manipulate and take advantage of.

But I think that the game is much more about driving the culture, like people can create corporations and religions and weird cults and stuff like that, and create an agenda, and drive it forward. So, there's roleplaying in that, that doesn't have to be too serious.

What we really want to capture is... There's a kind of social attraction that happens in the context of games that just doesn't happen elsewhere. My dad loves to play bridge, and he doesn't like playing bridge against the computer at all, even though he likes the mechanics of the game.

At the same time, he wouldn't just invite those same three people over his house just to hang out if they had no agenda because there's something that happens in the context of the game. There's the competition, there's the friendly banter, there's the out-thinking each other, and it's just not possible outside [of that context].

I think that's a place where video games currently really actually fail a lot.

SB: I mean, poker is the same way, right? So much of the game is just looking at the other person.

How have you gone about developing the title?

SB: So, we invested a lot. Well, we started the company 18 months ago, and it was only maybe three or four months ago that we really switched more of the effort going into game-making rather than tool-making.

You told me when you were working on Game Neverending prior to Flickr, you thought it would take about 18 months to make it into a game.

And now you're talking about like putting a big investment into this in terms of launching it, which is basically completely counter to the way most people talk about approaching the social game space now. It's all about spending two to four months on something and kick it out there.

SB: Why spend so long on it? Because we want to make stuff that we know is way better. I mean, the games, I think people exaggerate how long they spend on them really. It's probably not that much longer, maybe three months to six months or something like that to two to four. But there's no doubt, right.

One of the reasons... The amount of stuff you can do in a game day is gated with the energy in pretty much every social game, because otherwise you burn through all the content in four days or five days or something like that.

Right. That's why it's appointment gaming.

SB: Yeah, exactly. But they're also super linear single-player games. They're social, but they're only social to the extent that you can play and I can play, and I can see what you're doing, and you can see what I'm doing. It's not like if in FarmVille, if I'm better at growing grain and you're better at growing fruit, we can trade and have some real economy or anything like that. It's totally linear, total single-player.

And I don't have an extensive background in game design -- I'm learning as I go -- but one thing that's really obvious is making a good single-player game is really hard. Making a good multiplayer game is like maybe not exponentially harder but much, like an order of magnitude harder. And making a massively multiplayer game is just fucking retardedly hard compared to either of them because it has to be fun in the minute-to-minute thing, and it has to be fun with other people doing stuff that affects the state of the game around you, and it has to scale.

Well, that's like when you look at MMOs, it's why when someone figured out the trinity of tank/healer/DPS, everyone just stole it because someone figured out a mechanic that will keep three people playing together.

SB: Yeah, yeah. I mean, this is not a unique insight; I'm just ripping it off other people, but it's so hard to deviate what from WoW has done now. You can basically reskin it, and you can change the dynamics a little bit, but the only...

Well, this could be wrong, but it seems to me that the only thing that's fundamentally, that has serious followers, is EVE Online. People will say Darkfall is different. I haven't played enough of them to know how different they are.

[Indicates another Glitch developer] That guy's got level 80 with like seven achievements remaining and every fucking pet and stuff like that. I played WoW for five hours, like actually myself with my own character, and I was like, "I can't do it." Not because it's bad -- because first of all, I think I just would spend too much time, but I just didn't see the payoff for me. If I was going to put 20 hours a week into it, it's still more or less linear content, and it's not really...

I mean, the stuff I'm imagining may be impossible. We may just totally fucking fail. I don't know because no one's pulled it off yet. But that wasn't it, right. The amusement park criticism is pretty much right on. You get to a certain level, then you're allowed to go on this ride. And then you get to a certain level, you can get on this ride.

It doesn't much if someone just killed the boss 10 minutes ago; you can go and raid there and kill the boss again, and that's not... The game itself may be fun in terms of mechanics, but then it reduces it to just a really elaborate lobby for a small multiplayer game. It's not really massively multiplayer.

One of my close friends was playing Ultima Online all the time, especially when we were working on Game Neverending. And hanging out with him as he was playing, it really kind of felt like no one knew exactly what was going to happen. And they had issues, they had bugs, and things screwed up, and stuff like that, but that was a lot more exciting to me because at least you didn't know. It wasn't pre-ordained how the world was going to unfold. It could be chaotic, and you could try and do something in the game world that no one had done anymore.

You were talking about going back to the MOO stuff and having the ability to actually have an effect of the world. How much of an effect on the world are players going to have?

SB: Well, I mean, some of it is interaction with us. It's not totally open. We have a tool for location editing. I can drag stuff around. We're not going to give that to players. At least for the foreseeable future.

So people can't draw a penis using those bushes.

SB: Yeah, exactly. Some day, we might want to give some of that, open that up. Instead, we'll design the locations. And by the way, they'll be better, right? We're paying for people who are talented to do it so it looks awesome.

Players unlock new locations. So they kind of lay down where they're going to go. As they unlock them, we'll be a little ahead in terms of content development. Depending on which direction they push things, we'll go and develop more in that direction. We're also going to try a couple things like Tale in the Desert-style democracy for players.

The hardest thing of software development is prioritizing. You have a billion ideas. So, there are two things that we know we want to both of these, and we don't know which one we want to do first. Putting that up to the player, not as part of the game but outside of the game; a meta-game community.

This street [demos the game for Gamasutra] is a location. They're about 6000 pixels, like eight screenfuls wide. It was unlocked during the last test and upgraded a couple times by the players. They have choices in how they upgrade. All the different choices were pre-designed by us, but which ones they actually choose and how the world gets developed is up to the players.

It's almost like Wikipedia as a game, right?

SB: Yeah. It's a big sink of resources to develop this stuff. It requires a whole bunch of gathering stuff and working. People are learning skills specifically because there's a skill required to do this. The skill system is kind of half-Civ, half-EVE, but time-gated. As they push out in different directions, all these streets are new. We'll develop more of those.

Some people are just playing the game, leveling up. There's groups, there's a whole bunch of quests and stuff like that. People who are a little more advanced often go out to the periphery of the world and decide which direction they want to push it. And as they push it in that direction, we'll just develop more in that direction.

So, in that sense, they haven't. There's a real economy that's guided by vendors where we set the price for now. Eventually, we'll figure out how to like unpin each bit of the economy so that it floats free, but that's obviously really, really hard, and it will take us a while before we can completely float. That will be part of the gameplay as well.

You're spending a lot of time, especially for a free-to-play game, with a fairly high headcount. When we talk about the conversion rates that are typical in the industry, will those sustain you?

SB: I think they will. I think we'll have, I'm not sure if I want to say "a lot better conversion rates," but we're aiming for something that's a lot more deeply engaging.

I mean, you saw the trailer. It will have appeal to some people, and it won't have appeal to other people, and that's totally fine. I mean, we set out to do something that's strongly flavored in a sense... Because you want the people who love it to really love it, and that will probably apply to other people aren't going to like it.

Kotaku posted the trailer, probably 50 percent of the comments were like, "What the fuck?" or "Whoa, that was weird." And of the remaining, 30 percent were like, "Gay. It looks like MapleStory meets FarmVille," or something like that. And 20 percent of the people were like, "Wow, that looks really awesome." That's fine.

I think that it's possible to do something that -- like if you have the scale of maybe $1 per user per year that Zynga gets to the $280 that Blizzard gets for WoW -- if we can be like $30 or $40 per player per year on average, then we don't need tens of millions of monthly active users. You know, we're profitable at about 120,000 people playing the game, and having a million people makes it a really solid profitable company. Having the low number of millions would be a phenomenal success, and that's what we want.

So, there will be microtransactions for virtual items sales. There will also be subscriptions. And we're going to try a bunch of other things as experiments. So, for example, purchasable minigames on iPhone and Android that when you buy them, you unlock a new skill in the game, and when you get a new high-score, it increases your skill points.

Companies like Zynga that have a pretty good idea of what people will buy, it's still only a set of answers for a specific context of their specific game and their specific audience.

SB: Yeah. You know, it's pretty easy to burn people out completely, I think, if every interaction kind of resolves into increased purchase conversion experiment, then it's just not fun anymore. We can head back to that stuff in a while.

When you look at the successes of FarmVille, that's as anodyne as it can possibly be. I mean, that's a core part of its appeal. It's deliberate.

SB: I mean, Subway is a very, very popular restaurant... That's one way of being successful. There are definitely other paths.

You've shown me the player's skill tree; it's surprisingly complicated.

SB: Yeah. The great thing is... It depends on how much content we want to build around it, but adding a skill itself takes minutes. But building stuff off of that, like buffs that are based on whether you have the skill or not, items that are gated on whether you have the skill, again, you can point and click in a form-based UI, and anything that's a little complicated, developers can do like that.

We just added -- this is kind of a silly skill -- penpersonship, which is leading to a bunch of bureaucracy skills. Because I think bureaucracy is hilarious. When you want to build a house in the game, you have to get a permit. Because you can just make like all kinds of weird Kafka-esque quests and stuff like that.

So, if you want to build a house, you need to get a permit, and so that means you got to find someone. And the bureaucratic arts level three is a skill that someone spent a week learning when they could have been learning something else, so they'll sell you the permit but at a high price.

Yeah, I like that.

SB: So, you'll like that. I mean, you're the kind of person I think would enjoy the game, and there will be a lot of people...

Which may be a bad sign for the game. [laughs]

SB: Well, you know what... Are there a million people like you?

I haven't met them all.

SB: [laughs] There's a lot of people in this room and a lot of our friends will play it, and even... Again, if it's a low number of millions, it's a fucking phenomenal success, and like I'm going to like fly around in helicopters and stuff like that.

That's the headline. [laughs]

SB: Sorry. I forgot you were recording.

We've reached this point where we're asking "what is a game?" We're asking these fundamental questions that seemed solved a couple years ago. What is a game? What is a player interaction? What are we trying to engender? What are we trying to communicate?

SB: My background, academically, is in philosophy, and there's this great book by a Canadian philosopher. It's called The Grasshopper.

He has this theory that there's a good, solid definition of a game, and it's not very difficult, that there's a goal that you can describe outside of the context of the game. Like it's an objective goal people can understand -- like get the ball in the hole, or something like that. And then there's some way you come up with to make it harder to do that.

Like if you play golf, it would be super easy just to drop the ball in the hole. But instead you decided you have to hit it with little sticks and you have to stand a thousand feet away, so it makes it much harder.

And then three, you engage in it voluntarily.

So, you know, we had this big face-to-face where everyone who's not based in San Francisco came down to this office, and on a break, we were throwing paper airplanes off a mezzanine there. Someone decided that the game is try to get your airplane over the second beam. It's really hard; you almost get it. We spent like 45 minutes or something like that playing this game. But it's like the perfect example where there's like an objective goal, you try to make it hard on yourselves, and everyone buys into doing it...

Maybe being a platformer helps, maybe the avatars looking good helps. The art style could help. The actual mechanics. Whatever it is, there can be some kind of hook, but once you're in, you just decide that "I have this goal to get the most money", "to get to level 40", to whatever it is, or you decide to, ideally collectively with other people, that we have this goal...

There are 11 giants. You're inside their heads, in their imagination. You can donate to shrines that are in favor of the giants or that you can use to accelerate the rate at which you learn skills and stuff like that, so people are able to roleplay in a totally ridiculous way, that they're devotees of Lem, one of the giants, or Cosma or whatever, and come up with some goal, get other people to buy into it, and then it's a game. It's totally game on, and it's not like Second Life where we had to start that from scratch. There's enough of an environment... There's enough physics... physics is kind of the wrong... like socio-physics, I don't know.

I see what you're saying. There's there there.

SB: Yeah. We've already done part of the imaginary work, and then players, in the ideal case, they take it from there and it's emergent. We don't know exactly what's going to happen, and we're there to support the game.

When you're talking about a really broad social audience, people talk about, again, like more and more directed, you know. More and more directed. You know, like if you talk to people who work at social game developers, they're like, "if people don't understand what they have to do in five minutes," or even two minutes, they'll close the tab.

SB: That's true. Like I said before, that's a huge disadvantage for us because it's not like it falls into some familiar genre. Never mind first-person shooter or RTS or something like that. Even the social games are their own genre now that people totally understand. You can say what this game is about right away. We can't do that.

Once people get into it, if they're willing to get over the hump... If there's something, whatever it is -- either one of their friends play, or the style appeals to them enough, or who knows what it is, once they've played for like 15 minutes, then some of them still don't like it, but many people are like, "I get it. This is awesome."

I think that certain philosophy like I just described is again sort of predicated on the retaining the lowest common denominator as your audience, right?

SB: Yeah.

Ultimately, all the metrics in the world don't tell you why people quit. All the split-testing in the world don't tell you why people chose A or B. It just tells you that they did, right?

And that's where metrics fail and where it's sort of creatively galling as a  working methodology. Maybe people quit doing the tutorial because they got halfway through their tutorial and they're like, "This game is bullshit," not because your tutorial didn't effectively tutorialize them.

SB: There's a lot of obvious criticisms of what Zynga is doing. Like GDC this year was basically...

A Zynga hate-fest?

SB: Yeah. Essays were written about it and stuff like that. I don't know. I just read another one last night. At the same time, they're doing something right.

Sure.

SB: And I feel like we owe them something just for expanding the number of people who considered that they could play a game.

Absolutely.

SB: And that will be tremendously helpful for us, you know? And I think of the 80 million people or whoever the fuck played FarmVille, there's gotta be 10 percent of them, still 8 million people, who, for them, they weren't gamers before, they like this experience, and they want something deeper, more engaging.

And the numbers are just so huge that if it is 10 percent, that's great for us. And if it's as high as 25 percent, I'm just going to rent all the helicopters and just fly them around. I'll buy carbon offsets, whatever. That's fine. We can do tricks.

You'll have Facebook Connect, but you guys made a deliberate decision to get away from being a Facebook game, which is, I would say, not precisely conventional wisdom at the moment.

SB: I'm trying to remember where it was. Some site, one of the comments was like, "This game looks really cool, and I hope they succeed, but I think they're doomed because you have to be on Facebook now. Facebook is such a juggernaut, blah blah blah."

Which seems crazy like me. I mean, just Angry Birds. Red Dead Redemption. I mean, there's a billion things that are going on that connect, that aren't on Facebook. Maybe it will be more difficult to be on the web and be successful than on Facebook. I don't think so.

That's what people think. They think it's very hard to get people to come to something that isn't Facebook, because they're already there.

SB: Yeah. Yeah. But there's 500 million people using Facebook, and Facebook is a quarter of all U.S. internet traffic. It doesn't account for three quarters of all U.S. internet traffic. There's plenty of time that people spend on the web that isn't on Facebook. That maybe isn't true for everyone. Maybe some people...

For some people, it's like they go to Facebook, they go to ESPN, and they go to Hotmail, whatever, I'm just making it up, and those are the only websites that they'll ever go to. They're probably not going to be people who are going to like this game anyway.

We get asked all the time about demographics, and I can't answer. I did answer once, and it blew back in my face, in Rock Paper Shotgun. I said "people in their 20s and 30s, above average intelligence," or something like that. And then all the people who were commenting on the blog were in their 20s and 30s and probably above average intelligence, and no one likes being marketed to.

No one likes feeling like they're a demographic, but it's not like we had that in mind. We didn't think like, "Let's set out to design a game for this particular..." We just want to do something that we think is awesome, and hopefully there will be enough people who will also think it's awesome.

People who come from a web background who end up doing games sort of look at gameplay mechanics as sort of a scalpel to remove people from their money, right? They don't look at gameplay mechanics as an intrinsically valuable. Of course, that's starting to change.

SB: Right. I do not look at them that way. When we started at Game Neverending eight years ago, the inspiration was Homo Ludens, and the whole idea of the value of play as an element of our culture in the broadest sense. I mean, in the broadest sense like flirting, or witty banter.

I don't play golf -- maybe I'm down on golf players sometimes. I can imagine that it's really nice to go for a walk with three friends in a wooded area with the pretext of a game to keep you structured to it. Poker, playing music.

I remember, I was giving a talk on player creativity in 2003 or something like that, and on the way to the place where I was giving the talk, I was walking past the elementary school, and the recess bell rings, and all the kids pour out of the school and immediately start with something, right.

Like they're playing cops and robbers or they're playing like a structured game like four square, tetherball, or something. Or they're just chasing each other, to see how fast they can run. They're playing like I'll be mommy, you be daddy, playing with identity roles. For kids, I think, most learning comes from play, most really learning who they are. They play at the edge of their physical ability, their sexuality, their relationship with other people and how you're supposed to interact with them.

The most satisfying parts of my life as an adult are usually about play in some form, and that's what game mechanics are for. Because if people like the game a lot, they will pay us money. Some percentage of them will subscribe, some percentage of them will buy virtual items, some percentage of them won't pay us anything, but they'll enjoy the game and ideally be good players so that the game is funner for other people that are playing.

You talk about meaningful social interaction. In most "social games", the social interaction is fairly limited. Obviously it's going to be limited in certain ways that are intrinsic to whatever the limitations in this game are, but it sounds like you want more. 

SB: Yeah. And part of that is allowing group-based dynamics at all, like real ones. Like you're able to form a corporation or you're able to form a religion. When I say "religion", there will be a structured way of representing that.

When talking to you, pay-versus-don't-pay obviously makes me think of Flickr. I pay for Flickr. You don't have to to use it, but I do, partially because I like it and thus I want to give it money. Partially, it's the features. It's partially so I can have sets. It's both.

SB: For subscribers, you'll be able to have multiple characters per account, which you can't do with a free account. We'll give them more choices of skin color, a few things like that.

It's not like the game is crippled if you don't pay. I think that's enough. I mean, we'll find the right balance, but I think it's enough that the people who actually enjoy the game will say, "It's not an unreasonable amount of money."

There's some things that we talked about that really in an indirect way would affect game mechanics because you paid money.

So, one example is in-game advertising, which we'd never do for anything outside the game, but in the context of the game, buying a literal billboard space in a busy transit hub for something in the game, like you just started a new cult and you want to evangelize, or you opened to a store and you want people to go there or whatever, people pay money to do that and will get the attention to other players, which is what the feature is.

But if you're clever, you can turn other players' attention to game currency. That's like the closest we've considered, where you pay money and it has an effect on actual gameplay. The rest is just aesthetic.

This is like the unholy marriage of Animal Crossing and EVE Online.

SB: Yeah. That's actually a very good way [of describing it.] LittleBigPlanet is obviously an inspiration as well, in the aesthetics. I wish that we had a PS3 underneath this and that we're a lot better on 3D. But EVE, MOOs, and Animal Crossing have a cult following [here].

There's a couple of other ones, too. I guess the original... The early Sims, not that it's exactly the same thing, but just that idea. Yeah. There's a lot of them, but those probably are the two best. 

Well, I've never played EVE before. I never got into it because it just seemed too hard to me. It's my favorite game to read about.

Most games are boring to play and boring to read about. I'm not sure if EVE's boring to play; it's just an investment I don't want to make. But it's fascinating to read about.

SB: And I've always imagined, as well, that while the fights can be exciting and it can be cool, you know, to have victory in one of the fights, it's not really what it's about. I mean, people are playing the game to create the world. They're part of the corporations because they're buying into the agenda, even if it's roleplaying, against some other agenda. That's where the fun is.

In EVE, you might have a very small role, but you know you're a link in a bigger chain.

SB: Yeah. And you get to work it with other people and actually change how the world unfolds, which is something that I think every little kid who played games always wanted to see what would happen, like if it was just not open-ended in a Grand Theft Auto way because that's not really open-ended at all...

I mean, the people just walk like this, and you can bump into them and stuff like that, and 98 percent of the buildings are just hollow, textured shapes. If you could really change it, if there's a potential that you could fuck it up, if there's a potential you could create things that people haven't thought of yet, then it's a lot more interesting.

You're right; it's going to be a challenge communicating what this game is to people. Well, you've got two really big challenges, which are, A, making a game that does that, which is non-trivial, and B, communicating to people that they can do this.

SB: Yeah. Maybe we'll stumble upon a way of communicating it. More likely players will stumble upon a way. It's interesting to see what people say about it like, just reading Twitter where people are mentioning it and stuff like that, reading reviews and comments about what impressions they get.

We never had that with Flickr either, by the way, you know. Four years after we started it, we had to come up with a vision statement for Yahoo execs, and that was "eyes of the world", and that was pretty good. But we never had it at the beginning.

We could never explain why it was better than Shutterfly to people. Oh, we could, but it takes 15 minutes. We needed the five-second version. And we were successful despite that. So, hopefully either we'll come up with something good in the next six months or someone else will, or we'll suffer from not being able to explain it in 30 seconds.