Talking the Future of Minecraft

In December, Markus "Notch" Persson handed the reins of Minecraft development to Jens "Jeb" Bergensten, who became the lead developer so Notch could move onto another project. "[It's a] scary move, but I feel strangely confident," Persson told Gamasutra at the time. "I guess we've worked close enough for long enough for me to feel confident about it."

Gamasutra traveled to the Mojang offices in Stockholm in December to speak with Bergensten about his plans for the game -- a busy day for the developer, because in his other role, as the co-founder of Oxeye Game Studio, he had made the decision to launch Cobalt that day. Cobalt, still in alpha, is the first game from another developer Mojang has chosen to publish.

Still, Bergensten's primary focus is Minecraft development. In this interview, he explains what the future holds for the massively popular indie game, how he feels about taking control of the project from Notch, and what he expects to concentrate on as far as additions go.

You're the cofounder of the Cobalt developer, Oxeye, but at the same time you're in charge of Minecraft, so you must have a lot going on.

Jens Bergensten: Yeah, yeah. It's kind of busy. [laughs] I work full-time on Minecraft, and I work on Cobalt during evenings and weekends.

You took over for Notch, basically, is the way to put it?

JB: Yeah. He wanted to try new things. I took the design lead for Minecraft so he could have free hands to do other things.

How is that going to affect the game?

JB: That specifically won't affect the game very much because Notch and I have very similar game design philosophies and ideas. The things we are doing are quite similar.

But it's 1.0 now, and we are such a small team that we can't compete with the rest of the world with content. So, there's a change in priorities, that we really need to open up the game for other developers to add mods, and share mods, and run servers more easily. So, what I mean is I will work less on features, and more on the engine part of the game.

Is there a difference between someone building something versus just sharing it, and actually a team just working on something within the context of Minecraft? Have you seen that?

JB: You mean like the comparison of building stuff in Minecraft and adding stuff to Minecraft?

Yes.

JB: Yeah. There will definitely be much more content -- tools and blocks and mobs and stuff added to the game -- that other people can use in their constructions. There are already a lot of mods that add a lot of stuff, but they are quite complicated to install. So that's what we're trying to do.

Simplify that process, so more of the community can actually enjoy those things.

JB: Yeah. And also, Minecraft is a sandbox game, so people have very different opinions about what you're supposed to be able to do in the game. Like, some people really hate the adventure and RPG part of the game, and some people want more of that, more dragons and whatever.

Some people want more engineering tools. Some people hate engineering tools because they don't understand how Redstone works anyway. So, the good thing about mods is that then we can let people who really want to specialize on one part of the game, we can tell them, "Here's a really great mod. Just install it. You'll have fun."

Are there any plans to curate or point users to mods? Because right now, I guess, all of that stuff is out in the wild.

JB: Yeah.

Do you feel like it's part of your responsibility to point people towards things?

JB: Well, I don't feel I have the responsibility to do it, but I kind of envision that in the game, there will be featured mods, popular mods, and then you just click on them, and you can play them. In that way, we can put popular mods into focus.

You want it to be that simple.

JB: Yeah. Preferably.

Do you think there's a point where -- this is a really weird analogy, but it's sort of the best I can do -- you could enable a Counter-Strike of Minecraft, where it becomes a new game.

JB: Yeah. I mean, that would be really cool. But we would have to put a lot of effort on the mod API because that will have to allow the game to really change. So I'm not sure if you've heard of a game called Ace of Spades. It's a Minecraft game but with rifles, so it's kind of like Day of Defeat combined with Minecraft. And it would be really cool if you could actually do that without actually hacking the client. If everything just sort of started the game in that setting, and that worked. I mean, it will take us a while before we have reached that point, but it will be nice.

You guys tend to release often.

JB: Yeah.

Is incremental functionality something we're going to see, or are you going to save this until it's a little more done? Do you have to make these kinds of decisions?

JB: Yeah... We're used to doing things incrementally. I think that's probably how it will turn out. Sometimes we will have to say, "Sorry, modders. We broke this. You have to adjust your mods." Sometimes that will probably happen.

I think the initial mod release will take kind of a leap. There are already mod APIs that are based on modifying the JAR files. And I'm trying to get help from those communities to, already when we do the first version of the API, all the mods that are working on those [laughs] are working in the full game as well.

So, moving forward, your primary goal is this modding stuff.

JB: Yes, for the moment, it is.

I guess you got to Minecraft 1.0. Certainly it's not done.

JB: No.

There's no such thing as done, is there?

JB: No. It will never be done. But, as I said, it kind of feels silly of me to sit and work three days to add a new animal to the game when there are thousands of people who would like to spend three days to add an animal to the game, so that's why I changed my priorities.

There's also this fuzzy line. You're talking about working with people who've already figured out ways to mod the game. You've always had this really tight community focus for the game. Do you think that certain things will cross that line and become part of Minecraft?

JB: Yes. I am hoping to be able to use some parts of the community solutions. The problem is, first, there are several different competing community projects. So, all of them say they are the best, and they say they have the most supporters, and nd they don't really communicate with each other.

The second problem is that they are usually projects based off of loose individuals sitting all over the world. So, when I go to their leader and ask them in what kind of legal sense can we cooperate, they don't know. They don't know who owns the code. They don't know who is supposed to be compensated if we use their stuff.

We're looking to expand Mojang, so we're looking at talented game designers and programmers, so I also try to be like, "So, who's really contributing to this project?" And they're like, "No, no, all nine of us are contributing equally!" [laughs]

Yeah, there are some obstacles. I will continue to communicate to them, but in the end, I will probably use them more as inspiration than actually just copy/pasting it because of these legal problems.

Did you ever consider open sourcing at least parts of Minecraft so people could actually commit changes or anything like that? Or do you all want to centrally control it?

JB: Yeah. Actually, that would probably help us a lot if we did that. I think we're just afraid of what it means to open source parts of the code. We've been talking about it a lot, but I think nobody really dares to actually put it there because you can't take it back. [laughs] I mean, we intend to expand on Minecraft for a long time. I think probably we're safe, but just in case, it's easier to just sit on the project and keep it internal. Even though the whole source code is already available on the internet. It's just not officially open source.

Did you put it out there?

JB: No. People have de-obfuscated the code. Since it's Java. First, they created their own names for all the classes. I have cooperated with some of the mod teams, and I have given them the proper mappings for the class names. So now it's more or less identical to our code. That's more fine, because it's kind of beneficial. They know that it's still proprietary.

You have this tight connection with your community, and you have so much enthusiasm, and you're attracting so many people, that you're going into uncharted territory and making decisions that are probably tough.

JB: Yeah. [laughs]

Like [CEO] Carl [Manneh] was saying, you're trying to grow in a very controlled way as a company and not get out of bounds.

JB: Yeah.

I'm sure you have more that you want to do than you can do at the moment, right?

JB: Yeah. Yeah, but I think it's good. I don't think we should rush things. It would be nice just to add mostly three people to the team and no more than that, at least for a while. And three is a lot, I think. And people will have to have some patience. But yeah, there's also this, as I said, with the sandbox thing, people want different things. People also feel entitled to demand things.

On the internet? No.

JB: [laughs] The problem is it's impossible to please everybody because one of the things we've been talking about even since I started working here is we need support for different languages in Minecraft. And now finally in the last week, I started implementing that, and got a lot of crowd-sourced translation advice. And then I got a lot of angry messages. "Why are you wasting time on translations? What's the problem with English? Go back to add features!" [laughs]

You can't win.

JB: No, no. You can't. But I got a lot of happy emails as well. Sometimes you please that part, and sometimes you please other parts.

At least from the outside, Minecraft seems to evolve naturally, and it evolves in step with what the community wants. How far out do you plan?

JB: Usually we don't plan very far ahead. It's like one week at a time. We have general ideas of where we're heading. Like in this case, I know what I have to do mod the API, but I don't know when it will be ready, or when we'll plan to release it. I will just let it take its time. Also, a lot of the features that we've added during the last year have been in the morning, Notch just said, "Ah, I want to do a snowman," and then in the afternoon we have a snowman.

Do you have a development methodology like Scrum or Agile? Or do you just do things? Is it as simple as, "Notch says snowman, snowman happens."

JB: Yeah. So far it's been like that. It has only been me and Notch working on it. We started off using Kanban for the whole team. Yeah. It kind of fell apart. I know the other teams are still using Scrum. The Scrolls team is using Scrum. The web team is using Kanban. Since it was only Notch and me, we kind of discussed small pieces every day.

Sure. But things are changing, though, aren't they?

JB: Yes. So... When I'm working now, I'm kind of planning to release when I feel I have a full feel of the release. So, it's not that I work towards "When I've done this, I have a release." It's more like, "Okay, now I feel..."

Comfortable. You don't want to lose what makes Minecraft Minecraft, and part of what makes Minecraft Minecraft is that it feels right. You don't want to lose that creativity and that spontaneity.

JB: Right. But, I mean, I'm slightly more... I'm not sure of the word, but I think I'm more...

Methodical?

JB: Yeah, than compared to Notch. No, not methodical... I'm more hesitant to throw things in [laughs] ... maybe. Maybe that will change as I grow more comfortable being the lead, because previously, if I wanted to do something weird, I always felt like I must ask Notch first. That automatically created slightly more lag from idea to implementation.

I'm assuming you're still going to talk to him?

JB: Yeah. Yes, of course.

But you're making the decisions now, ultimately. Right?

JB: Yeah. I think that's also why I'm focusing more on non-gameplay content features. Like languages, mod API, and maybe friends lists when we get time to do that. And leaving the game a little bit like it was when it was released, only fixing bugs and doing smaller changes.

Just until you get comfortable with this role, or until you see a direction?

JB: I think it's more that there are a few things that are unfinished, I think. As I said, we planned one week ahead at a time, more or less, but we had a list of things that we wanted to include in the game, that we counted as "boring" stuff to add.

It feels like I don't have time just to fool around and add stuff just because I think it's fun. [laughs] But maybe it will change... Now after Christmas, we will have a new programmer starting who will work both on Scrolls and with me. I will ask him to improve the AI of animals. That's not an engine bit. That's a gameplay change. That will probably also make me feel like, "Oh, I have to do something in the game for fun as well." Yeah, I'm not sure.

Well, it's continuing in the spirit of what you've already achieved, right?

JB: Yeah.

The game drives the game.

JB: Yeah.

And the community drives the game, too.

JB: Yeah.

It's not so much the plan drives the game.

JB: No.

I mean, you have some goals. Obviously you have goals that you want to get to.

JB: Yeah.

It sounds like you have a list of things, too, that you know need attention.

JB: Yeah. Also we have the community is organizing all of the bugs lists. So sometimes I go there and fix those that feel relevant. Usually I fix those that seem easy to fix, more quick stuff.

How many programmers do you have working on it right now?

JB: Minecraft?

Yeah.

JB: It's only me.

Only you?

JB: Yeah.

And then one and a half after Christmas.

JB: Yeah.

You and the other person.

JB: Yeah. I mean, we have Aron [Nieminen]. He works on the [smartphone] Pocket Edition.

Right.

JB: And we have 4J in Scotland working on the Xbox version.

Right. I talked to Daniel [Kaplan] about those, but it's different. They're not working on the same codebase. Well, at some point it was the same codebase, but then it split, right?

JB: Yeah, exactly.

And their stuff isn't going to get committed to the PC version. It's going to stay split?

JB: Yes. it would be very difficult for us to merge those because it's not even the same programming language, but the Xbox team, they have made a lot of optimizations to make it work on a console. We probably want to get them into the main game as well just for performance reasons.

That sounds like a challenge in itself.

JB: I think they will have to describe the optimizations, and then I will have to re-implement them in the game, yes.

No wonder you don't have such a process-driven atmosphere here. It's just you. [laughs] I'm just so used to talking to game developers who have to have a certain way of working because it's not just a person. It's easy to forget how indie this is just because it blew up so much. Do you ever feel that way too? I mean, you can't, because your perspective is "I am the dude."

JB: I actually feel it's kind of pressuring. Like I said, we're trying to release Cobalt now, and we have some problems. On Cobalt, we've been mainly three people working on it, and I've been working on it maybe four hours per week, or something? And then we have one full-time developer and one part-time graphic artist.

So, we're really small. The game will probably not work for most people. There will be lots of bugs. We'll be kind of like, "Here's a game!" -- that kind of style. [laughs] We're releasing it, or [rather] giving alpha access to millions of interested people. ... It's not a triple-A product, but we have a triple-A audience, if you understand what I mean.

There's nothing wrong with the game, but we don't have a QA team, we don't have a marketing department, and so on. And also we don't have a legal department either. We have to ask lawyers when it's necessary, but I'm sure that there are things that we don't think of that big companies would have thought.

But that's the atmosphere that allows a game like Minecraft to thrive. It couldn't anywhere else.

JB: No.

Or at least not at a big company.

JB: No.

I can think of a million of reasons Minecraft could never happen in a big company.

JB: Hmm.

I'm sure they all want it to now, but I don't think they're going to allow the atmosphere that would make it possible.

JB: No, it would probably be a very different game if at a large company. But I'm kind of looking forward to bigger companies taking on the same kind of genre, if you should call it that. Because I think it could make quite cool games, you know, the sandbox environment.

One thing that I like about Mojang is that we're allowed to communicate directly to everybody. We don't really have many secrets. It's just sometimes we say that it's probably not a good idea not to talk about this yet. And then, we say, "Everybody, kay, it's fine."

I mean, this legal dispute with Bethesda, or actually Zenimax, I found some people that worked at Bethesda on Twitter, and I Tweeted them, "Hey, how's it going? What's the atmosphere at the office?" But I never got any replies. And I was thinking that maybe they have some policy that they aren't allowed to mention anything that's going on.

I'm sure they've been told not to talk about this.

JB: Yeah. And I don't see why it would hurt. [laughs]

I could tell you.

JB: Yeah?

It's because anything they say, you could then use it.

JB: Ahh, okay. Yeah.

And they don't want to get fired.

JB: Yeah. If everybody is allowed to say whatever they want, there could be a few people that don't say some clever stuff. [laughs] Of course. It's a big responsibility to put on your employees.

You hit 1.0 with Minecraft. Is audience expansion something you're concerned about? I know there have been gradual moves toward making the game more accessible, making the game more directed, at least in the beginning. Is that something that you find important? Or is that more like something that the console version will be for that audience, the audience that wants a more directed Minecraft experience?

JB: Yeah, I think it could be better. I mean, you don't get much help when you start the game. You're almost required to have a friend to explain the game to you to get anywhere. So, it could be better, but I don't really plan to work on that in the current time span.

It comes back to the fact that you don't want to lose what makes Minecraft Minecraft. A lot of people must be telling you different things, "This is the opportunity you should be following," but you have to really stick to what you feel confident about. And you have limited resources.

JB: Yeah. You know, sometimes I get Tweets or emails that say, "Oh, can't you bring back the old Minecraft feeling?" Then after that I think, "What is the old Minecraft feeling?" It must be something that when I first played Minecraft, I was really fascinated. And when I first found my first like dungeon with a chest, I was like, "Wow, this is so cool!" And if I created a new world with every release, I probably would get gradually less excited every time.

So, I think it will be hard to live up to expectations and at the same time... It's hard to keep the game alive and still not change it. Some people will complain about it.

Yeah, I'm not really sure how to answer your question. I think I will probably work more on things that I find more enjoyable. Maybe that's dangerous, because I remember when I started playing the game, I thought building was most enjoyable. But after a time, now I find farming is more enjoyable than building. So, I want to work more on farming, but maybe I will start working on things that are more advanced, that a new player wouldn't understand.

But that's also why it's so fun to work on Minecraft. Because every time I add something, you're almost never forced to use it. The only difference that we made recently was to add a hunger meter to Survival mode because you can't ignore it. You have to do something about it. So that changed the game for everybody. But if I add like, "Oh, now we can grow potatoes," it doesn't have any change at all for people that never grow potatoes. But it's fun for those who do.

You always have the community holding you accountable for the changes you make. So if you end up going down a road that doesn't satisfy the community, you're going to find out.

JB: Yeah. That's the thing.

But there is always a vocal minority.

JB: Yeah. But I can see when a feature gets more than normal amount of complaints. We have a few of those. I'm not sure if you're familiar with that, but in the recent versions, if you dig very deep, when you get close to the bottom of a level, it gets really, really dark. People have been complaining a lot about that because they have been building bases and stuff, and suddenly you can't see anything.

Even with torches?

JB: No. It's very dark. You can only see a few steps ahead of you. So, what I did was that I disabled it in Creative mode, but I will probably end up removing it completely. I'm fine with removing it, but it was Notch's feature, and it kind of becomes sacred ground. [laughs] And he really liked it. So, I'm a bit hesitant.

Right now, Notch's feature feels sacred because you've just taken over. But there's going to be a certain point where you're leaving Notch's features way behind you.

JB: Maybe. [laughs] I can't really tell yet. I think I will probably always ask him about things that I know he has been working on, most likely. At least if they are added.

I think there's also, when a feature is aged, it becomes easier to remove or change it. So, for example, if I should go back and change something that was added really, really early, I don't think Notch would have a problem with that. I don't think he has a problem if I remove new features either, but it's more that I don't feel that I have to ask him about changing old stuff.

Right. I'm sure that if he didn't feel confident, he wouldn't have handed the game to you, right?

JB: No. He has said that he has 100 percent confidence in me, so that's pretty good.