Jonathan Blow interview – the father of indie gaming on video games in 2525 AD

GameCentral speaks to the creator of Braid about his new game Order Of The Sinking Star, working from home, AI, and future dystopias.

The Game Awards last week had many surprises but one we knew about beforehand was Order Of The Sinking Star from Braid and The Witness creator Jonanthan Blow. By the time the game comes out next year it will be his first game in 10 years, which is both exciting and a reminder of how few games he’s actually made – despite being one of the founding figures of the modern indie games scene.

Order Of The Sinking Star is a complex, narrative-based puzzle game, reminiscent of everything from The Legend Of Zelda and Chip’s Challenge to more modern titles like Cocoon. Or at least that’s the impression we got, since we haven’t played it ourselves yet.

However, we did have the chance to speak to Blow about not just the game but his role in the early days of indie gaming, the current state of the industry – for both smaller companies and giant publishers – and whether or not he uses AI to make his games.

GC:  I understand you don’t actually like to be called an indie developer, which seems surprising as I think many would consider you one of the founders of the whole movement.

JB: I mean, I was definitely there when indie games were taking off. The games that I’ve made are considered significant indie games, in some way or another. I think when I said that, what I was sort of reacting to was just… back at that time, so in around 2009 through 2011 was when indie games really took off. Because there had been, back in the Braid era, there was Castle Crashers, there was N+.

There were all these developers who had successful games in ’08/’09 and then it showed everybody else that this was a thing, and then this nascent indie community kind of took off and exploded. But it kind of became more about… it’s not like computer game nerds can ever be that cool. But it became more about being cool ’cause you’re an indie, as opposed to… for me it’s always been about look, you have a job, you’re telling people that they should spend their valuable time playing your game and spend money on your game, ’cause you’re trying to make a living doing this.

And so actually being an indie developer should have this very strong work ethic attached and this strong sense of craft. And some people have that for sure. But that whole scene that blew up around that time was much more just about identifying as an indie developer. And I think that’s what I was saying there, is I didn’t like any of that stuff.

Since then, all that has kind of past. There’s way more indie developers than ever before, but I don’t think there’s this weird self-congratulation about it anymore, in part because it’s so hard out there, right? If you make a random indie game and put it on Steam, you’ll get 10 reviews from players and you know… so I think all that has passed.

GC: There were a lot more indie celebrities back in the day, which you’re right, you don’t really get anymore.

JB: It’s because there’s just so many more developers right now. You will get indie celebrities if somebody has a super hit game, you know? What’s a good example? I mean, Balatro was a super hit game…

GC: But that was one guy and I don’t know that I’ve ever seen a photo or video of him, or seen an interview…

JB: Yeah, I just thought of a counter example, but people who are a little more famous, maybe because their games were a little earlier, is the Undertale guy.

But I think also the internet is also just a lot more noisy, right? People have enough things to do, like waste time on social media all day, that they don’t need indie celebrities anymore.

GC: Is this something like what you imagined the indie scene would develop into? The idea that there’s a lot of well-worn formulas, like Metroidvanias and roguelikes, and, to be frank, an awful lot of slop in amongst the truly great stuff.

JB: Yeah, I would say I kind of predicted it and kind of didn’t predict it. The things that I didn’t foresee is that at the same time triple-A style games would diminish in importance. When I got into the industry, or even 10 years ago, if you were thinking about what games are coming out this year it would mostly be triple-A games and there would be some indie games in there. And there were smaller indie games that wouldn’t make that list and nobody really thought about them then.

And now like most games are technically indie games in some way but, as you say, most of them are kind of slop. Or maybe that’s mean to say but most of them are just, from a player’s perspective… most players are not interested in most of the games. But I would’ve thought that that meant that triple-A games would still be big. But I think the problem is there’s just not enough triple-A games anymore, because they’re so hard to make and they’re so expensive that there’s kind of this vacuum that nobody’s filling.

Triple-A games fill them once in a while, and then hit indie games fill them once in a while, but I think a lot of the time there’s just no game to satisfy people.

GC: I agree. It’s so strange how halfway through this generation Sony and Microsoft, and others, suddenly seem to realise that games were costing too much and taking too long. How could something like that possibly creep up on you?

JB: Well… it can happen. I will say, obviously, I don’t work for any of those bigger companies. I don’t have an inside view, but we had our own smaller versions of this problem, you know. So on the one hand this game is tremendously big, and so you would expect it to take a long time. On the other hand, it took longer than I thought and was more expensive than I thought.

And I think there’s a bunch of reasons. One of them is, there’s been a lot of shift to remote work. I personally don’t think remote work works as well and is not as efficient. There are advantages. So as a small company, we are able to hire people, who are very skilled, who don’t live in the United States, and that’s normal now.

And I’m thankful that we can do that, because we make our own engine and our own programming language. [laughs] And so we can’t just hire regular gameplay programmers. We need people who can work on engines and so forth. So there are advantages, and I’m not fully complaining about remote work, but at the same time it is even… I’m not talking about people silent quitting, necessarily, although that’s a problem.

Like, in the old days when you were working on a game with people and there would be this magic moment where somebody gets something really cool on the screen and everybody comes up and looks at it and they’re like, ‘Wow, that’s what we’re doing!’ And they’re motivated now and you get this energy and it’s just hard to get that.

And so that’s part of it. But I think, especially as a company gets bigger and bigger and people are less invested on average, you know, at a big company, I think they’re working less at home. And you know, even if you sort of do these two days in the office thing… so if people are 50% as productive you would expect cost to double, right?

At least. Actually more, because then you need to coordinate more people to get the same thing done. I think there’s other stuff too. I think since Covid, somehow, people are just a little bit demoralised. They don’t expect society to work in the same way and so they don’t… they don’t take social roles and work ethics as seriously as they once did, broadly speaking. And that’s not true for everybody, right? Obviously, everybody’s different.

But just on a population level, I think that’s a little bit true. And then thirdly, having nothing to do with any of that stuff, I think just as the internet has progressed… like the internet’s been very different. It keeps changing over its arc, right? In the early days of the internet, it was like, ‘Come read my blog’ and people would read your blog and stuff.

And then it went through this period where everything was centralised into a few small places, but it was still kind of alive. Now you have AI results pre-empting the reason that you would go to a webpage in the first place. You have just all these social media outlets where people spend their time.

And so I think less of the population’s time and energy is interested in games, and looking at games, because it’s going into these other places. You even see that in stuff like Twitch. Twitch is still the main streaming place for games. There’s competitors, but it’s the main one. And I feel like Twitch peaked like five years ago.

It’s still there, it’s still alive, but viewerships are like half of what it was. And so I think our attention is just getting diffused a little bit. Interestingly, that is part of the plot of this game, as it’s a complicated game with a complicated story. ‘Cause there’s like six different worlds. But one of the worlds, the one that built this facility, with the overworld that the queen is wandering around and stuff, is contending with this issue that in the even further future, where we’re post scarcity and you can just manufacture matter. Like in a Star Trek way, you can just create food with a transporter beam or something.

What do you live for and do people have a common purpose anymore and do they even talk to each other that much anymore? So it’s been an interesting… it’s not an idle question, it’s been an interesting thing for me to think about.

GC: The game seems to have a cod-medieval setting but it sounds like you’re actually talking about modern society. Which is interesting because it’s changed drastically in just the last two or three years.

JB: So, whenever I make games they are, at some level, about whatever issues I’m contending with. But I don’t quite do it in the normal way. There is this normal societal thing that we understand when someone does art, if it’s a game or a movie or whatever, that there’s like a social commentary aspect to it.

And I try to not dive fully into that because I do think it dates a work a little bit. Whatever particular thing I’m thinking about, is anyone gonna care about that in 20 years? Maybe, maybe not. Right? When I make games, I’m always trying to think about, ‘Is anyone gonna want to play this 20 years from now?’ And that seems to be successful so far. Although we’re not even 20 years from Braid yet, we’re getting close.

But there is this aspect of, they have to be issues that are important to me and I am also embedded in our society. And so that trickles over into these societal issues. And then I think when you’re making a game like this, that has a lot of characters and different settings and stuff, those characters need to have motivations and they need to be doing something interesting and that needs to come from somewhere.

Part of the hypothesis of this game is that there is some far future society that is us. It’s us like 500 years from now. But whenever you’re doing science fiction like that, you can’t be too serious about trying to predict the future because you don’t know.

It’s a way of dealing with some of these ideas. What if in the future it goes this way and this kind of issue becomes prominent? And how are those people dealing with that?

GC: I’ve said this to a few developers recently, when dealing with games that deal with society at large and predictions of the future, and it’s how all the science fiction I read in my youth – all those warnings of a future dystopia have been completely ignored. More than that they’ve actually been accelerated towards.

JB: [laughs]

GC: You get people saying oh, wouldn’t it be great to live in the world of Blade Runner or whatever. No! It would be awful; didn’t you see the film?

JB: [laughs] Even ignoring societal level issues, with small functional issues we’re not even heeding the warnings. So we’ve had all these cyberpunk novels about how cool is it that if you could hack anyone’s computer and crash their car or whatever. And obviously you don’t want that if you’re driving a car.

And then, like 15 years ago at this point or something, there was a relatively big publicity stunt where somebody hacked into a Honda or something. That was one of the first, more computerised ones and it was a reporter’s Honda and they made it stop. This was all planned in advance, it wasn’t by surprise, but it made the car stop on the highway.

And when that happened I was like, ‘Okay, thank god this has brought attention to itself and now we won’t do this. We won’t be making our cars work this way.’ And it just went in complete opposite direction. [laughs]

So, I don’t know. Maybe we just can’t learn. Or we can’t learn pre-emptively. Unless we touch a really hot stove and get burned we can’t learn.

GC: What is your position on, on AI? Because talking of societal issues, that is a massive one at the moment that almost seems to have come out of nowhere. And gaming is affected by it a great deal.

JB: To answer the more immediate version of that question… ’cause again, who can predict the further future? But right now, I would say simultaneously the current AI that we have is indeed a legitimate, large technical breakthrough, in the sense that if you look into the math and all these things of what’s actually happening, it legitimately is stuff that we had no idea how to do even several years ago. And that’s great.

At the same time, it can’t really do most of these things that it’s being hyped up as doing. Even image generation, right? It can generate cool images that look neat, but if you actually try getting it to generate the actual image that you actually want, it’s incredibly difficult and at some point you have to give up and just settle for what it gives you.

And that’s true for all these things. For programming, especially, there’s all this hype about, ‘By the end of 2025, AI will replace all programmers.’ And that’s so far from true it’s not even funny, because if you’re an experienced programmer you can see that it doesn’t actually know how to program at all.

It’s just generating output that looks like programming to people who are not that deep into it. I think that’s true for all of these creative aspects of current AI. So I would not use it for most things. Now, that said, when we’re making games there are legitimate uses. So for this game we’ve got a little bit of a workflow set up where I can try out scripts – ’cause there’s a lot of dialogue in the game – if I want to try some dialogue.

One of the things that happens when you work with actors is you write something that you think is really cool and then the actor reads it and you’re like, ‘Oh yeah, that’s way too many words for this situation or that one word really sticks on me’. But when you’re there in the performance session… you can try to fix it, but it’s not necessarily the highest quality fix ’cause you didn’t have time to ruminate on it.

But with AI temp voice, which I think for the most part does not sound very… it’s not remotely as good as an actual actor. But, if I want to get a draft into the game and hear the words as I play the game, it’s actually really helpful. And so our workflow now is that we’ll prototype everything with AI voice and then we’ll, of course, pay real actors to actually act for real for the final game.

And I think that’s a valuable process. And so there are things like that. There are legitimate uses, but at the same time there’s also way too much hype.

GC: You touch on an issue I worry about a lot, which is how people are willing to put up with a major drop in quality with almost no complaint, especially if it involves a small increase in convenience. I’m thinking of a big name game from a major publisher I played recently and the voice acting was awful, just because they were being cheap, but then you go down a rung further and things like Arc Raiders, which has plenty of money, is using AI voices.

JB: There’s a weird way in which games are still kind of infantilised, right? So I won’t name the game and I won’t name the developer, but there are quite large game companies that, by habit, with real actors who are talented, they produce the most horrendous voice-acting because they do… especially weird stuff with accents, where they have people who are supposed to be in some far away country and they speak really slow.

Like they’re really stupid and they can’t speak their own language, because they’re trying to do accents. And it’s the worst thing ever. And it’s like… just watch a good play or a good movie where people don’t do that and see how the acting works. And somehow in the games industry, it’s almost like we’re past the peak of that.

Early in games we couldn’t do high quality cinematics. And then we started doing them and they were very cheesy compared to film or something, and then they got better and better. And then at some point they were actually pretty good, for a game. And now, in the same way, as you say, that all these projects are getting more expensive and maybe failing to reach gamers in the same way… even specific things like acting in games is worse now than it was five or six years ago.

GC: I think that’s true but there’s also… there’s a certain charm, in many cases, to games not being good at the things that are not directly related to gameplay. Resident Evil is the obvious example, where the originals are beloved because the dialogue and acting are awful, it’s what gives it so much of its character. But they struggle to replicate that in modern games because how do you decide how bad is good enough? [laughs]

JB: Well, that’s a weird problem that I’ve never had to face. [laugh] But yeah, I can see that.

GC: Seeing your game, I can see you’ve obviously been influenced by old school Zelda. But a developer that hadn’t played those games, they wouldn’t make something like this, with the tile floors and the single screen areas. But how long do those classic affections last? If you want to talk 500 years in the future, what does a video game become then? What’s the holodeck equivalent of a single screen room with tile-based movement? Do video games as we know them just cease to exist at that point?

JB: That is a little bit of the point, approached by the fiction of this game, but I will also say, if you are interested, you can look up a little five minute rant that a friend of mine did at the Game Developers Conference a long time ago. He was very visionary and could foresee these things. It’s called the immersive fallacy, and it was about how all these game designers of that time, when graphics were still worse than they are now, all they envisioned was like simulating reality exactly.

And he has this rant that’s very funny and also very right on about why that’s not really what you wanna be trying for. And so I think that’s interesting.

GC: So over the last nine years, have you literally just been sitting there with pieces of graph paper, designing this game? Did you want to spend 10 years making it?

JB: Well, it, it hasn’t… so, The Witness came out January 26th, 2016, I believe is the right date, beginning of 2016. So we’re not quite, we’re at nine years, 10 and a half months or something. But by the time this game comes out it will have been more than 10 years. And yes, I’ve been working very hard on it.

I did some other things. So we made this game engine as well. We made the programming language that the game engine is written in as well. We also released the anniversary edition of Braid. So it’s not like it’s the only thing we did over that time. But, yes, it’s been a period of very intensive work on this game. Not necessarily with graph paper most of the time, ’cause when we design puzzles we have an in-game editor that we pop open. [laughs]

GC: I remember talking to a developer about eight years ago – so things will have got even worse now – but he was maybe late 30s/early 40s and he said I’ve got my next four games planned out and then I’m going to retire, which was kind of shocking to me. How many games are you ever going to have time to make in your lifetime? Because so many now are only doing one per generation.

JB: Isn’t it funny? You look back to the 1980s and Sid Meier released all these games between 1986 and 1990 or whatever. And it’s just not like that now. This is a question I think about all the time. Part of the reason why I wanted to make a programming language, for example, is to help accelerate the programming part, because modern games are so complicated, but there’s a lot of parts to deal with and, of course, it took a time investment to make the programming language.

But yeah, I do ask myself this question: how many games am I gonna be able to make? I know what the next, at least, two games are. [laughs] So we’ll at least make those. We’ll hope that AI can invent some medical interventions for us and we can live a little longer.

GC: [laughs]

JB: But yeah, it’s a very real question and it’s a very real observation.

GC: Alright, well thank you very much for your time. It’s been a pleasure to meet you.

JB: Yeah, thank you. Take care.

Email [email protected], leave a comment below, follow us on Twitter.

To submit Inbox letters and Reader’s Features more easily, without the need to send an email, just use our Submit Stuff page here.

For more stories like this, check our Gaming page.

Comments Add as preferred source