Single-Player, Multiplayer, MMOG:
Design Psychologies for Different Social Contexts

Ernest W. Adams

2010 Game Developers' Conference

This is an approximate transcript of the text of my lecture at the 2010 GDC on March 13, 2010. I present it in this form because the nature of the material does not lend itself to the traditional paper format. Also, because the lecture is informal and to some extent ad-libbed, this is not a verbatim document.

The final part of this lecture has been adapted as a Designer's Notebook column, "Selling Hate and Humiliation."


Hello, and thank you for coming. This talk is called "Single-Player, Multiplayer, MMOG: Design Psychologies for Different Social Contexts or "It's Not About You." I should begin by warning you that this lecture may be of more use to people who teach game design than to those who do it for a living. This lecture is actually a continuation of a talk I gave her back in 2004 called "The Philosophical Roots of Computer Game Design." It is not going to contain anything of tremendously practical value, so if you came in here wanting to learn how to maximize your profit margins or to ship games on time, I'm not going to be offering that kind of advice.

In "The Philosophical Roots of Computer Game Design," I made some observations about the nature of the task of game design. Along the way I observed that there were certain differences between English and French philosophy; that English philosophy tends to be driven by deduction and by hardcore rational thinking, and that French philosophy tends to be more inductive, and to be more about feeling. I also highlighted the "two cultures" debate, started by the scientist C.P. Snow in 1959 in a famous paper, in which he talked about how the academy is divided between the hard sciences and the social sciences or the humanities. There are these two cultures growing up in the academy, moving progressively farther and farther away from each other, and having a hard time talking to each other. I also discussed the distinction that Robert Pirsig made in his book Zen and the Art of Motorcycle Maintenance: the distinction between classical thinking and romantic thinking. The classical is rather like English philosophy. It's dedicated to deductive logic, while romantic thinking tends to be more free-form.

The conclusion I reached in that lecture is that one of the reasons game design is so hard is that we're actually trying to straddle these dichotomies. That what we do in the video game industry, what I do in my job, is to write technical documents that enable the creation of romantic fantasies. That is a completely bizarre idea. To an ordinary engineer, who thinks in terms of meeting requirements specifications, that is really strange. And to an artist, who thinks in terms of artistic expression, that is also really bizarre. So I've come to the conclusion that game design is neither art nor engineering. It's a craft, because it has both aesthetic and functional elements.


Leonardo da Vinci, the game designer's hero

I concluded in my "Philosophical Roots" lecture that our hero in this industry really needs to be Leonardo da Vinci. He was a Renaissance man who was both an artist and an engineer. He, too, straddled these various gaps: the C.P. Snow gap, the classical versus romantic gap. I think he's the person that we should look up to. So I ended the talk on an upbeat note, feeling that this was an important thought.

This lecture is a sort of précis of my thoughts since then about the nature of game design. And I want to talk for a little while about how I do game design myself, and how I consult with companies to teach them how to do game design, and how I teach students at a variety of institutions to do game design.

Now, you might ask yourself why anybody would bother to think about "the nature of game design," rather than just getting on with the job and doing it. The answer is two-fold. First, you really do need to know what you're doing. You wouldn't expect someone to just "get on" with architecture without first thinking about what we build buildings for and what we need to modify our landscapes for, what we want out of them, what we're trying to achieve with them. An architect who went into constructing buildings and modifying landscapes without having thought about that might make some very expensive mistakes.

Well, guess what: In the game industry we do make a lot of very expensive mistakes, and we pour a lot of money down rat holes. We kill two out of every three projects that we start. That's not a good way to do things. Any architecture firm that had to demolish two out of every three buildings that they started would very quickly go out of business. And you know what else? A lot of game companies do very quickly go out of business. So there is a point to thinking about this kind of thing.

It has been my experience that the biggest game design mistakes, the most expensive disasters, do not result from minor mistakes or technical problems, but from the major mistake of failing to actually think about what they were trying to accomplish in the first place. They get part way through the game and the company changes its direction because they haven't committed themselves to a particular thing that the were trying to do. So this philosophical noodling does have a practical purpose.

Fundamentals 2e coverThe second reason that I need to think about these underpinnings is that I have a university-level textbook on game design that I have to keep revising. I need to keep it up to date, both technically and as things change in the business of game design. There are thousands of college students all over the world who are buying and reading my textbook, and it's up to me to think about the background. I brought out a new edition just this past year, and I've already come to the realization that I need to bring out another new edition sometime in the next three or four years because things are changing so rapidly.

Let me start with a couple of points that I begin with when I teach game design, two more ideas that inform how I go about it. These ideas are in my book and my workshops and classes. Then I want to talk about how these ideas, which have informed the way I do it for a long time, have actually gotten me into trouble. In certain respects, they don't work any more.

Player-Centric Game Design

One of them is a concept called player-centric game design. I ask the designer to imagine a representative player of the game. The designer accepts two duties to this representative player. The first is the duty to entertain. The designer asks of every single game design decision made in the course of developing the game, "How does this entertain the player?" If a feature doesn't help entertain the player, then maybe it doesn't need to be there. (In some cases it does, because there are some features that we need for reasons other than entertainment—saving the game and bookkeeping functions and so on, that are not specifically directed at entertainment.) I believe that a game designer should regard any feature that doesn't entertain the player with deep suspicion, as guilty of superfluity until proven innocent.

The other is a duty to empathize, to ask, for every design decision that you make, how it's going to make the player feel. Will he become frustrated, or bored, or will it make them triumphant, or happy, or frightened, or what? Ask that question. Think about it. Because the designer has a duty to empathize with the player in player-centric game design.

I want to make a distinction here between the concept of the representative player and "the market." You might think that this business of thinking about a representative player is the same as thinking about what the market wants. I don't really want designers thinking about the market as a large, faceless statistical aggregate. I want them thinking about the player: a real person who is sitting there on the living room floor with the controller in his or her hands—not a statistic—who has chosen to play this game, and has certain beliefs and expectations and hopes about the experience. I do this particularly because I want to emphasize that it's necessary to make games for people who are other than ourselves. I'm very often dealing with young students who have been playing games all their lives, and the very first thing they want to do is make exactly the sort of game that they really like. In students this is not necessarily a problem, but as our market expands more and more, we need to be able to start reaching people who are different from ourselves. And if these students want to get jobs, they have to realize that young people are now a diminishing percentage of the overall market. When they graduate, they need to able to make games for girls and young children and senior citizens and people with disabilities, and all kinds of other people who are now game consumers that we always used to ignore. That is the reason for thinking about this representative imaginary player. Who is my imaginary player, and how do they feel about the way my design decisions work? What kind of emotions am I going to create in them—that's the empathy part—and am I entertaining them?

The Tao of Game Design

The next idea, that I came up with a little while ago—I wrote it up in one of my Gamasutra columns—is called "The Tao of Game Design."

To begin with, let's look at the way the Japanese language works. Japanese uses suffixes on words that modify the meaning of the first part of the word. There are two particular suffixes that I want to talk about. One of them is -jutsu. Jutsu means approximately methods or techniques. The other is -do, which is cognate with the Chinese word tao, which literally means way or path, but has come to mean a kind of underlying philosophy: how should we think about this thing when we do it? -do has come to connote a more refined or sophisticated version of whatever activity it is that you're talking about.

A very good example is jujutsu, which is a form of martial arts. The original form of jujutsu was a particularly brutal violent form—fighting to the death—for use in war; a desperate, no-holds-barred means of hand-to-hand fighting without weapons. Later, they invented a higher form, essentially a form of wrestling, called judo. This illustrates the distinction between these two terms. Jujutsu is the original methods and techniques of hand-to-hand fighting; judo is the more advanced form, the form that is informed by an underlying philosophy. It has additional rules, and so forth.

I feel that in game design we have a whole lot of jutsu. We have a lot of methods and techniques. When the player has consumed most of their resources when trying to accomplish a challenge, then you have to refill their resources again, and there are various techniques for balancing, and positive feedback, and fairness, and that kind of thing. We have a lot of jutsu that tell us how to design games, in a kind of methodological and technical sense. But my question is, what is the do? What is the tao of game design?

Tao_mediumI concluded that the Tao of Game Design is this. Every designer contains within himself a player, and every player contains within herself a designer. Every designer has to be thinking about that representative player all the time, and every player is trying to figure out what the designer had in mind. They're trying to figure out "What did the designer want me to do? How do I beat this game? What was the designer planning for me here?" So these two work together in the dance of creation. They're locked in this mutual exchange, this closely-coupled relationship.

I concluded that the Tao of Game Design is Know Thyself and Know Thy Player. Know what it is that you want to achieve, and understand what it is that the player wants from the game. Neither one can exist without the other. A designer with no player only creates an abstraction, a collection of pieces and rules. In my book I say that a game is an activity, not a thing. A game really only comes into existence when somebody starts to play it. If you paraphrase the old Taoist question, "If a game sits in the forest and there's nobody around to play it, does anybody have any fun?" the answer is no.

If there's a player without a designer, then there's no game to play. So each requires the other in order to make himself or herself whole. So that is the Tao of Game Design: know thyself and know thy player.

Limits of These Approaches

So these are the precepts that inform the way I teach game design. However, they are insufficient. They're not wrong, but they're incomplete. You might have noticed that all this time I have been using the term the player and not the players. That is, I have been thinking of the player in the singular. In effect, what I was doing was the very thing that I warned students against doing themselves, which is designing for myself. I was privileging single-player games, and that's because I prefer single-player games. And that's not right. It's inappropriate. I did that unconsciously because that's my preferred form of computer game.

When I was 16 or 17, I had a close friend named Terrance Druggan. Terrance and I were really into The Lord of the Rings, and we decided that we would build a Lord of the Rings board game. There were some already on the market, but we didn't have any money, and we thought we would create our own. So we went out and bought a blank hex sheet from Avalon Hill. I took that hex sheet and I opened up the front of The Lord of the Rings, and I got a whole lot of colored markers, and I copied line-for-line J.R.R. Tolkien's map from The Lord of the Rings onto this hex sheet, adjusting the features a little bit so that rivers ran along the boundaries of hexes.

Terrance and I got into defining some of the rules, but we kept kind of knocking heads, because it was very clear to me that Terrance was trying to fix it so that the good guys always won. Terrance identified with the good guys. He wanted the good guys to win. But I was talking about balance, and I kept saying, "No, no, no! We have to make it so it's possible for the bad guys to win too!" Terrance really didn't like that. So we abandoned it, and the hex sheet has remained in my parents' house from that day to this.

What I realized recently about this experience was that Terrance was failing to create a good two-player game; he wanted to create a single-player game. He wanted a game that the player wins; that's what happens in single-player games. We just didn't have a term for that back when I was 16. He wanted to create a game that you play by yourself, in which virtue prevails.

So I came to realize that the meaning of player-centrism varies significantly depending on the social context of the game, and that's what this talk is really about. I have come to the conclusion that the task of the game designer in these different social contexts is almost entirely different. It's not quite orthogonal—there's definitely some overlap—but it is profoundly different. So I'm going to talk about how the job of the game designer varies from one to another.

The Single-Player (PvE) Game

I'll start with the classic single-player game. I should explain that by single-player game, I really mean player-versus-environment game. I worked for six years for Electronic Arts on Madden NFL Football. Madden Football has a single-player mode, because it has an artificially intelligent opponent, but it's not a single-player game. It's a multiplayer game with an AI opponent.

In a single-player, player-versus-environment game, the nature of the designer's job is interaction design, where interaction has to do with the player's relationship to the environment. The designer sets up exploration, sets up puzzles, tells stories. Fairness, in the context of a single-player, player-versus-environment game, is very complicated. If we take a look at fairness in single-player games, players will feel that a game is unfair if any of the following things occur:

  • The difficulty of the challenges suddenly spikes.
  • The player suddenly loses the game in a way he could not predict or avoid (learn-by-dying). There was a time when this was commonplace in the game industry, but it's now considered bad form.
  • The game gets into a stalemate or deadlock.
  • The player has to make critical decisions without enough information.
  • The game requires factual knowledge from outside.
  • The types of challenges change unexpectedly. So you work like crazy killing aliens to get all the way through Half-Life, and at the very end of Half-Life, there's a jumping puzzle.

So the definition of fairness is really complicated, and the longer the game, the more these matter. You can get away with a few of these things in a short game—except for stalemates and deadlocks, which are always bad—but if it's a longer game it's really important to avoid these things.


The designer-player relationship in PvE games.

I feel that the single-player, player-versus-environment game is as close as we get to Art with a capital A. The relationship between the player and the designer is intimate and personal. If I'm playing a single-player game and the designer cheats me, I'm offended by that individual. And if the designer does a spectacular job, then I admire that individual. This is the Tao of Game Design. It's a very close relationship.

The Multiplayer (PvP) Game

Let's move on to the multiplayer, or player-versus-player game. These games are not really about the player's relationship with the designer; they're about the players' relationships with each other. The designer's work consists largely of competition design, and of managing interactions among others. The designer is an enabler of other people's fun. Your work as a designer consists very much of mechanics and balancing. There's a lot less of the storytelling and puzzles and exploration and all that kind of deep immersion, because it's really about the players' interactions among themselves.

The definition of fairness in player-versus-player games is much simpler:

  • The rules give all the players an equal chance of winning at the start. Of course, they don't have an equal chance of winning all the way through, because some players will be ahead and others will be behind, but at the beginning—ignoring issues of talent and skill, or prior experience—all the players have an equal chance of winning. (Interestingly, amateur golf includes handicapping, which enables bad players to play with good players. This is really quite unusual. If you play poker and you're a really bad player, they're not going to adjust the rules to make it easier for you.)
  • Players must not be able to cheat each other. If the players do something that is within the rules, it's not cheating , but if they do something that is prohibited by the rules, or they're deliberately trying to hassle the other players, that is cheating.

That's it. That's all there is to fairness in player-versus-player games.


The designer-player relationship in PvP games.


In these kinds of games the relationships are somewhat different. Balance is a question of competition. It's not about managing the pacing or the environment, but about managing the competition and the interactions among the players. They can't pause or reload; the game always goes forward. They don't necessarily expect to win short games, and in fact there tend not to be any long PvP games. If they get really long, they turn into persistent worlds.

Here I feel that the designer is more of an architect than an artist. You construct the building, but other people decide how to use it, and in fact, you have no control over how they use it. Players quite famously change the rules of multiplayer games to improve them. In Monopoly, the practice of putting all the money from fines in the center of the board, and then giving the money to whoever lands on Free Parking, is not in the official rules of the game. That's a house rule that a lot of people play by because they think it makes the game more fun and moves the money around a bit more. Another good example is forbidding the tank rush in Command & Conquer: Red Alert. Command & Conquer: Red Alert is badly balanced because it has the problem of the tank rush, so people just make a rule: no tank rushes.

As you see, I've drawn the designer and the designer's relationships here in light grey, to indicate that the players' relationships to each other are much more important than the designer's is to them.

The Massively-Multiplayer Online Game

Now I want to move onto massively-multiplayer online games and Raph Koster's Laws. I am not an MMOG designer. There are other people who are a lot better qualified than I am to talk about these kinds of things. Richard Bartle and Jessica Mulligan and Raph Koster and Sheri Graner Ray, for example. But I'll give you the benefit of what experience I do have with large-scale online games.

The very first job I ever had in the game industry was coding the PC-side client for an America On-Line game called RabbitJack's Casino. RabbitJack's Casino was a pay-per-minute game. The players logged on, played in this casino, and they paid by the minute to play. It cost 10 cents every minute, or 6 dollars an hour, to play this game. In EGA graphics.

I feel as if pay-by-the-minute games are the most honest business model of all, because as a designer, your butt is on the line every single second. You are keeping people happy and entertained, and if they are not happy and entertained, if they get tired or frustrated or bored or angry, they leave and you stop getting the money. It's a very direct measure of your success. Am I entertaining people? Yes or no. They're paying or they're not paying; they're logged in or they're not.

Another interesting phenomenon in those days is that people were nice to each other. [Laughter.] It was impossible to be a griefer in RabbitJack's Casino. The worst thing a player could do was wait the maximum amount of time allowed to place his bet, which forced all the others to wait too. But it was only 12 seconds. After that the timer would run out, and they would automatically fold and lose their stake. There was also lots of staff around to help out. They kept an eye on the conversations and threw out anyone who was being obnoxious.

I was really fired up about online games at that time. The thing that I found most exciting—we're talking 1989 here—was that there was no need for artificial intelligence. On a 4.77 MHz 8086 machine, the need for AI becomes a problem. If we could make multiplayer games in which the human beings provide the intelligence, then we don't have to, and that was very exciting. I gave a talk at this conference, the very first talk I ever gave, called "The Problems and Promise of Online Games," and I discussed all these opportunities and the various issues that we had to deal with. Most of them were technical and have long since been solved.

What I did not anticipate at the time was that online games would become so totally evil.

Let me talk about what happened when I started Second Life. Second Life is not a game in the conventional sense, but it is a massively-multiplayer online environment. I wanted to go in and see what Second Life was all about. It got a whole lot of press a little while ago and seemed to be a big deal.

In Second Life you don't start with any graphics already on your machine, because everything is mutable. You don't go down and buy a disc full of graphics at the shop, as in other games, that stay with you for the rest of the game. Second Life has to download all the graphics of the universe all the time. Constantly. Now I live in England, and the Second Life servers appear to be in Botswana as far as I can tell, so there's a terrible lag. I stand there on an extremely foggy island, the island for newbies, and then suddenly a brick wall appears out of nowhere right in front of me, and then suddenly a tree pops in, and other things pop in a little at a time as the bits crawl their way slowly from wherever the server is to where I am. So I didn't know what was going on. I was having trouble figuring out how to move around, and I spotted this other guy, so I thought I would try out the chat feature. I typed in "I seem to be having technical problems."

He turned around, and he said to me (in Spanish), "You seem to be having mental problems." [Laughter.] Now, I did not know this guy from Adam. I had never met this person in my life, and he has just gratuitously insulted me for no apparent reason whatsoever. I had not done anything to him. So what is this about?

I'm a grownup, so it's not as if I'm heartbroken about this. I don't consider myself to be a terribly thin-skinned individual. But what did occur to me was, "This is not OK for my mother." I might be prepared to tolerate this gratuitous rudeness, but she won't be. And I want my mother to be able to play games too. I want my mother to get into Second Life too. (God forbid she enter certain areas of Second Life. [Laughter.]) But if people are going to be crappy to you the moment you arrive, then what's the point? So that informs my thinking about this.

When we think about these online games, community-building becomes a major goal. That's really the point, and there have been tons of books and articles and discussions and talks at GDC all about community-building. In this context there's no such thing as a short game, and fairness become a very complicated concept again. Very complicated. Players don't start symmetrically. In Monopoly, everyone starts with $1500, and they all start on Go, so it's all symmetric. But in massively-multiplayer online games, everybody starts with different stuff, and some are clearly going to be ahead of the others because they've been there longer.

This suggests that the game shouldn't be about competition at all. But of course it still is. Sometimes players compete, sometimes they cooperate. The players expect at least to advance, if not to "win." You definitely wouldn't want people to win at World of Warcraft, because then the game would be over. You have to support cooperation in various forms, and usually some form of competition, and generally you also want to support solo play. As with other PvP games the players cannot pause or reload the game.

I was trying to find out more about massively-multiplayer online role-playing in order to write a chapter for my book about it. Of course you could write whole huge books about nothing else, but I did need to say something about it. So I spent some time looking into the topic and I came across Raph Koster's Laws. Raph Koster has been designing online games, MUDs and so on, since God was a child. He assembled his laws with a lot of other MUD and MMOG game designers that he had talked to over the years.

The vast majority of Koster's Laws are about trying to survive, as a designer and a game administrator, with a rude, unruly, whining, cheating player base. Here is one of them:

    Violence Is Inevitable. You're going to have violence done to people no matter what the facilities for it in the game are. It may be the combat system, stealing, blocking entrances, trapping monsters, stealing kills to get experience, pestering, harassment, verbal violence, or just rudeness.

So I took that on board. Here's another one.

    Baron's Theorem. Hate is good. This is because conflict drives the formation of social bonds and thus of communities. It is an engine that brings players closer together.

I might also call this George Orwell's Theorem or Adolph Hitler's Theorem, because he had this insight a few years before Mr. Baron, and it sure worked for him.

    In-Game Admins. … no matter how scrupulously honest [the in-game admin] is, no matter how just he shows himself to be, no matter how committed to the welfare of the virtual space he may prove himself, people will hate his guts. They will mistrust him precisely because he has power, and they can never know him. There will be false accusations galore, many insinuations of nefarious motives, and former friends will turn against him.

I don't know about you, but I didn't get into this business in order to lose my friends. There is not a single one of Koster's laws that gives a good reason for creating an MMOG. Not one.

Koster points out, quite rightly:

    It's a SERVICE. Not a game.
    It's a WORLD. Not a game.
    It's a COMMUNITY. Not a game.

Now, John Perry Barlow—one of the founders of the Electronic Frontier Foundation, sometime lyricist for the Grateful Dead, and all around cyber-pundit—turned up at this conference 18 years ago with a six-gun on his hip and gave a speech at the banquet in which he asserted that he was going to make the Internet safe for women and children. Leaving aside the gratuitous sexism of that remark, the only conclusion I can reach is that nearly two decades later, his effort was an abject failure. Viruses, worms, botnets, trojans, spam, phishing attacks, identity theft, Nigerian scammers... as far as I can tell the Internet is not only not safe for women and children, it's not safe for anybody at all. And when you build an MMOG, you are building an entertainment enclave in a place that is already pretty hostile. So long as Koster's Laws remain true, online games are going to suck for a lot of people. For people who are not prepared to tolerate being gratuitously insulted.

In spite of what this may sound like, I am actually an optimist. I believe that Koster's Laws don't have to be true in all cases. Club Penguin does not have these problems. In Club Penguin there really isn't any way to abuse others, even verbally. And maybe that's what I have to play. Maybe in order to get basic courtesy and decency from the player base, I have to restrict myself to the kiddie wading pool of online entertainment. But I do think it's possible to make a Club Penguin for adults. Some sort of online entertainment experience, or massively-multiplayer online game for grownups who are prepared to behave like it.


The designer-player relationship in MMOGs.

So here's the diagram for MMOGs. When you're building an MMOG, you're a social engineer. You have an absolutely vast number of players. You can't actually think of them as individuals at all. You have to treat them as a statistical aggregate. Based on what Koster has said—and Koster has forgotten more about online world design than I'll ever know, so I have to take his word for it—game design in the conventional sense is very much a secondary activity in MMOGs. The game is the hook to get people in and to keep them in, but that's almost not primarily what they're there for. So if it wasn't "about you" in multiplayer games, it is really not about you in MMOGs—until you screw up.

The Massively-Multiplayer Free-to-Play Game

This last part of the lecture is not mentioned in the title. Massively-multiplayer free-to-play games are new to me, and they're a relatively new arrival to our industry, when compared to the other business models we have used over the years. "Free to play" actually means "sort of free." It doesn't mean "truly free." The game doesn't cost anything if you have a whole lot of time, but if you want to advance at anything other than a glacial pace, you have to pay money, to buy virtual goods and things that enable you to get ahead faster. And I know even less about free-to-play, or F2P, than I do about MMOGs, but I've learned a lot in the last few days.

In particular, I want to talk, or rant, about a particular lecture that was given by a man named Zhan Ye at the Virtual Goods Summit 2009. I do not know this individual and I've never met him. These are his slides, which he has published online. You can find them at

Traditional Game Designers Are from Mars, Free-to-Play Game Designers Are From Venus

In his lecture, Zhan Ye asserted that in F2P game design, every feature must be measured by two metrics: is it fun, and does it make money? The designer is no longer free to make a fun game. The designer must be a businessperson. He asserts this explicitly. He says that fun is kind of a desirable goal, but it's about monetization.

I had this idea confirmed when I had a conversation with Matha Sapeta, who's an old friend and a designer at Playdom. She knows a great deal about free-to-play gaming also, and is the lead game designer on Sorority Life. She told me that at Playdom, every game feature must drive one of three things: daily average users, or DAUs, which simply means "number of logins";  re-engagement, which is fancy biz-speak for "the number of people who come back"; and monetization, which is a nice way of saying, "how much people spend." So every design feature must drive one of these, and is measured against that. You'll notice that there's no sign of empathizing with the player here.

Now, I come from the retail business. I worked for Electronic Arts. We made games, we put them in boxes, we put them on the shelf and hoped that they sold. The designer of retail game also thinks about whether the features will be popular or not, but he or she is free to take a more holistic approach to the whole thing. You don't have to measure moneymaking potential on a feature-by-feature basis. You don't decide that this year you're going to put a new playbook into Madden, and for each new play that you add, it will earn you 15 more cents from each player. We don't think that way.

So let's take a look at Zhan Ye's lecture, because I found it extremely enlightening. It taught me a lot about free-to -play. For one thing, he said we had to get over these conventional notions of fairness.


My response to this is, God forbid the game world is a reflection of the real world. Who the hell wants a game world with all the misery and oppression of the real world? Why don't we just throw in cancer and Alzheimer's while we're at it? [Laughter.] They're not fair. Maybe in the context of a game you can make money selling people artificial cures for their artificial cancer. "Oh, you're a newbie and you didn't pay, so you've suddenly got cancer, but we'll sell you the cure."

Regarding fairness, he also says:


Then he goes on to mention a solution that didn't work, which I'll skip—this is a bit out of context, I'm showing you whole slides, but I am skipping some slides. Here is his solution that he says does work:


Oh, great. This is gangsterism. This is warlordism. This is tribalism. This fantasy game world that they've constructed is essentially Afghanistan or Uganda or Somalia, where children and the poor are forced into militias at gunpoint, abused, and made to fight. "Fight in our army or we'll kill you." It doesn't seem to have occurred to him to create a game in which nobody kills anybody at all.

Now, maybe this is popular in China. Clearly he says that a lot of people will pay a lot of money for it. Maybe when they want to escape from their day-to-day lives in an oppressive totalitarian centralized regime, what they fantasize about is being peasants forced to fight for a brutal overlord in an oppressive totalitarian decentralized regime! [Applause.]

Zhan Ye defends all this in his lecture by likening it to Las Vegas. [Laughter.] He points out that gambling takes advantage of a human weakness and never goes out of fashion. These free-to-play games take advantage of another human weakness, the desire to dominate and oppress other people. Apparently that never goes out of fashion either.

I think this is a dangerous sort of analogy. Gambling is very heavily regulated. Do we really want free-to-play games to be regulated the way gambling is? He comes from China where everything is regulated, so maybe he's not aware of the difference, but in a liberal democracy we have different expectations. Also, the analogy is very inexact. Las Vegas is not free to play. It doesn't have to charge the paying players enough to cover the expense of supporting the non-paying players. In fact, the whole essence of the experience in Las Vegas is that you must pay to play. Las Vegas is actually much closer to the old pay-by-the-minute games I used to work on, the difference being that you can win real money.

Most importantly of all—and this is a key point in this question of fairness—Las Vegas does not deal aces to rich players and deuces to poor ones. Rich players can play for longer before they run out of money, but everybody plays by the same rules regardless of how much money they have.

Personally, I find this whole idea completely appalling. I first heard of this lecture from a guy named Rich Carlson, who's somebody you should know. He's one of the Digital Eel guys, and he's an old-timer. He designed board games and card games and video games a long time ago, and he believes in this concept of fairness, and he believes in players treating one another with certain minimum standards of decency. He sent me the link to Zhan Ye's lecture in an E-mail message with no further comment than the subject line, which simply read, "An obscenity." And I have to say that I'm inclined to agree.

Here's another example.


Zhan Ye also thinks that conflict and hate are good. They have this mechanism: "When people are emotionally unstable they're more likely to make purchases." Well, that's a really desirable state of affairs, isn't it? And there's this virtual item called the Little Trumpet that you can use to curse other gamers. It will be broadcast to all the other gamers, so it's a public humiliation tool. That's just really delightful.

Is this what game design has come to? Creating things to sell players that enable them to be crappy to each other? Looking around for opportunities to make money out of emotional instability? The only people that ought to be making money out of emotional instability are therapists, and at least they're trying to improve the situation. Even the handgun industry tries to make the claim that they're only selling them for self-defense. They're don't say, "Go out and blow the hell out of people because that'll make you feel better, and besides we can sell you virtual bullets." [Laughter.]

Now, I'm not against competition. Competition is fun. I like a hard-fought game. But there's a social convention called sportsmanship that is designed to keep competition on the right side of the line. I realize that I'm beginning to sound like a crotchety old man, and that some of you are sitting there thinking, "Oh my God, sportsmanship? What century did he crawl out of?" But I have to tell you something: When competition turns into hatred, you have gone too far. If you are building games that foster tribalism and hatred, you are doing evil. There is no such thing as artificial hatred. All hate is real. [Applause.]

Summary and Conclusion

So, the GDC insists on lectures including some takeaway, and here it is:

If you're a single-player game designer, you are an artist, and your relationship with your player is as close as it will ever be in any game. Everything that I have thought and taught about Player-Centric Game Design and the Tao of Game Design is still correct for you. I believe these approaches are the right ones in single-player game design: know thyself; know thy player; you have a duty to entertain; you have a duty to empathize.

In multiplayer games you are the architect of interactions among others. Your first concern is not how the player perceives your game, but how the players use your game to enjoy themselves together, either through competition or through cooperation or team play. Fairness is critical to your experience. Balance is key. It's only partially about you.

In massively-multiplayer online games you are a social engineer. You are attempting to build a place where people will want to live over a long period of time. In your case, you can't provide fairness in a competitive sense because some players have been there longer than others. A lot of games have realized this, and they have removed direct competition between players. They made player-versus-player interactions either impossible or consensual. They create situations that tend to group players together who are of like skill level. They cooperate and go out raiding together, and they're roughly equal. The Tao of Game Design does not apply to you because your relationship is with a very large community of people and not with any one person. I still encourage you to try to entertain and to empathize with your players.

Massively-multiplayer free-to-play is an area that is new to me, and as far as I can tell it's chiefly about economics; about predicting and manipulating the spending patterns of people en masse.  If Zhan Ye is correct then as a game designer you are an economist. The question of monetization infuses every design decision that you make. Your job is to create currencies and study their flow, and find ways of encouraging more spending on virtual goods.

So far as I can tell, neither the Tao of Game Design, nor Player-Centric Game Design, applies to you. You want to entertain the player, certainly, but you empathize with the player only to the extent that it is profitable to do so.

I think the only way to make these games fair is to remove the element of competition from them, so the rich players can't just defeat the poor ones by spending money. It becomes about growth and advancement, not about death and destruction. And that's how Farmville works. They should not be zero-sum games.

It's also possible to place caps on the players' ability to compete using money. The NFL salary cap is an excellent example. The NFL salary cap was put in place because rich teams could always hire the best college players coming up, so naturally they tended to win more games and got richer still. The NFL said, "You can't spend more than X amount of money on hiring players," and this tended to level the playing field and create a better experience for everybody. What that did was make the entire NFL more interesting and balanced. By contrast, the America's Cup, and F1 motorsports, have turned into technology races where it's more about money than it is about talent or skill. I mean, the America's Cup has just gone weird. It's hardly about sailing the boat any more. It's about designing the boat.

Ultimately, game design is fragmenting. The new business models mean that the way we go about it has changed dramatically, and it's increasingly difficult to teach the subject, or approach the subject, with a single unified philosophy. There's a lot to think about. But there's one thing, I think, that we ought to try to preserve. If I ever make a game in which there's a feature whose sole purpose is to humiliate other players and make them feel bad, then take me out and shoot me on the spot. If I ever make a game whose purpose is to enable players to be crappy to each other, hang me from the nearest tree. Because that's not what I'm here for. I think we at least try to keep one thing in the backs of our minds:


Thank you very much.