EVE Online: The Next Steps

[Running a game with a powerful and engaged fanbase is tough, and improving it and making it more accessible while attracting new users is even tougher -- and in this interview, EVE Online lead designer Noah Ward explains how he balances these influences.]

EVE Online lead designer Noah Ward has a tough job. Not only does he have to keep hundreds of thousands of players who enjoy the service happy by constantly updating and improving the game for them, he has to try and expand its audience -- while also not ruining the game for its existing fans. He has to figure out how to get the online console shooter Dust 514, under development in CCP's Shanghai studio, integrated with the main MMO without disrupting it -- in fact, make it additive to the experience.

CCP's EVE Online Fanfest, which takes place in Reykjavik, Iceland, is coming up soon, and the team has been working extremely hard on all of these fronts and more. In this extensive interview, the lead designer speaks to Gamasutra about the influences that are leading to changes in the game.

It's important to Ward to make sure he balances out relations with EVE's player government, the Council of Stellar Management, with the need to attract new users.

He also talks about how the company is doing research and metrics-gathering on both player and economic fronts, which is impacting the design of the game, and how social games are influencing his decisions on the design of EVE.

Where is EVE Online right now in terms of expansions?

Noah Ward: So, what just happened -- Incursion's gone out. And at CCP, we're running enterprise Agile. The whole company is pretty much on the same sprint cadence right now, which is pretty good.

Slightly staggered, and when the releases start and finish, but keeping everybody on the same sort of cadence really helps just for organization across the whole company. If you need to move someone at a team, it happens at a boundary.

So, all the things are running full steam in that sense. We're putting that out. Now we're just jumping into the next year of stuff there. And EVE does two big expansions a year, but this year, since Fanfest is in March, we've split this first release cycle up into four sprints and a hardening sprint, then Fanfest and four more sprints and a hardening sprint, then it's the summer expansion.

If we have stuff that's ready to go out, we can put it out at Fanfest and wow the players with it, but that's not necessarily a requirement of this first release cycle. Then we have the other expansion we do in the winter. So, we're just pretty much ready to do that.

The whole agile process has different things. We've done release planning, which is a big meeting where everybody goes in the same room and puts stickies on walls and decides what we're going to do or the first part of the year. I mean, now is a crazy time. We have no idea what's going to happen. You can make plans, but... What is that quote. "Plans are worthless, but planning is essential," or whatever. And they never survive first contact with the enemy.

So, yeah, we've got big plans. We're just going to have to see. Come March, when we've done the first part of this release, we're going to be able to have a lot clearer picture on what we'll be able to do.

Is that because you'll see what you have achieved via player response?

NW: No, not so much player response, just see how far we are on the technology. So, we're building a lot of stuff for Incarna right now. We've got the awesome-looking characters. We've to a lot of environment art. But the layout of how stuff is and how you interact with it and all that stuff is still yet to be made.

And there's a quality bar that we want to hit, and we don't know where we're going to be, so it's really hard to say how that's going to work. One of the main things that we want to make sure that we really nail is the new player experience, because if we don't nail the new player experience, then it's going to hurt us more than help us to do this whole thing.

We'll say, "Yay, you can walk around," and people have been hearing this for half a decade or whatever, and then this thing comes out... If it falls on its face, then people aren't going to give us a second chance. So, we're cognizant of that, and we don't want to screw that up.

Is that the real goal of Incarna? What is the goal?

NW: Well, our goal for EVE is to be the ultimate sci-fi simulator, and you can't be the ultimate sci-fi simulator if you can't go into dark seedy bars and make shady deals in back corners, if you can only fly around in a space ship... So, you know, there are a lot of goals.

The other thing is we've got a strong message from players, specifically ones who came and went, that didn't like see themselves as a spaceship. I used this example earlier; they wanted to be Han Solo, not the Millennium Falcon. So, we want to let people to do that.

And it's going to be an iterative process, and it's not going to be the perfect ultimate sci-fi simulator on the first go. I've almost been thinking in the very first release, it's almost like an extension of character creation. So, you make your character, your character is awesome, and now you can walk around in a little bit of a thing, and we're going to have to expand on throughout the coming years as we make more and more.

Your game is always touted as unique because unlike many MMOs, it's continuously expanded its audience and continuously expanded its gameplay -- sort of systematically evolved. They're intertwined obviously, in a way.

NW: One of the things that we're really turned on by in the future is just being more introspective and streamlining and improving on the systems we already have. So, EVE is a really complex game with a whole lot of systems. We can only go so far with adding more and more systems before we have to start looking at this and smoothing things out. So, that's what we've been talking about a lot lately.

Of course, you're not going to see that [in Incursion], but this is going to be like our new theme. We've talked about it a bit. We've mentioned it to the CSM [Ed. note: Council of Stellar Management, EVE's player government], and we've mentioned it to players.

And it's always a thing that players are telling us. They want us to start improving older systems, so that's something we're going to focus on in the future. Not dumbing EVE down -- that's not what I'm saying -- but simplifying and removing user pain, taking things and figuring out like the goals of the user and the process they go through, and trying to make it a nicer experience.

You're very careful not to say you're not dumbing it down; I think people get really wary of that kind of thing, and I think it probably goes extra for your audience.

NW: Well we've always been... I know I've just actually been reading a lot about these different user types lately, and the expert user type, we really cater to that, and we have for a long time. So, we've built up this community of expert user types,

And then the CSM is that type. As we listen and improve the game in directions that help the expert user type, we might be doing a disservice to the more mainstream people.

We feel like there is a way that we can cater to those mainstream people without alienating expert users. But it's a challenge. It's harder to build something that's nice and usable for mainstream and expert.

That's an interesting question. So is the implication that the CSM would be packed with people who have the biggest stake in the game and the highest, highest levels of knowledge and engagement with the game.

NW: Yeah, for sure. You don't run for... internet government spaceship thing... without being heavily invested into the whole thing. Or at least you don't win. Maybe you run, but you have to have this pretty big social network, and you've probably devoted a lot of time into EVE in order to win a seat on the CSM.

So it's up to us to really sort of listen to what they're saying and interpret what it is they need, rather than just doing exactly what they're asking for. I think that's just something that businesses have to do in general, to figure out what it is your customers need, not exactly...

Sometimes they're asking for one thing, but if you read into it deeper, it's really that they have a more fundamental concern that wasn't coming out. They're saying like "Why don't you move this button here?" when they're really saying like "Why don't you make this easier for me to do?"

You are totally at the forefront of this kind of interaction that potentially could become more and more relevant, as we all migrate online for a bulk of our interactions, in a lot of ways. How long has the CSM been going now? It's been several years.

NW: Yeah, so, in its current incarnation, it's been three or four years. But we had a CSM a long time ago that was just a chat channel that we invited people to, back... Almost at launch, there was a CSM. It wasn't the player-elected big thing that it is now, where people are flown to Iceland and talk to us, and we have these big long meetings, and every department prepares and gives presentations and talks. But, yeah, I mean, just talking to your customers is an important thing if you want to be in a service...

Absolutely. But this is sort of taking it to an extreme that I don't think really anybody else has.

NW: Yeah. I would agree. Not in the gaming industry anyway, no one's done this... Bringing your users in to try out your product and then monitor them is just... This is best practice in a lot of design-related industries. I've been reading about lately, they just say what you definitely have to do is sit people down, watch how they use, figure out where they stumble, fix it, talk to your users about what they're trying to accomplish.

So, we've been using the CSM. EyjoG, he's our doctor of economics, and he's got a research team, and they also research the users, not just the economy. We do surveys to the player base in the newsletter, and we get usually over 4,000 people responding to it, which means it's scientifically viable. And we're asking them, constantly, different things, and we've sort of built personas of our user base...

Based on behaviors?

NW: Yeah, based on behaviors, and based on answers to questions. So one of the personas is the "unwinding professional." There are about seven personas that we have, and the unwinding professional has a really hectic busy job, and he's the boss of everything. When he comes home to play EVE, he just wants to sit down and have some escapism and relax. He actually doesn't want to be like the leader of the fleet or anything because in his day job, he is the leader of whatever -- the thing. He just wants to kick back and shoot some stuff.

Then you've got this other persona of "the maven," and he's the guy who's been in every beta of every game ever, and he can tell you what the best printer is to buy because he's totally into like all the tech stuff, and maybe he's got a 3D TV already. And, you know, you ask him like "Is it time for me to buy a 3D TV?" "No, no, wait until the Sony blah blah comes out because that's going to be..." He's like one of these super nodes in the network that's just bringing everybody together. He's the expert user type that, you know, sort of the CSM is as well, these super nodes.

But looking at these different personas and seeing like what is their goal when they play EVE, what is it that they do, and what's the game loop that they're in when they sit down and play it, and trying to like optimize that for them. This is sort of just something we're starting to look at.

And for the longest time, EVE was built just on a hunch. You know, we were building a game for ourselves. Reynir [Harðarson] had the idea, and Hilmar [Pétursson] and the other guys who started CCP.

They were playing Elite back in the day. They played UO back in the day. They play a lot of Magic, and they just said, "Wouldn't it be cool if you had a game that was like the big procedurally-generated space thing of Elite, but the non-class based, you can do whatever you want PVP sort of thing that UO is, but you would fit your ships like a Magic the Gathering deck? That would be the ultimate game ever. Let's build that."

And, you know, they started building this thing, and everybody on the team just started building what they thought would be cool. But now we've got all these customers that aren't us. They're completely different people, and so we need to see what it is our customers are doing, and build something that's going to be nicer for them.

And also, you know, we've got 350,000 subscribers, because we've been growing from much smaller. It would probably be an order of magnitude more if we would build something that was a little bit easier to get into, if there wasn't that "learning cliff" that everybody jokes about.

And it's not because we shouldn't have complicated systems, and it's not like it shouldn't be a harsh world. It's just that people are presented with everything up front. Incarna is going to go to some way to like slowing it down and letting you feel a little like, "Okay, here's my captain's quarters, and I've got my little bit of the world, and then I go out and explore some, and then I come back."

It's not like the very first tutorial, and you were just sitting in your ship in space getting shot, and it was sink or swim, throwing people into the deep end of the pool. Some people really loved that, but probably most people didn't love that. "Oh God, I'm dying! I don't know what to do or press!"

The pace of learning, and the pace of the introduction of complexity, is an interesting question in games right now because we're reaching this point where people who don't accept the premise that it's either got to be casual or hardcore, that very reductive premise, are now wrestling with these kind of audiences.

NW: Yeah. That's the thing we've been wrestling with at work. A whole bunch of reading... Like, About Face is one of the really good books in this regard. It's about building things for the mainstream user and talking about what an expert user wants and how users move into being an intermediate user.

From the perspective of what industry was the book originally written?

NW: It's a usability book, and it's web and software applications, but they use examples from all different [places]... A lot of these web usability books -- the really well-regarded web usability books -- you can read the book and think of it, "If this was EVE instead of a website," and glean a lot of good info.

Yeah, just over the holidays and stuff, myself and some of the other UI designers were delving into these things deep, and just talking about how we need to clean up the stuff.

Another book that's really good called Simple and Usable... It's another web thing. It talks about ways of simplifying. Like, you can hide, you can organize, you can remove, and you can displace -- different ways of simplifying things. The example that book uses is a TV remote control, and talking about ways that you might simplify it.

It shows the TV remote control, like all these buttons, and talks about different ways -- like displacement, that's taking the things off the remote control and putting it on the TV so you would have a menu instead. Or hiding things in a little drawer, so the advanced stuff that you never really use is there, so you're not like "Ah, all these buttons!" Or organizing things better together, you know, like moving some stuff around. So, thinking about that stuff, and then you start looking at the EVE UI, and you go, "Oh my God. There's so much stuff we can improve with these techniques."

One of the positive influences of social games is the understanding of feeding back user data really effectively. Have you guys been getting that kind of data, aggregating it, and analyzing it yet?

NW: We are more and more. So our research and statistics team is on that, and I'm doing a lot of work to try and get the game designers -- research and statistics lives in publishing, and the game designers live on production, and we sit on a different floor, and it's almost like we're a different company. But at least we're the same company, and at least we're in the same building.

But you know how it might work at a publisher and a game producing thing somewhere else -- you wouldn't talk and you would have two separate key ways and everything. But I'm trying to get our designers to talk more with the researchers and to do that kind of stuff. The researchers have been looking into, the money flow and the sinks and faucets, and seeing how all that stuff is -- really delving into that stuff lately.

But we're starting to look at all of that, like how people are using the various UIs, and also where people are when they log on, and play, and log off. Are people starting their session in a secure space, and then moving to the unsecure, and then back again? We're looking at all that kind of stuff lately.

Sometimes when you just start looking at the data, you find things you didn't expect, and it sends you down paths to try to find out more, try to find out why things aren't that way.

We're looking at metrics, but we're also just polling. The polling is pretty interesting... the different polls are typically themed, and we ask the different Scrum teams, "Is there any specific thing about your feature that you added that you were really wondering what the players thoughts about it?" And then we'll poll them and get that info back. If it's pretty strong, we'll maybe tweak the game some.

You guys have the real challenge, which is keeping the plates spinning while also trying to improve. Either one or the other would be difficult, but both...

NW: And our increments aren't as rapid as the social games in Flash or whatever, I mean, sure, FarmVille changes stuff every week. You know, they add new stuff, and they're able to do...

Frequently, yeah. It's that rapid.

NW: And they just have a shitload of data. But, yeah, our sort of framework is a little bit harder to work, you know. It's not Flash.

Well, also, it's a more complicated... Ultimately, the changes you make are further-reaching because they tap directly into things like economy and real time social interactions, which are things the social games mostly so far have not had to butt their heads against. So, you're a more complicated, a more delicate environment.

NW: Yeah, so we tread a little lightly. But we're also not afraid to just make pretty big changes. I think it's necessary to do so.

If most social games add an item or change the value of an item, it still only sort of impacts, fundamentally, one person at a time. You could potentially screw up, or improve, the world dramatically.

NW: Yeah, we added planetary interaction, and you could suddenly get starbase fuel from planets instead of getting starbase fuel from highsec, and all of a sudden the money that was being spent on starbase fuel is not being spent, and you see the people in these big alliances that were spending money suddenly have more money, and now the game is just changed.

So, monitoring that and saying, "Okay, is this money that they're now accumulating, is it good? Do we need to find something else for them to spend money on? How should we deal with that?" And also just trying to be cognizant of the changes we're making and how they might affect the economy.

So, that's something we're going to be constantly in touch with the research team and just saying, "Okay, here's what we're going to do. What do you guys think that would change?"

How do you aggregate the data internally for your design team? Do you put it in a wiki? Do you send reports?

NW: We have a wiki, an internal wiki, obviously. But we've got OLAP cubes. That's where all the data is stored. You go into Excel, you configure the server, and then you can pull the data back in. It's a pretty slick sort of new system. It's in pivot tables, and then you just drag sections... and then it makes the data. So, yeah, it's pretty cool.

That's the thing about game design. It's an art and a science, and it is an inspiration, and it's also based on fact and technology and other things. It's where everything comes together. Game design is sort of the nexus point...

NW: Exactly.

...of where all the interactions occur, so you need something that's readable to people who then take it and turn it into a creative impulse.

NW: Yeah. So, a lot of people outside of the design department say, "Well, you should really use metrics!" It's like, "Well, we can read all the metrics in the world, but if we don't have a vision where we want to take it and a goal, it doesn't matter if a thousand people use this ship compared to that. Do we nerf that ship -- or did we want it to be that way?"

Yeah, it is definitely a bit of an art at the same time. Yeah, it's the same with EyojG. He tells us things that are happening in the economy, not necessarily if it's good or bad, just this is what's happening. It's like, "Okay, well, if you do this, money supply will increase. Beware of that."

I've talked about this a little bit with several games people, and I feel like so far they don't exactly have an answer. Yes, you can get metrics about what behaviors are being exhibited, like you said -- like a preponderance of people are selecting this one ship. But, like you said, that doesn't necessarily imply anything about what the reason behind it, nor what the effect is.

NW: Yeah. And I think to some extent, social gaming, it feels to me like big evil companies just putting people in a Skinner Box, and it's an aberrant conditioning sort of thing. And they're just sort of like, "Click this stuff, and then get the desire to pay us money!" And we're not trying to be quite that evil. We're trying to make people enjoy themselves, I think, and feel like they're this space captain badass dude, not just making people pay us for virtual chickens or, you know, that Cow Clicker game.

Well, Cow Clicker is a satire.

NW: Yeah, yeah. I know. But the thing is... I saw this talk at GDC Online, and he [Ian Bogost] was like saying how it was a satire, and yet here's people that are actually doing it. They were paying money. And I don't know if they liked the satire so much they wanted to pay for it, but it's weird that people could just -- even though, it was satire -- still get caught up in the loop.

Sure. I played it for a while.

NW: Yeah. The other thing, and this is mostly just because at GDCO there were a couple good talks about intrinsic motivators, that's one thing I've been really looking at lately.

I'm glad that people are looking at it. As soon as it came up as a focus point about how far we've been going toward extrinsic motivation, it became a concern.

NW: Yeah. Scott Rigby, I've seen him speak a couple times. He has a lot of interesting stuff to say about it.

Some of the really interesting stuff he says is like, it's not fun that keeps people around, and it's not their level of engagement. So you can see how engaged people are, but like six months later they might not still be playing. It's how much you're meeting these intrinsic needs that keeps people around. So, looking at that kind of stuff...

And there was another one that was like economic game design stuff. And he was saying, you know, we can use the stuff that we know about economics for evil, or we can use it for good. One of his examples was that people experience loss at like 2.5 times more than the experience gain, so if I would give you 100 of something and take 50 away, or if I would give you 50 or give you two 25s, you don't experience those the same even though economically it's 50 on all of them.

You can use these techniques to make people have a better time. You can make them enjoy themselves more. And that's the kind of stuff that I'm really interested in. How can we have people enjoy themselves more? Not, how we can trick them into buying fake hats?

Incarna opens up the possibility of selling people virtual goods. They could be superficial or non-relevant from a pure gameplay perspective, but people can still enjoy them.

NW: So, one of the things we’re trying to have is really "realistic" or "lifelike", I don't know exactly... but the avatars feel real, and when you look at a scene, it just feels like something out of a movie. I really, really don't want to have quest givers standing there with a big exclamation above their head. I want it to just look like this really cool sci-fi dark setting, and everybody fits in.

Hellgate: London had these towns, and there were just quest givers all around. I was like, "They're just standing there. They don't even look like they fit into their thing at all. My immersion is just ruined." So, I don't want to have that sort of thing happen, and part of making the realism, making the illusion work, is gaze attraction and how the various avatars respond to their environment.

Something we've been talking about a whole bunch -- we haven't made it yet, but this is something we really want to do -- is have metadata on all the characters so, for instance, if a big alliance leader walks into the room, everybody goes, "Oh, that's that guy," and you can just see the scene change, and that sort of thing.

And maybe he's wearing some interesting clothing, you know. And it actually has an effect on the world in some sort of way that makes them feel... And this is an intrinsic sort of motivation, you know. This is like the relatedness...

The way it would be, say, if like Bill Gates walked into the room or something. Everyone would turn and look.

NW: Yeah. Exactly. Everyone would turn and look and maybe "Oh, that’s... Ooh, can you believe it? Wow. Hey, it's him!" So, we want that sort of thing. See, this helps the relatedness of the newer players, too.

It would be interesting for us if he walked in the room just as much as Bill Gates would be like, "Here I am in the room, and people are looking at me and I'm this rich guy," but he's probably used to it by now. But if you were like, just sitting at some bar, the most famous person in EVE walks in, that's a cool thing for everybody, especially if the scene makes it feel like it actually fit together.

And it would communicate to the new players that there's somewhere to go also, you know what I mean? It's not just atmosphere; it's also information.

NW: Yeah.

They'd say, "Look, this is like an achievable thing within the context of this game," right? That's kind of interesting.

NW: Yeah. Who knows how it's going to pan out? [laughs]

I'm really curious about this, but maybe I don't know if you're the person to ask. The thing that I find really fascinating about Dust is the way it's intended to play into EVE. That sounds really innovative but also scary.

NW: Yeah, it is scary. [laughs] I mean, I don't know, I don't think we should delve too much into Dust because we haven't really been talking about it much, but yeah, as the lead game designer of EVE, I'm very protective of the EVE players.

My whole goal when designing the EVE side of the link to Dust is, I want it to be something that the EVE players want in their game. It has to feel like this makes sense, and just basically, you know, "I fly a battleship, but of course there's marines on the ground, and of course if I would want to attack this installation, I would pay marines to do it, and they would just run and do it." And who cares if they're on a console or another PC somewhere? This is just the fighting that happens.

But, yeah, it's a challenge, but we're having a lot of wins lately in getting this link. I mean, we showed it off at last Fanfest, and I suppose we'll show something else off at the next coming Fanfest and have, like, whatever the next stage is, and how it is.

Yeah, it's cool. We've got a couple test kits around the office in Reykjavik, and we're able to play. Not that we're necessarily developing it, but we can hop in, run around, and shoot at stuff. It's mostly just vertical slices of various stuff. You wouldn't show it off to people, but it's definitely interesting times.

It seems like between that and sort of the stuff with Incarna, you're broadening the scope of what EVE represents. You're saying like to your users, "Look, EVE has had a very obviously robust gameplay system that's been very attractive to tens of thousands of people, but we're expanding their universe to include new perspectives."

NW: Yeah, for sure. I mean, that's our journey from being the ultimate space trading MMO to being like the ultimate sci-fi simulator. A while back, when somebody took Star Wars and Star Trek and Battlestar Galactica and Babylon 5 and put the EVE logo up next to it and said like, you know, "We want EVE to be one of these sci-fi household names at some point in the future that people just go, 'Yeah, EVE. That's that thing that has all this stuff in it.'"

Just like a lot of people know about The Force and Wookiees and things like that. You, know, same with people know about the guys with red shirts in Star Trek die. Eventually some day we want people to know little tidbits about EVE, and they don't necessarily play it, but some day we want that to be... And people to play it on a PS3 and a PC and on iPhone and just, you know, everywhere, a sort of pervasive thing.

The complicated thing to do is to do it with the intelligence and thoughtfulness that you're already known for. If you wanted to make bad Resident Evil-style movies, I'm sure you could.

NW: Yeah, yeah.

I'm sure you've been approached, right?

NW: Yeah, yeah. Absolutely.

But you don't because that's not the foundation of your success.

NW: Uh-huh. Uh-huh. And we're like... I think we would do ourselves a disservice if we would do that. So, Star Wars is big enough that they can come out with crappy video game after crappy video game, and it's really not going to hurt...

Most Star Wars computer games, or games on consoles and stuff, are pretty bad. There's only a few Star Wars games that you can name that you can say were even good games. X-Wing, maybe, is one of them. But I don't think we could just come out with a whole bunch of bad EVE stuff, because we're not big enough yet to do that. But, I mean, at least we're getting in a stage where EVE trailers are played before Tron. I mean, that's a step in the right direction.