IndieGames Blog | Best of Features | Interviews | Reviews


Interview: Jonathan Blow (Braid)

DECEMBER 23, 2007 - Article by Eden Bradfield


I interviewed Jon Blow, creator of upcoming XBLA (and PC eventually!) game Braid, and found him to be an wonderfully nice and insightful fellow.


EB: How do you go about designing a level? In Braid, the levels are sort of no-nonsense. They're very efficient.

The game is about understanding what it means if time behaves in certain ways -- exploring the consequences of these hypothetical laws of spacetime, and the puzzle pieces you collect are concrete tokens representing the understanding you have gained. So, every puzzle in Braid has a very specific point; it is there to tell you one thing.

Because I wanted the game to be focused (and not long and bloated with filler), I decided early on that the levels would only contain the elements necessary to create the puzzles inside them. There aren't lots of random enemies to jump on or big levels to just sort of wander through. Everything is in the game for a specific reason. After being in the game for a while, the player might start to pick up on some of the nuances (why certain puzzles are grouped together, for example).

What I've said here only applies to Braid, though; if designing a different game, I would probably take a different approach.


EB: Define "drug", in terms of games. (WoW, etc)

Games like WoW are very deliberately designed to dole out rewards in order to keep players glued to the game. (It's not just MMOs that do this; for example, the Diablo games do the same thing, and so does Peggle, though in a simpler way). These tend to be very similar to the systems modeled by B.F. Skinner; anyone interested in this might start by reading the Wikipedia article on "operant conditioning".

I claim that some rewards are natural (you automatically make yourself feel good for having accomplished something; or you received acknowledgement for building a skill you didn't have before you played the game; or else you learned or experienced things that will make your life richer) and some rewards are artificial (the game tells you that you are doing a good job, gives you lots of eye/ear candy to celebrate your progress, or pretends to give you valuable things when in fact it is giving you nothing).

I think of natural rewards as food, and artificial rewards as drugs.

All games provide some mixture of natural and artificial rewards. My problem with games like WoW is that they take a huge amount of time and give you only artificial / empty rewards. In other words, they grab players' attention for long periods, but don't feed them much. I think that's a very bad thing, when you look at how many people are playing these games, for so many hours. With Braid I very deliberately tried to base the game design on natural rewards, using artificial rewards sparingly and only when they support natural ones.


EB: So in your eyes, natural rewards are sort of about "self improvement"- even if this improvement is only related to the game itself. Perhaps why this improvement is only related to the game itself is because games are so detached from reality (although Braid appears not to be from what I've played).

It doesn't necessarily have to be self-improvement, but that is a major aspect. Much of it is evolutionarily grounded in some way, even if it's subtle. For example, you look at dogs, they like to play-fight with each other. Why do they do that? Well, there are a lot of reasons, but some of them have to do with keeping healthy and physically fit, and practice-fighting so that they are better off when they get into a real fight. (But of course there are social aspects too, and others... it's complicated).


EB: In your MIGS talk, you said "Games are trying to achieve a goal". Do you think that we really need to try an achieve a goal? Doesn't this limit the sort of games we can make? The Endless Forest is pretty much goalless, unless you say that there's "implied goals". Nevertheless, it doesn't really have any stated goals.

The Endless Forest is not a game, by my way of thinking. It's an "electronic interactive experience" or something like that. I don't think that's bad, it's just different. It's much harder to make something like that, that will hold the player's interest, than it is to make a game. That's why you see relatively few of these things.

I fully agree, though, that we need to expand our notion of what games are. When I go give a lecture like the recent one in Montreal, guys on message boards start posting things like "that guy is dumb, because if you do what he says then games won't be fun any more, and the whole point of games is to be fun". Well, I actually disagree with that statement from several different directions. For one, I think that the majority of games that people feel compelled to play are not "fun". Counter-Strike is one of my favorite games, and it is not fun at all, except on rare occasions. Raiding in WoW is not "fun". They are gameplay experiences that people want to have for other, more-complicated reasons.

But also, I think it's important to get rid of the idea that a game should be "fun" or even compelling. Ian Bogost wrote an essay about this recently and I agree with him completely. The problem is that the word "game" is so invested with prior connotations; we need to come up with a new name for what this medium is. But nobody's really come up with a good word for it yet, so I end up saying things like "electronic interactive experience".


EB: Yeah, I agree with you on the word "game". All the other suitors sound kind of pretentious, though. There's a quote that goes something like "to teach them, you must entertain them." What do you think?

I do think there is something to this. So many "Edu-Games" are so terrible because they are bad games. I spoke before about how natural rewards are evolutionarily grounded. Our minds tend to reward us for doing things that have survival value to ourselves or our society. Thus if we are doing things that are really interesting, we ought to be entertained, automatically. Raph Koster's book "A Theory of Fun for Game Design" has something to say about this.

Somehow we have developed this educational system that manages to be rote and boring, that somehow leeches all of the entertainment out of learning. And then we try to artifically restore it via stuff like edu-tainment. Well, instead, why not figure out where we dropped the ball in the first place, and just make the learning itself interesting again? I bet that would work a lot better.


EB: You mention that a game (or interactive realtime electronic experience if you want) doesn't even have to be compelling- what's the rationale behind that?

All the time in everyday life I have experiences that were not "compelling" but that I am glad I had. "Compelling" is a super-high bar to try and reach, and if we only shoot for that, we are missing a Long Tail of value.

An experience that is just Mildly Interesting can be pretty good, if it only takes 15 minutes of our time. Our mistake is that our idea of a game is that it takes 8-80 hours to play... but we don't know how to put real material into a game to make it interesting for that long, which is why we pump them full of artificial rewards. What if the average game cost $1 and took 30 minutes to play and was interesting and fresh for the entire 30 minutes? Wouldn't that be nice?


EB: What do you think about the Game Grammar/Game Chemistry crowd? Are they trying to reverse-engineer art?

I don't pay much attention to schemes like this, because they are fundamentally unappealing to me. They always seem like they start by wanting to establish a formal theory, then ignore 75% of what makes games interesting in order to come up with a simple-enough picture to admit any kind of theory at all; then they kind of make up the theory and in the course of bending that 25% to fit the theory they've made, they lose most of that too, so at the end maybe you have 9% of games.

That's fine, but it bears almost no relation to anything I experience on a daily basis.

(Raph Koster is a friend of mine, and I say good things about his books pretty often. But I am really not into the Game Grammar thing.)

I like Formal Abstract Design Tools better, because it's not trying to be a closed universe. It says, "hey, we ought to have a language for talking about game design; if you want to make a formal description of a tool and plug it in here, hey, that's great and useful." It does not say "Here is a structure that describes all games!!!" which is what Game Grammar, Game Chemistry, etc try to do.

I have my own game design viewpoint that I have never formally written up, which I call Loop-Oriented Design. It's not really about trying to classify games; it's more about just trying to highlight gaps in a design, to find things that are missing (and which are not obvious because it's a lot harder to see something that isn't there than something that is).


EB: I'm glad you don't pay attention to game chemistry - There's no solution because there is no problem. Can you explain a bit about Loop-Oriented Design?

It's really a long topic that I can't go into at sufficient detail here. But, I think of it as a "reality-based" design approach, in that it starts with the basic core of what a game really is: it's a program that runs in a loop, from Input to Simulation to Rendering, then back to Input again. But then the player is in that loop too, and you can make up some boxes that stand for the different stages that go on in the player's perception/action cycle.

Then you can say that an event happens in the game world -- starting at Simulation -- and then start tracing that event through this loop, seeing what the effects are as it propagates through the different boxes. If things are designed well (and have good "game feel", as Steve Swink put it), then this loop will go continuously forever (though the effects of your event might shrink asymptotically to zero). Maybe you'll realize that there is a piece of hidden game state that's important but that is never rendered via graphics or sound, so the player has to guess at it... which makes your game feel bad.

This can be applied at the low-level (game feel) or the high level (long-term goals, story, etc) or any blend of those. However, it doesn't really approach the art question. It doesn't try to talk about how to make a game meaningful or important. It's just to help look for mechanical problems.


EB: A couple of people have complained about Braid's main character. Any reason he's designed the way he is?

Yes, he's designed that way because he's the right character for this game. He adds something significant to the mood and the themes. The reason why is delicate and subtle, but I think upon playing all the way through, at least some players will kind-of understand. If people want to be a talking wombat, or a guy that carries 17 guns, they can go play a different game.

Games are already so full of bullshit escapism that sometimes it's hard for me to take the medium seriously any more. Some days I just look at the games that are out there and think, "Why am I devoting my life to this again? Surely I can come up with something better to do, like go help some kids in Iraq or Somalia stay alive."

If games are ever going to be a medium that is intellectually and emotionally important to people -- not just otaku man-children, but, like, well-respected and powerful people too -- then they need to do something besides escapism. This means there has to be room for games where the lead character is not a 7-foot-tall guy in power armor, or a chick with guns whose tits are bigger than her head.

So if anyone is dissatisfied with the main character in Braid, I would just encourage them to think about what they are asking for, and why.


EB: Why do you think game developers shy away from commenting on reality? From making games that aren't just sheer escapism? I see this in the indie scene too- there's very few games that aren't just escapism.

It's easier to make something that is escapism, than something that isn't. It's easier to make something flashy and empty than something deep and meaningful. And the more we make flashy but empty things, the easier that becomes, since we build those kinds of audience manipulation skills, and new generations of designers grow up with just the flashy empty stuff as their main model of what a game is... so it becomes hard to imagine something outside that box.

Most Hollywood movies are empty escapism, too. But the difference is, movies have shown that they are able to affect people in a way suitable to what we might call great art. So you can have all these lousy Hollywood action flicks, but there is this foundation that something more meaningful is available, and people know where to go if they want to look for it (even if they do that much less frequently than watching Hollywood flicks). Games have not established that foundation. Until we do, every escapist game is one more nail in the coffin of comic-book-style nerd ghettoization... too much of this, and then even if someone makes a very meaningful or important game, society won't pay much attention, because games are just about pretending you have a giant penis. When a typical American thinks of comic books today, he doesn't think of Maus; he thinks of Spider-Man or something like that.


EB: What would you call "great art"? Something can affect people a lot, and not be artistic.

Well, sure. An AC-130 gunship can affect people a lot, and that's usually not art. Art is one avenue that we have by which to affect people and the world. I'm really not going to get into the business of defining art here. When creating things, it's something you either feel or you don't.


EB: What's wrong with escapism exactly, in your view?

There's nothing wrong with a little bit of escapism. Sometimes people are very sad or depressed, they have a hard time dealing with some event in their life, or with life in general. Escapism in these cases can be useful therapy.

But it's like ice cream. If you eat ice cream for every meal of every day, you are going to be a very sickly person. If you're a little kid, maybe you only think ice cream and candy taste good. But actually, steak tastes good. Broccoli and cauliflower taste good. Some people even think haggis and gefilte fish are edible!

All the time we spend escaping detracts from the available time we have to build richer lives for ourselves and to make our contributions to the world. This is extra-nefarious because it's so hard to see -- it's harder to see something that is missing but that would be beneficial.

How did Bill Gates get to be the richest man on Earth? How did Mother Theresa or Mohandas Ghandi come to be the well-respected historical figures they are? How did Albert Einstein discover the Theory of Relativity? I guarantee you it wasn't by playing World of Warcraft.


EB: You say "and new generations of designers grow up with just the flashy empty stuff as their main model of what a game is". Isn't it a fallacy to base your model of what a game is on previous games anyway?

Well, it's not a good idea, but that's how people are. We build our mental models of the world from the things we have seen and experienced. It is often very difficult to break away from those preconceptions and do something new.


EB: It's just that I think if we learnt design from games, and used that as our model, we would end up with a really narrow possibility space of games.

That's exactly where I think we are right now. I am not saying it's a good thing, I am saying it's what happens.


EB: Shouldn't we take our models from Media as a collective whole, from life, from the world around us?

That's one way to do it. But it's hard to even have the mindset to do that. Certainly, I don't think the indie community has a very good track record of that, though there have been some recent works doing very interesting things (like Rod Humble's "The Marriage" and Jason Rohrer's "Passage").


EB: Why do you devote your life to games? To make the life of someone a little better?

I have deeply ingrained impulses and drives that I don't really understand. And I don't think that understanding them is necessarily possible or would be good, because there's a limit to how profound straightforward reasoning can be.

In part, I work on games because I see a lot of potential in where they can go, and how they can shape the future. But that is kind of a rationalization. I can't tell you the real reason because I don't exactly know, but I do feel that it's important.


EB: Maybe we shouldn't take games too seriously. The first Katamari game (apparently, it was never published in PAL format, but Tim Rogers seems to think the following is true) is quite an emotional experience, and had a strong message. Yet the creator- Keita Takahashi, doesn't take games seriously at all.

I do think Katamari Damacy was a very strong and emotional game. In contrast, Beautiful Katamari is just embarrassing. I turned it off after about 20 minutes. It's just trying to mimic what came before, but it fails in every conceivable way.

I don't claim to know much about Mr. Takahashi's attitude toward games, but when you play the original Katamari... it's very sophisticated. All at once, it is a playful game, but it is also an ominous game of universal doom. It's very fun, but it's also deeply scary in a way that you can't quite put your finger on. So, at least the way I see it (and we know there is always a big difference between audience perception and authorial intent), the game has some deep emotions behind it. If Keita doesn't take games very seriously, then perhaps games were just the medium by which this kind of expression came out this time, and it will come out via some other medium next time. Who knows!


EB: "How did Bill Gates get to be the richest man on Earth? How did Mother Theresa or Mohandas Ghandi come to be the well-respected historical figures they are? How did Albert Einstein discover the Theory of Relativity? I guarantee you it wasn't by playing World of Warcraft." Of course. Critics would say: "if that's the case (which it is), shouldn't we just burn all our games and start trying to change the world?" Otherwise, you imply that games the people who play these games of the future will play them and somehow change the world or whatever. That's a beautiful thought.

That is exactly what I think. Games have the power to change the world; I wouldn't spend so much time designing them if I didn't believe that. If you can emotionally or intellectually impact someone at some point in their life, then every decision they ever make in their life after that, everything they think and feel, is influenced (in at least some small way) by that thing you did. Isn't that interesting?


EB: It's kind of scary really. Imagine how people who play WoW are affected. Scary and wonderful.

Well, that's one of the things that makes it feel important to me, is the sheer scale of what's going on. Things that are just fine in small quantities can be horrible and scary in mass quantities. And things that are just little and kind of insignificant on an individual scale, can become very powerful when multiplied like that.


EB: What advice would you give to someone wanting to be a game designer? Do they need to play a large amount of games? (The second question comes from an argument I've been having with another designer).

I think it is very helpful to play a large amount of games. For me this is a big part of how I formed my design sense -- what kinds of things work, what kinds don't, what I think is interesting or boring. It's possible to create something good without this, of course, but I think that would be flying blind and hoping to get lucky.

There are a lot of things I don't do because I have seen them not work well in other games, or else kind of work but not ultimately be that interesting. If I hadn't played those other games, how would I know not to do these things? I would think they were potentially fruitful ideas, and spend a lot of time working on them.


EB: What are some of the key design things that you learnt from Braid?

It's hard to summarize in a pat way. Braid is the best project I have ever been involved with, so I've gotten a lot out of it. But a lot of it is subtle and not conducive to being conveyed in words. I think if people play it (when it becomes available, sorry it's not out yet!) then some of the ideas will come across.



IndieGames Blog | Best of Features | Interviews | Reviews



About The IGF


UBM TechWeb (producer of Game Developer magazine, Gamasutra.com, and the Game Developers Conference) established the Independent Games Festival in 1998 to encourage innovation in game development and to recognize the best independent game developers.

The competition, now in its 13th year, awarded a total of nearly $50,000 in prizes to deserving indie creators in Main Competition and Student Competition categories at the IGF Awards Ceremony, held in March 2010 at the Moscone Center, San Francisco, CA, as one of the highlights of the Game Developers Conference.

The Independent Games Festival will return in March 2011 at GDC in San Francisco - look for more information soon. [More information...]

About The IGS


Featuring lectures, postmortems and roundtables from some of the most notable independent game creators around, the Independent Games Summit is a yearly event, with iterations thus far taking place in March 2007, February 2008, March 2009 and March 2010 at Game Developers Conference.

UBM TechWeb (which curates the IGF and runs GDC) plans to continue and grow the Independent Games Summit in subsequent years. [More information...]