Digital dystopia:
Player control and strategic innovation in The Sims Online

by Francis F. Steen, UCLA

[This is the draft of a chapter for Avatars at Work and Play, edited by Ralph Schroeder and Ann-Sofie Axelsson. London, UK: Springer, 2005 (in press).]

1. Introduction

Around New Year 2004, a quiet year after the official launch, The Sims Online hit the global headlines. Disappointingly, the topic was not a celebration of how this "new virtual frontier," this "daring collective social experiment," had succeeded in bringing "our divided nation" together, as Time magazine had blithely prophesied [1]. Instead, the media reported that the online game had turned into a Biblical den of iniquity, a Sin City, a virtual Gomorrah -- and that the whistleblower who bore witness to this had his account terminated [2,3,4,5]. Rivaling mafia organizations were practicing extortion and intimidation, pimps running brothels where underage girls provided sex for money, and con artists scamming newbies out of their start capital. Maxis, the company that designed and operated the game, maintained a relaxed laissez-fair policy of light and somewhat haphazard intervention. The first mafias and in-game brothels had been established already in the early days of beta-testing; they continued to operate unchecked. Pretend crime paid well and recruitment was good, leading to a rapidly mutating series of inventive scams targeting the inexperienced and the unwary.

It wasn't meant to be like this. Gordon Walton, one of the chief designers, had spoken in glowing terms of the game's potential to provide opportunities for better relations between people. While "all of our mass media positions us to believe our neighbors are psychopaths, cheating husbands, and just bad people," The Sims Online would short circuit our distrustful negative stereotypes [6]. Echoing McKenna & Bargh [7], who had found that relationships initiated online benefited from transcending the limitations of spatial proximity and physical appearance, leaving more room for a creative identity construction that might act as a guide to a real self, Walton envisioned a game where people would "interact with others anonymously, have physical distance, and not be judged on your outward appearance. You interact with people on a pure intellectual and emotional level, devoid of all those filters." If his team did their job right, The Sims Online would feel like Disneyland [6].

Careful steps were taken to forestall "griefers", gamers who derive enjoyment from anti-social behavior. In line with the vision of a privately run amusement park as the contemporary image of the good society, The Sims Online would have no common areas or public property where griefers could harass people. The habitable landscape would be divided into lots, each of which would be owned and controlled by an individual gamer. If a player was giving others a hard time on your lot, you could throw him out or permanently ban him. If that seemed too much trouble, you could restrict access to a registry of friends. The programmers even anticipated the problem of one player blocking another's exit from a room; to avoid this they decided the trapped avatar would be able to escape by walking straight through its would-be captor. Implementing Lessig's (1999) dictum that on the Internet "code is law," [8] the makers of The Sims Online (TSO) sought to prevent crime by writing software code that made crime impossible.

These twin elements -- the redemptive vision of the game's potential for creating friendships across barriers of distrust, and the proactive, structural legislation within the computer code itself to remove the threat of anti-social behavior -- represent two of three main pillars of the game's utopian project. The third is more subtle: the freedom of the players themselves to create and to govern their own virtual society. Its spokesman is Will Wright himself, the lead designer and originator of The Sims family of games. Even before the game was released, he had begun imagining a self-governing society, with local governments and elections. But these features would not be built into the game. "All of this political stuff has to come from the bottom up," he insisted from the start. "We can't do it from the top down and dictate structure" [6]. A key inspiration was the architect Christopher W. Alexander's work on the emergence of communities. Alexander speaks of a "pattern language" that evolves organically from people's small acts. The patterns that define a town or community "can never be 'designed' or 'built' in one fell swoop -- but patient piecemeal growth, designed in such a way that every individual act is always helping to create or generate these larger global patterns, will, slowly and surely, over the years, make a community..." [9]. Could TSO also be designed so that each individual act added up to a whole? After putting a basic framework in place and building some incentives into the game, it must be left to the players to establish their own political and civic cultures. According to Wright, "totally planned cities don't work. It's sort of like the Utopian society movement, where there were these guys who went off and started building planned cities. For the most part the cities were total failures" [6].

The uncertainty that such freedom entailed, however, was also a cause for worry: the team was painfully aware that nobody knew how the virtual world within the game would develop. Walton and Wright envisioned a society in which human relations were -- paradoxically -- more direct, because they transcended space, physical appearance, and entrenched identities, where crime was banished by the very architecture of the game, and where human freedom would express itself in self-organizing cultures. Could this vision could be realized? Within the first year, this utopian dream shattered, at least provisionally, by persistent reports that human relations had taken a turn towards an eerily familiar catalog of exploitative behaviors, were crime flourished and spread in the face of attempts to remove its very conditions of possibility, and where freedom had led not to democracy but to warring mafias. "How would people act if they were freed from real life laws and social constraints?" the BBC asked rhetorically, responding with reports of "child prostitution, rampant crime, mafia-controlled neighborhoods, shadowy self-declared governments struggling to maintain order and runaway inflation" [3]. "Hobbes in Cyberspace: Life in an online game world proves nasty, brutish, and short," Reason Magazine concluded [10].

In the following, my goal is to understand the role of the unanticipated outbreak of crime in the dynamics of cultural creation in TSO. I start by examining the origins of the game in the single-player offline version of The Sims, focusing on the key design features that later become incorporated into the multiplayer online version. In the second section, I examine the three main dimensions of player involvement in the game: the immediate-term mechanical control of the avatar, the medium-term control of strategic moves, and the long-term control of the goals and ultimate meaning of the game. I argue that at all three levels, control is inadequate and tends to produce dissociations between player and avatar. In the third section I discuss the architectonic constraints on cultural development, the scope for rebellion against these constraints, and the significance of the strategy of crime. I end with an assessment of the future of the game and the lessons to be learned from this vast and exorbitantly expensive experiment.

1.1 The origins of The Sims Online

In order to understand the surprising dynamics of the online world of TSO, I propose to begin by examining its history. The official birthday of the game is the 17th of December 2002, when the game was first released to the general public. At that point, however, The Sims Online was already a bustling world, thanks to the activities of tens of thousands of beta-testers. Starting in mid-September, they had been invited to join the game for free, to uncover any show-stopping programming bugs, to ensure that the system scaled adequately, and to populate and settle the vast and virgin electronic landscapes. Paying users, Maxis reasoned, would prefer to join an existing world to constructing one from scratch. These beta-testers, in turn, encountered a world that was already highly structured, even if this structure would become actualized only by the gamers' own activities. Some of these structural features are odd: key design decisions of The Sims Online make little sense until one realizes they are the result of code inherited from the earlier, off-line version of the game, The Sims. I will suggest, in my analysis in sections two and three, covering player control and the higher-level cultural dynamics of the game, that the high degree of path dependence on code written for an off-line, single-player environment is a significant contributing factor to the problems that subsequently unfolded.

It was the Oakland fire of 1991 that provided the impetus to what was to become The Sims. Will Wright, a game designer at Maxis, lost his house in the blaze, and in the following months he became fascinated with the process by which his family gradually drew up the plans for the new house and its furnishings. Wright abstracted the sequential characteristics of this process and realized it could be turned into computer code. The program would supplement and extend the power of the imagination, simulating the construction and decoration of a house. He began to design Home Tactics: The Experimental Domestic Simulator. In 1993, after two years of in-house development, he presented a prototype to Maxis executives during a focus session. The idea was so unlike anything that had been done in computer games up to that point that they rejected it outright. For the next four years, the game was only worked on by Wright himself in his spare time [6]. When Maxis was acquired by Electronic Arts in 1997, however, Wright's reputation as the designer of SimCity earned him a blank cheque with his new bosses. Home Tactics got all the attention it needed and was released under the new name The Sims in January 2000. Once scorned, The Sims and its expansion packs became the most popular computer game of all times, selling more than 20 million units over the next two years [11].

At the core of The Sims is the act of building, landscaping, and furnishing a suburban house. The natural environment is an invariant subdivision: a rolling meadow by a brook, a road circling a dozen lots ready for construction. The game invites the player to build a house, or a succession of houses, and to move ready-made families into them. The building materials, the plants and trees for landscaping, and the furnishings and interior decorations are selected by the player from virtual shopping windows and assembled onto the three-dimensional canvas of the building site at the click of a mouse. A profusion of bricks and wood sidings, roof shingles, potted plants, wall papers, chairs, and kitchen utensils makes this part of the game straightforwardly enjoyable. The combinatorial possibilities are finite but astronomical; for all practical purposes infinite. You can move the walls around freely, put in a kitchen, a bedroom, a bath, a pool, a TV lounge, an exercise room. These practices don't build genuine engineering skills -- according to a classic text of structural engineering, "A deep, intuitive appreciation for the inherent cussedness of materials and structures is one of the most valuable accomplishments" of an engineer [12]. In TSO the materials are flawless, each brick or chair the spit image of any other of its kind: there is a perfect and predictable match between the real and the ideal. The very act of design is an act of building. This simplification of reality allows the player to focus singlemindedly on the task of architectural design, maneuvering through a gratifyingly vast possibility space. The task of design is further aided by a series of elegant program features: while you build and decorate, you can zoom in and out and view your progress from different angles. As you adopt a particular view, the walls in front of you becomes selectively transparent, allowing you to see the entire layout of the house. This aspect of the game bears the quality stamp of sustained iterative design [13] and surprises the player with delightful features.

The second layer of code in the game is the sims, or simulated people. They are the dolls that inhabit the houses built by the player. Controlled by artificial intelligence (AI), the sims are imbued with a rudimentary form of agency. They can walk from one location to another, perform a finite but expandable repertoire of tasks, express emotions, and communicate desires. The player interacts with the sims in a manner very different from that of the building materials. Where the latter move only when the player moves them by clicking and dragging her mouse, the sims behave as if they were alive, folding their arms impatiently if you leave them standing, as if chafing at the bit. While they do not actually move from place to place or perform tasks on their own initiative, you also cannot move them directly, by grabbing hold of them with your cursor and dragging them to a new location, as you can the building materials. For a sim to do something, you must give him an instruction to act. This is done by interacting with the sim's virtual environment. Objects in the sims world are endowed with Gibsonian affordances [14], or activities and behaviors that can involve the sim in some way. By right-clicking on the object, a contextual menu in the form of a cloud of affordances is displayed, and you make your choice. Clicking on the sim itself will give you a menu of possible behaviors valid for that particular location and circumstance, or you can click on a distant object and instruct the sim to interact with it. For instance, you can wake your sim up in the middle of his night and tell him to go for a swim in the pool simply by right-clicking on the pool and selecting "Swim." Like a truculent child in the face of a stern parent, the sim will get out of bed and stamp his foot on the floor repeatedly in a bout of displaced aggression and frustration, a fume of anger rising from his mind. He finds his way through the house and out to the pool on his own, without any player intervention: this shows his behavioral routines are pre-programmed and robotic. As he gets into the pool and begins to swim, the image of a bed floats in a thought bubble above his head, informing you that he is tired and needs to sleep. In the morning, having so rudely been deprived of sleep, he may become unresponsive to the instruction to go to work and require time to regain a functional level of comfort.

These responses create the impression, gratefully accepted in pretense, that the sim is a dynamic, homeostatic system, whose behavior is regulated by a simulation of causally connected mental and physical states such as hunger and the need for food. Dominating a sim's life is the physiological needs at the bottom of Maslow's pyramid [15]: the need to sleep, to eat and drink, to go to the restroom, to be comfortable, and -- this is after all suburbia -- to take regular showers. Each of these needs is tracked by a bar on the player's screen, turning slowly from green to red if it is neglected; taking care of your sim so that his basic needs are being met is called "greening." Higher needs have a rudimentary presence: if you don't provide your sim with regular social company, he will become despondent and slowly refuse to function. With some variations, this is as far as what Wright calls the "economy of motives" has been elaborated [16]. Sims do not live in a world where they need to worry about safety, and although in some versions of the game they can fall in love and marry, they do not complain if this doesn't happen to them. The built-in motives are of a ground-level nurturing kind. The sims have no aspirations to achieve social recognition through outstanding acts -- to kill a dragon, to become president: they don't have a life project, a mission. Nor do they show an interest in learning and understanding the world they live in, though if you make them read employment-appropriate books, their skill levels increase and they get promoted at work. They do not seek to actualize themselves artistically or spiritually. They are, after all, dolls.

What are the design goals realized by creating sims with this particular circumscribed degree of autonomy? A sim has a fixed set of clearly defined needs, in part conveyed directly through emotional responses and through the display of the simplistic and generally predictable content of his mind, yet he is entirely incapable of taking care of even the simplest of these needs himself. Devoid of independent initiative, he relies on you, the player, to instruct him how to meet his own needs. For your benefit, a special "greening" panel tracks the precise progress of the sim's needs from moment to moment. If appropriately instructed, the sim will be healthy, energetic, and promptly carry out your commands; if his needs are not consistently met, however, he will drift towards a non-functional state, become unresponsive to your commands, and eventually die. This design creates -- and is clearly aimed at creating -- a distinctive dependency relation between the sim and the player. The function of this dimension of the game is to encourage the player to care for and nurture the sim. The game in effect provides opportunities for a kind of practice parenting, similar to playing with dolls, but with a more realistic feedback. By nurturing the sim, the player experiences the systematically differential consequences of proper care and neglect, and acquires skills relating to taking care of others. Care and nurture are behaviors with a deep natural history and they remain vitally important for any society, yet they had never before been the target of a sophisticated computer game. The Sims tapped into a vast and hitherto neglected audience of young girls who were left cold by the typical competitive or adversary shooter games favored by the boys [6].

Experienced players of The Sims, however, take the game far beyond the elementary task of keeping your sims green. As a player, you are in a position of directing the sims' lives as a dramaturg directs his actors, creating dramatic situations and developing extended narratives. Your capacity for absorbing information, for seeing connections, for opening up new possibilities, exceeds that of your charges by orders of magnitude. Like a god you can control whole neighborhoods of sims, staging their marriages, births, quarrels, reconciliations, and breakups. The writer Monique van den Berg's illustrated Sims diary [17], where a dozen families interact in intricate and often comic situations, provides good examples of the game's potential for staging complex narrative scenarios in richly elaborated environments. If you so choose, the godlike power of the player can even be used in the spirit of Gloucester's "As flies to wanton boys are we to th’ gods, They kill us for their sport" [18]: you can command your sim with the terrible voice that the God of the Old Testament used to instruct Abraham to sacrifice Isaac, wall him into a closed room, electrocute him when he changes a light bulb, make him drown in the pool. If you adopt the sim's point of view, the situation is distinctly odd: in the middle of the night, with nobody in view, you are commanded to perform some meaningless act, strongly aversive to you, and you feel the anger welling up in you at the pointless imposition. But the command is as ineluctable as it is mysterious, and you are entirely incapable of disobeying. Such is the power of the player in The Sims.

In summary, the immensely successful offline, single-player version of The Sims developed out of a program designed to simulate the hands-on building and furnishing of a house, and was elaborated to incorporate robotic agents dependent for their basic welfare and continued existence on the constant nurturing behavior of the player. The game facilitated player psychologies ranging from an architect and interior decorator to a doting mother caring for her family and a movie director staging elaborate narratives. The secret of the game's phenomenal appeal lay in providing players with a godlike power to explore and innovate in two complementary and fully developed permutational spaces: that of building, landscaping, and decorating houses, mansions, and castles, and that of caring for, directing, and narrating the lives of simulated humans in evolving tangles of complex social relations. When The Sims was to be taken online, Electronic Arts sought to build on the success of the existing game and reuse tested code by preserving the core features of the game. Just like The Sims, The Sims Online would present a suburban housing division, albeit a much larger one, where players constructed and furnished houses on lots, using the mature code inherited from the offline game. Just as in The Sims, the simulated people within The Sims Online would respond robotically to instructions selected by the player from clouds of affordances surrounding in-game objects. The secret of the game's phenomenal appeal lay in providing players with a godlike power to explore and innovate in two complementary and fully developed permutational spaces: that of building, landscaping, and decorating houses, mansions, and castles, and that of caring for, directing, and narrating the lives of simulated humans in evolving tangles of complex social relations. The Sims would go online.

Yet massively multiplayer online games rely on a very different dynamic in the relationship between players and on-screen characters. Key to a multiplayer environment is that each player is represented in the virtual world by a single character, his or her avatar. A relatively tractable dimension of this change is that the game now needs to have a unique avatar for each player. Users typically want to control their avatar representation and have input into its design, yet the need to download tens or hundreds of thousands of unique avatar designs onto each player's computer would cause severe network and storage problems [19]. The Sims Online solves this problem by providing more than a hundred different heads and bodies with outfits for the user to mix and match. This combinatorial space is sufficiently large to minimize the risk of two avatars looking exactly alike, yet avoids storage problems, as all avatars can be represented by a number referring to the graphics in the selection. While this solution provides sufficient differentiation between avatars, it does not allow players to contribute their own graphics, thus limiting the work the representation can do in defining and channeling a particular identity.

More troublesome than avatar design was the requirement of The Sims Online that the player identify with his or her avatar. The mantra of the development group was that in TSO, "the sims are real" [16]. Massively multiplayer online games like EverQuest and Ultima Online had shown that players became emotionally engaged in their own avatar and formed strong bonds with others through their on-screen representations. Such identification would be absurd and inappropriate in The Sims, where each simulated human being was endowed with a carefully circumscribed autonomy, expressed in robot behavior, designed to be endearing and to elicit a caring and nurturing stance. As Will Wright himself put it during the alpha phase of the development process, "That's never been an issue in any of my games before. Most of the time I'm dealing with little simulated AI people that pee on the floor all the time" [6]. In The Sims, the player was in charge of the entire virtual world and all its inhabitants; in The Sims Online, tens and hundreds of thousands of players would interact with each other. If the player was god in The Sims, what would be his role in The Sims Online?

The conversion of The Sims to The Sims Online, starting in 2000, was an enormously complex and expensive undertaking, with a development staff of a hundred programmers, three million lines of code, and a rumored budget of $25 million [6]. It was undertaken at breakneck speed, and achieved its ambitious goal of shipping in time for the Christmas season 2002. Subscription numbers shot up instantly to 40,000 and by New Year to 80,000, continuing to mount until topping out around 105,000 in June 2003, far below the projected numbers [20]. In the following section, I examine two players' responses to the game over the course of the first year, focusing on identification and player control. How did the experience of player control change as the core features of The Sims were ported to a radically different, multiplayer environment?

1.2 Player Control in The Sims Online

When The Sims Online was released in December 2002, Patricia Greenfield, with the collaboration of Brendesha Tynes and the research team at Children's Digital Media Center at UCLA, had already recruited two players and initiated data-collection. The intention was to treat the game as a laboratory in which the spontaneous emergence and evolution of culture could be documented. By starting participants at the game's beginning, Greenfield hoped to be able to observe not just the adaptation to an existing culture, but the actual creation of a culture from scratch. Steen and Davies joined the project in early 2004, and when the assembled team began the task of examining the collected data, it became clear that we needed to change the research focus. In the participants' diaries and captured gameplay we found little or no evidence of the players creating a shared cultural world of meanings, norms, activities, and physical environments through processes of social interaction and communication [21]. The virtual inhabitants of Alphaville, the first city within The Sims Online, appeared to have surprisingly little to say to each other, and the game did not provide our study participants with opportunities or tools to engage in sustained collaborative cultural creation. As we observed their interest in the game slowly fade, we shifted our focus to investigate what had gone wrong. In the following, I present a summary of our findings, with selections from the data and some new analytical points; for the details, see Steen et al. [22].

A massively multiplayer online role-playing game (MMORPG) can provide its participants with control along a spectrum of time horizons, ranging from the immediate to the long-term (figure 1).


Spectrum of player control


Fig. 1: The spectrum of player control in a MMORPG

At the far left of the spectrum we find the immediate-term mechanical control of the avatar. Is there a one-to-one correspondence between the player's manipulation of the game controller and the behavior of the on-screen avatar? Such synchronous control is a mandatory feature of fast-paced shooter games; Kirk argues it is critical for a low-level, physiologically driven player-avatar identification that powerfully enhances game immersion. [23] The code that The Sims Online inherited from The Sims handles avatar behavior through artificial intelligence, thus removing a base level of support for an immersive experience. In contrast to other MMORPGs, such as There [24], you cannot directly control your avatar -- you cannot, for instance, use the keyboard or mouse to turn your avatar to face a certain direction, move its arms, or move it step by step in some direction. All control is done through menu interaction, where you give instructions the avatar acts out.

In the diaries and videotaped gameplay of our first study participant, KM, a 23-year-old female graduate student, we found that the lack of synchronous avatar control generated a series of dissociations between the players and their avatars. KM described herself as a recovered addict of the off-line version of The Sims and was thus familiar and comfortable with the sims' robotic behavior in the single-player environment. In the spirit of the new multiplayer environment of TSO, she began play on 30 December 2002 by creating an avatar to represent herself, an alter ego:

I created this character based on myself. It took me a while to go through all of the hair and outfits to pick one that I thought resembled me. (KM, p. 16)

In short order, she was confronted with what for all of us was the most surprising feature of TSO: the low level of social interaction and conversation. KM began by identifying the avatar she encounters with a real person:

I said "hi" to the other person playing chess, but little conversation happened. I noticed that no one in this house was talking everyone was just earning skill points (KM, p. 17).

Skill points can be earned through activities such as standing in front of a mirror ("practicing charisma"), playing chess, and cooking. In order to encourage socializing, a key novel feature engineered into the game is that skills are acquired more rapidly if practiced in the company of other avatars. As with other avatar activities in TSO, the behavior itself is robotic, and skills points take hours to build. KM's suspicions were soon aroused:

After about 10 minutes of playing chess, and seeing how long it took for the skill meter to go up I could see why no one was talking -- probably no one was there! (KM, p. 19).

The substance of this insight is a dissociation between the avatar and the player: the presence of one frequently does not entail the presence of the other. In the absence of synchronous control, avatars in the online chat rooms of TSO can be intermittently attended; it is common practice to go "afk" or "away from keyboard" and leave your avatar, robotic slave that it is, stay behind and accumulate points. An unfortunate downstream consequence of this behavior is that players begin to see the presence of other sims in purely utilitarian terms, as an opportunity for earning skill points more rapidly, rather than as a chance to meet people and socialize. This second dissociation separates your own needs for real human social contact from the avatar's more pedestrian need for the proximity of other avatars, attended or not.

The absence of synchronous control also created anomalies in the immediate-term mechanics of social interaction. We analyzed a recorded conversation at Lucky Luc's Slots, where AJ, the proprietor's roommate, gives KM instructions on how to play a particular gambling game. During most of this conversation, the two avatars have their backs to each other, yet it would be incorrect to infer from this that they had lost interest in the conversation. At one point, AJ refers to a game score result that she could not have seen, as it appeared behind her. These incongruities open up a third dissociation, that between the panoptic perspective of the player and the embedded, in-game perspective of the avatar.

Zooming out leaves the avatar behind
Fig. 2: Players can leave their avatars behind and like disembodied spirits view the world from above. In the meantime, the avatar continues to engage in activities as instructed.

Finally, robotic behavior can generate dissociations of intentional states. In one session of recorded game play, we observed KM giving instructions to her avatar to play the guitar. Once instructed, the avatar persists in the task until it is completed. Other sims talk of leaving; half in jest, as she isn't actually playing, she says, "I am going to go too, cant stand listening to my own music." Zooming out of the building, she ends up with a bird's eye view of Alphaville, from which you can see all the different properties and decide where to go next. Her own intentions have dissociated from those of the avatar, who is left behind playing the guitar (cf. fig. 2).

In KM's diaries, we begin to understand why so little conversation takes place in The Sims Online: typical gameplay is characterized by long absences from the keyboard, as the robotic work of skilling is itself experienced as boring. Our analysis of her game play captured on video indicates that this boredom forms part of a series of dissociations between player and avatar, which act cumulatively to weaken the bond of identification. These dissociations have a primary cause: the absence of immediate-term, synchronous player control.

At the other extreme of the spectrum of player control is the long-term goal or overall purpose of the game. In the single-player, offline version of the game, The Sims, the player is comfortably in charge of the purpose of the game, whether it is to build castles in the air (there's a trick to getting them up there), to raise a sim family, or to start a gay karaoke bar [17]. In keeping with the hallmark of Will Wright's game philosophy, the promise of The Sims Online was that it, too, would allow the gamer to formulate his or her own goals within the game.

TSO, however, had introduced something co-designer Chris Trottier called "a real secure economy" [6]. In The Sims, the economy is insecure. If you enjoy the constraint, you can abide by the rules and hold off on that swimming pool gazebo until your sim has boned up on his mechanical skills, developed some friendships, and can start earning some real money. But if you don't, a Google search for "sims cheat codes" will pull up 283,000 hits. These cheat codes, which are built into the game, allow a player to get unlimited resources with a few keystrokes. The insecure economy of The Sims allows players the flexibility of setting their own goals. The secure economy of The Sims Online, in contrast, introduces a resource scarcity into the game that tends to swamp all other goals, as illustrated in the following analysis.

Our second study participant was a 28-year old man (SH), a free-lance producer and part-time actor. On the 19th of January 2003, after spending the first play sessions exploring the game, he creates an avatar, "Sammar," to represent himself:

I chose to develop this character because he is the closest thing to my alter ego. I needed an outlet for that ego in order to help myself in real day to day life. I'm hoping that I'll be able to learn from my other self and take those characteristics that I feel I lack and forge them into my real life. (SH, p. 9)

His entry suggests a very rich vision of the game's potential. By creating a virtual self, he imagines he will be able to explore and to cultivate modes of being and responding to the world that he can subsequently incorporate into his own life in a selective manner. This is a goal he sets himself, demonstrating his confidence in being able to control the overall purpose of the game. After recounting Sammar's first day of play, visiting places, getting a roommate, and earning money -- "simoleans" -- making pizza, he states a subordinate and interim goal in the game, now projected onto his avatar:

His main goal at present is to make enough money to build a party pad by the beach. (SH, p. 11)

His entry contains the key elements of a narrative self: a goal (a party pad), an obstacle preventing him from realizing the goal immediately (not enough money), and a strategy for overcoming the obstacle (making money). The goal itself is envisaged in social terms: his house will be constructed in an attractive location and provide a venue for himself and other players to have a good time together. As he continues playing the game, he maintains a strongly positive disposition:

A lot of the people I've visited at their properties have been exceptionally nice. I imagine it has to do with their visitor bonus. The people I met in the pizza place are not nearly as friendly. It's amazing what greed will do. (SH, p. 13)

The effect of the game's incentive structure on the other players is beginning to transpire. Property owners are paid for the time others spend on their lots; his warm appreciation of their welcoming behavior is qualified by the suspicion that they, just like the unpleasant pizza makers, are driven by greed. A few days later he reports,

Sammar is feeling accepted in this community. He is still figuring out the finer details but it's coming along well. He aspires to make his skills at their peak and make as much money as possible." (SH, p. 19)

SH continues to view TSO primarily as a community, a place regulated by norms and common meanings, in which he can feel accepted and welcomed. At the same time, as he begins to master the rules of the game, his visionary goals narrow. The next day, we find him for the first time thinking of friendships in instrumental terms:

He's building a friendship base that's making him money and skill. (SH, p. 23)

When SH returns to the game on 18 March 2003, he reports,

"Sammar" has built his skill levels, mostly mechanical & logical, and is making a decent amount of money making gnomes. He has a home now and is in the process of building it up to be a place where other sims can come to relax and make money. (SH, p. 25)

He now shows little emotional involvement with his avatar, whose activities in this session are directed not towards forming relationships, but on building skills and making money. However, he sees these activities as a temporary means to a more attractive goal of building a house. The purpose of this house is still to provide a place for others, but he no longer imagines they will come to party. Instead, they will come to his house to hang out, and to make money. His basic motivation remains altruistic:

My characters main goal at present is to be a viable and successful character who can help other Sims in their money and skill earning endeavors. (SH, pp. 25 & 27)

At the same time, his interest in The Sims Online is starting to flag. The game, he writes, is "somewhat boring" and does not facilitate the social contact he came looking for:

The game would be more condusive to chatting if email were accessible while playing to swap pics and personal info. A real possibility of meeting these people off line would get the place buzzing. (SH, p. 39)

Avatar encounters, apparently, do not have the emotional and intellectual qualities of real encounters. His suggestions of introducing e-mail, swapping pictures, and meeting people off-line indicates that he experiences the on-screen characters as poor representatives of the players' social agency: identification has become unattractive and the channeling role of the avatars is failing.

When he decides to give the game another try in April, he abandons Sammar and begins Freakstick, a skeleton-like character to "express my off the wall personality" (37). He attempts to reformulate his strategies:

Now that I've learned the main tricks and tips in suceeding in The Sims, I have a new way of going about things. I plan on amassing large amounts of mechanical and logical skill. Those skills have the greatest amount of financial profitability with the least amount of constant attention. (SH, pp. 37-39)

He soon learns, however, that TSO provides very limited opportunities for rapid progress in the game, gets bored, and plays infrequently. On 27 July 2003 he goes on for a brief session to build skills. There were few people online, so the effort didn't pay off as much as he had hoped. "Maybe there will be more people the next time I log on," he comments (SH, p. 45). He is now speaking of people as a simple means to speed up the gaining of skill points; he is no longer interested in socializing or meeting friends. The incentives of the game appears to have ground sociality out of him.

On 4 August 2003, his goal is subtly reconceptualized, even as he represents it to himself as unchanged:

My ultimate goal, still, is to gain enough skill and money to build the ultimate house where I won't have to work at making money. Rather I earn money by collecting the revenues given to me by the Sims for visitors coming to my house. Also I will get residuals for every dollar that my guests make. (SH, pp. 47, 49)

Shortly after, complaining that the "time it takes to build skill is a little overwhelming, not [to] mention boring" (26 August 2003, p. 51), the record trails off and he abandons the game.

From start to finish, SH displays an active desire to define the long-term, overall purpose as well as the intermediate goals of his gameplay. Step by step, he lets go of his own goals and adapts to the goals built into the game. He relinquishes the desire to become a better person by trying out new modes of relationship within the game, he gives up on the idea of helping others succeed and of becoming appreciated by a community of friends. Unable to remain motivated by the time-consuming and mindless activities of skilling and making money, he is drawn towards the ultimate goal provided by the structure of the game: to become a "sim lord" (SH, p. 41) and live off the labor of others. At that point, human relationships have deteriorated beyond the level of instrumentality to something reminiscent of exploitation. An early review spelled out the emerging culture of the game:

Since a player can earn money simply by enticing other players to congregate on their property, and because all the other players truly want to do is earn money, the object of the game is reduced to building -- not a "house" in which your Sim will live, but a labor camp in which other Sims will come to earn money. Providing beds, showers, food and a pool table persuades your guests to stay longer and spend more of the money they are earning, owing their souls to the company store, so to speak, and never truly needing a place of their own. The result is a "city" in which nearly every house is a sweatshop. [25]

Skilling
Fig. 3. Avatars practicing "charisma" before mirrors in a lavish interior. Clicking on another avatar gives you a "cloud" of interactive options; the sequence of tasks to be carried out by your avatar accumulates in the upper left corner. A needs panel monitors your avatar's state along eight dimensions.

What The Sims Online failed to do, then, was to provide the players with the tools to control the overall purpose of the game. From the early interviews, it is clear that Wright and his development group fully intended to build this freedom into the game. When large numbers of beta-testing citisims started spending their time making pizza, the developers were distressed. "A few weeks ago, we thought we'd have Disney World. But right now, everyone is just making pizza," Trottier lamented. Wright worried that too many people were chasing money in the game -- "we might start to lose the creative players" [6]. However, no satisfactory solution appears to have been found.

Taking The Sims Online involved two major design challenges, one social and one psychological. The psychological challenge was the shift from a godlike player to an embedded avatar. In KM's experience, we see some of the player-avatar dissociations that make this transition unsatisfactory. On the social front, the challenge was to establish some kind of resource scarcity. Kollock's seminal "Design principles for online communities" argues that scarcity is an important dimension of a vibrant online community, not just to keep things lively, but because "moderate amounts of risk are required for the development of trust ... and encourage the formation of groups and clubs as a way of managing that risk (or exploiting it, in the case of a guild of thieves)" [26] (Kollock, 1996). Using only a "secure economy" to create scarcity, however, backfired. It set up a dominant incentive gradient that funneled most people's energies into mindless and boring money-making routines, destroying the fun and creativity of the game.

The effect of the design choices made in these two major transitions was to curtail player control at both extremes of the spectrum -- the immediate-term mechanics of avatar movement, as well as the long-term goals and purposes of the game. There remained, however, the central portion: player control of intermediate-term strategic innovation. This area provided the developers with ample possibilities of expansion, but they were not exploited. A rich set of tools and opportunities for medium-term, strategic innovation would have given players something to focus on, compensating in part for the loss of control at the extremes of the spectrum. We see in SH's diaries a mounting frustration at the ineffectiveness of his actions and his inability to come up with strategic shortcuts, indicating a strategic deficit. Player control at the intermediate, strategic level involves the ability to formulate complex and clever sequences of moves that help further the player's goals in effective and original ways. In TSO, plodding along a number of roughly equivalent routes was the only legitimate way forward.

When the players of TSO found, to everyone's surprise, that they had nothing to talk about, the explanation may lie in the glaring absence of strategic opportunities. In most multiplayer online games, it is strategy development that provides the gamers with something worth communicating about. Indeed, a distressing effect of the failure to develop the central, strategic portion of the spectrum of player control is that players had no way of becoming uniquely valuable to each other. Human beings are irreplaceable resources for each other in part because each one of us gathers our own information, adopts an idiosyncratic perspective, and develop our own strategies -- complex sequences of actions that reliably achieve improbable results. In a society rich in strategic opportunities, information becomes the critical scarce resource. In the next section, I examine how a relatively small number of innovative players found a way of mining this middle of the control spectrum, with dramatic effects.

1.3 Strategies and limits of cultural evolution in The Sims Online

Full-scale social experiments are expensive. In order to put a new model of social organization to a realistic test, you need tens of thousands of subjects over a period of years. Finding a suitable and willing population is almost impossible; funding the enterprise is prohibitive; gaining human subjects approval is impracticable; the logistics a nightmare. In the initial phases, you will need to insulate your state to some degree from its surroundings, to give it time to develop its own economic practices and civic institutions. The continuity of history must be broken for your ideas to be implemented and given full play. If your experiment goes awry and your ideas turn out to produce a monstrous society, you will be held personally responsible, exposed to the murderous ire of your own captive citizens. Until recently, short of a full-scale revolution or a military invasion, the only viable alternative available to social visionaries was the thought experiment. Where past ages where constrained to chose either the expensive realism of war, colonization, and revolution, or the cheap but fallible instrument of utopian thought experiments, we now have a technology that permits us to create imaginary worlds and to populate them with real people. What was simulated in the mind for millennia can now be simulated in public, in vast simulated online cities.

The utopian question is not "What emerges from a state of nature?" [5]. A utopia is a social experiment that involves the explicit and deliberate manipulation of certain structural parameters in establishing and running the imagined state. The intricate thought experiments of Plato's Republic, More's Utopia, and Butler's Erewhon exploit the imagination as a cheap and readily available vehicle to explore the possibility space of social arrangements. They ask the question, "How do the choices we make in setting up the structure of society affect the behavior of individuals and thus its course of history?" As computer-mediated active worlds proliferate, this question is pivotal not only for understanding the evolutionary trajectory of a particular multiplayer online game, but also for being able to develop the kinds of worlds we want. The utopian question has become reformulated as a challenge of software design: in an active world, "what features must the environment have in order to enable particular types of social interaction?" [19]. The evolution of culture in a massively multiplayer online world cannot develop from scratch; it must necessarily emerge from the complex dynamics of interactions between the programmers, the structural framework of the game they create, and the gaming activities of the players, which may take place both within the game itself and -- as we shall see in a moment -- extend beyond it. A game is not a clean-room implementation of a new society, untainted by preexisting values, beliefs, and conventions, but imports these dimensions, explicitly or implicitly.

In the case of The Sims Online, we begin with a richly featured environment, structured on multiple levels. I argued in section one that the secret of The Sims' phenomenal appeal lay in providing players with a godlike power to explore and innovate in two complementary possibility spaces: that of building, landscaping, and decorating houses, and that of caring for and directing the lives of simulated humans in evolving tangles of social relations. In section two I argued that the shift to TSO involved two major innovations, on the opposite ends of the spectrum of player control: one social and one psychological. The transition from an insecure to a secure economy effectively barred most players from the joys of building, reducing their game play to sweatshop labor, their godlike status to serfdom. Adding insult to injury, the transition from the dollhouse to the avatar model barred players from the godlike power to control and stage the lives of the sims within the game. These architectural decisions were in large part forced upon the developers: going online meant that you had to remove the godlike power of the player. In this section, I will argue that TSO's critical shortcoming was that these constraints on the extremities of the spectrum of player control were not compensated for by new powers in the middle: by new strategic opportunities.

The co-creation of culture within a multiplayer online game is not necessarily cooperative; the goals of developers and gamers may be only partly overlapping. When the implementation of a game fails to provide attractive avenues to success, inventive players may seize the strategic initiative and attempt to move the game in new and unanticipated directions. The game designers' vision may be incoherent or produce unanticipated results, and the developers may falter in the implementation -- in May of 2003, EA spokesman Jeff Brown acknowledged that "The people who make The Sims [Online] believe that its execution isn’t what it should have been when it was launched" [27]. As we saw in section two, casual players, working with the tools provided to them within the game, met with repeated frustration and boredom in trying to achieve their goals. Creative players, on the other hand, may extend the boundaries of the game, drawing in resources that supplement the game, and find ways to compensate for its weaknesses.

The absence of a mechanism to import customized graphics into the game, for instance, cut off an important dimension of user creativity and constrained the elaboration of in-game identities. Some of the players responded by creating off-game web sites devoted to their TSO avatars, in which they were able to utilize their own art work and draw freely on cultural references that added resonance and power to their role play. A striking example was www.thesimsmafia.com by JC Soprano, played by the 25-year-old Sacramento native Jeremy Chase. It sported dramatic flash animations on the mafia theme and a detailed list of available in-game services, from prostitution and gambling to debt collection and assassinations -- "Moe Green Specials" as he called them. Another group of players established the Rose Bush Gardens neighborhood in Alphaville devoted to "Bondage, Discipline, and Sadomasochism," amplifying this theme on the external Black Rose Castle Learning Center web site, with detailed instructions by the avatar Lady Julianna on how to be a dominant without being obnoxious, and how to participate in pony sub races within the game. Urizenus, played by University of Michigan philosophy professor Peter Ludlow, created The Alphaville Herald, an online newspaper covering in-game events. By conducting in-character interviews outside of the game, he raised the profile of the inventive players, created a wider audience for their role play, and provided them and other players with an opportunity to reflect on the significance and impact of their in-game behavior. In these and other ways, the limited opportunities for creating an arresting identity within the game were transcended, as the gamers recruited a range of off-game resources to reinforce and heighten their own game experience.

Expanding the game beyond the confines of TSO proper also increased the effectiveness of the in-game character. To circumvent Maxis' Online Community Representatives, the Soprano mafia family members chatted with other players on Yahoo Instant Messaging (IM) rather than using the game's own (logged) chat feature. This allowed JC and his recruits to refine and develop a string of strategies for mimicking mafia-type activities within the game, without giving them away to the developers. As presented on his web site, these strategies were mysterious, vaguely menacing, and clearly fun. Maxis tried to keep up: "Most of the behavior described in stories about these 'mobs' is no longer possible, actually; we've been improving the game with frequent updates," associate director Kyle Brinx claimed in a June 2003 [28], but six months later JC confidently proclaimed, "The city is mine.. I hate to say it, but I got the juice in AV [Alphaville] and have for awhile" [29]. By figuring out ways to 'warp' the built-in features of the game for mafia role-play, he had in effect seized the initiative from the developers and taken a strategic level of control of the game.

The blatantly anti-social character of the mafia role, adopted within a game that set out to simulate a real society, raised the question of how such a virtual society can be policed and governed. By creating a publicly available representation of his in-game activities, JC Soprano made it easy for the non-playing world to participate in contemplating this question. In the summer of 2003, the Associated Press did a story on "Sex, mob hits: Sims tests virtual morals" [30], reporting on the exploits of two rivaling mafia groups, the Sopranos and the Sims Shadow Government (SSG). CNN followed up a month later with live coverage [31]. In these interviews, both mafia families defended their activities by claiming that they dispensed a rough justice to discourage griefers. EA simply didn't respond effectively to protect innocent players from abuse. "Griefers ... find and utilize loop-holes within the game," Jennifer Mathieson of the SSG said, "And it happens very, very quickly. So what we do, we just fight back. We use the same tactics ... against them." [31]. Jennifer and her husband, who jointly played the avatar Mia Wallace, recounted that they had "ransacked apartments, sent out their 'troops' to urinate on others' lawns, and once drove another player from the game" [30]. By extending role-play into the media, these gamers gave people a new reason to play the game: to experience and to explore the ethical dimensions of online worlds. The "darker side of Sims life," Wright himself conceded, "makes the game more interesting. It is pretty playful and harmless" -- and the governance of virtual communities "is something our society is grappling with" [32].

At the same time, The Sims Online was revealing its potential as a breeding ground for a wide range of humiliating, anti-social, and exploitative behaviors. In a BBC interview, Ludlow later explained that his make-believe newspaper, created off-game but edited in-character, was founded to document "the emergence of economic, social and political structures in the game" [3]. As events unfolded, The Alphaville Herald turned out to be perfectly positioned to become the media hub of the seething underworld of TSO. As Urizenus, Ludlow covered the rise of the mafia families and their increasingly hardball tactics, such as sending a new roommate to an unsuspecting sim, who would then tear down his house. The interviews show that the players invested considerable emotion in the conflicts. In early December he conducted a series of instant-messaging interviews with Evangeline, a cyber-prostitute, who had set up a brothel early on in the game as a strategy to avoid the boredom of skilling and working. Describing her business in graphic terms, Evangeline let it slip that she was underage herself, and claimed that several of the girls that provided "sexual services" within the game were also minors. Using the proceeds from the prostitution racket, she had purchased the property at the top of the game's welcome list, naming it Free Money for Newbies. Here she cheated newcomers out of their money, humiliated them by caging them in small rooms and ridiculing black avatars as monkeys [24].

Of course it was just a game. The houses are pixels on a screen, "money" is a play currency called simoleans, and sex between avatars is no more than dirty talk in cartoon bubbles. The CNN anchors kept tongues firmly in cheek and concluded that "this is all taking place in a virtual world. We can hope that it stays there and that if you don't like it you can just leave the game and stop playing" [31]. Electronic Arts, the game publisher and Maxis's parent company, who had long been aware of these activities, handled the publicity angle by an appeal to unreality. Confronted with stories of online prostitution in an interview with The New York Times, Jeff Brown, vice president for communications at Electronic Arts, said, "If someone says that is going on in cyberspace, is it lost on anybody that it's not actually happening? No law was violated. It's a game" [4].

Yet the interface between game and world does not present a simple and clean-cut boundary. The motto at Maxis, as we saw in section one, was that in TSO, "the sims are real" [16]. A subjective act of identification with the sim is an integral aspect of the design of the game: from your perspective as a player, your sim was intended to function as a virtual self onto which your own subjectivity and agency is displaced, and you become emotionally invested in the sim's changing social relations, reputation, and resources. Equally, in order to interact with and understand the other sims, you need to model the sim as an avatar channeling a real person. Since the sims are real in the sense that each one is a real individual once removed, displayed onto an avatar, then ethical questions that could be entirely ignored as fictive in the offline version of the game have inescapably moved much closer to reality. Add to this the fact that creative players actively widened the boundaries of the game, extending their gameplay far beyond the confines of the game itself, and in various ways integrating their in-game character with the real world. Finally, they found ways of taking strategic control by devising their own methods of earning money within the game, bypassing TSO's intended gameplay. EA continued to treat the game as if everything were happening within the their proprietary controlled world, when this clearly was not the case. Ludlow's research assistant, Candace Bolter, pointed out that although prostitution was remunerated using simoleans, the in-game currency of TSO, this "fantasy money" was readily convertible into US dollars through money trading at online auction sites such as Gaming Open Market and eBay [33]. Electronic Arts, they argued, had a moral responsibility of governance that they couldn't simply walk away from: The Sims Online was not just a game.

Urizenus' The Alphaville Herald provides a case in point: it started out as an in-character blog, but in the course of documenting the conflicts and exploitation taking place within the game, it quickly morphed into a project of serious investigative journalism. The significance of this stance towards the game was underlined by the following incident. In October 2003, roleplaying as the priest Urizenus in a TSO church, he was contacted by another player:

A sim IM'ed me claiming to be a 13 year old boy and started asking me about God and forgiveness. He claimed that he had beaten his 8 year old sister because she had annoyed him, and that she had gone to the hospital with a broken jaw. I asked if he had reported this, and he said no, and then broke off contact with me [34].

Responding to the reality of the situation and casting aside an implied in-game seal of confession, Ludlow contacted EA to report a real-life crime, repeatedly requesting they pursue the case with the boy's local authorities. EA responded with a string of boiler-plate customer service replies, advising him that they could only take action on Terms of Service violations [34]. Bolter finally threatened to take the case to the media; EA relented and handed the case over to local police. They also terminated Ludlow's account on a technicality. In the following days and weeks, the termination story and the background corpus of interviews was picked up by the international media, from The Detroit Free Press to the Corriere della Sera and Izvestia, featured as an attempt to supress the public's knowledge of the truth and a possible violation of freedom of speech. In a recent academic paper, Ludlow notes that the real-world press incongrously treated The Alphaville Herald as a real newpaper, even though it was produced as an in-character blog as part of a game, and advances the thesis that "There is no such thing as fiction, and there are no such things as fictional objects" [35].

In this section, I have argued that taking The Sims Online entailed a loss of player control at both ends of the spectrum (figure 1). While large numbers of casual players, including our study participants, appear to have responded to the highly constrained game by attempting to adapt and eventually losing interest, a small number of highly inventive players found a variety of ways of taking charge of the game. They accomplished this by developing intermediate-term strategic control on two fronts. First, they found original ways of bending and twisting the built-in features of the game into complex series of moves to achieve their own defined goals. Second, they extended their gameplay beyond the confines of the game to include web sites, instant messaging, and media outreach, thus embedding their in-game character within a broader matrix of cultural references and meanings. Moving in-character communication out of the game allowed the players to develop their 'warped' use of the programmed rules in a conspiratorial spirit of secrecy, away from prying eyes of EA's monitors, thus deepening the gameplay and staying ahead in the conflict of interest between developers and adventurous gamers. Web sites served the function of reinforcing and avertising social role and player identity, creating a community memory of causally connected historical events, or a cumulative repository of instructions.

Because the regular gameplay within The Sims Online was so boring, most of what has been recorded about the game on web sites and discussion boards relates to the events generated by the small number of players who played against the grain. These players took charge of the game and moved it in new directions. Evangeline, for instance, who turned out to be a boy, exposed himself and other underage players to forms of adult sexual imagination that may have been harmful and developmentally inappropriate. Jeremy Chase started playing JC Soprano when he was unemployed; his masterful gameplay got him paid work to play online games and design web sites [36]. Peter Ludlow became world famous as the academic that exposed the seamy side of The Sims Online, and his career may take a turn towards active worlds research. In each of these cases, it was the originality of the gameplay that created the real-world consequences. The fictional world blended into the real world because they dragged the real world into the game.

1.4 Conclusion and outlook

The chief creator of The Sims, Will Wright, often cites the work of Christopher W. Alexander as an important inspiration for his work, in particular his A Pattern Language [37]. Communities, Alexander et al. write, emerge out of "a hierarchy of social and political groups, from the smallest and most local groups -- families, neighborhoods, and work groups -- to the largest groups -- city councils, regional assemblies" [9]. The enabling condition that allows such groups to form spontaneously and to constitute communities is that "Each group makes its own decisions about the environment it uses in common. ... Ideally, each group actually owns the common land at its 'level.' And higher groups do not own or control the land belonging to lower groups -- they only own and control the common land that lies between them, and which serves the higher group" [9]. By assigning resources to be held in common at different levels of organization, people would be challenged to institute appropriate collaborative patterns of governance at each level.

In The Sims Online, these architectural principles of community design were never implemented. The community has a flat structure of individual ownership, with no land held in common at any level. In part, this happened by default as the code was ported from The Sims; the single-player model of private, suburban lots was simply scaled up to a multiplayer environment. Far from acknowleding this to be a massive design blunder, however, EA championed private ownership of all land as the key to forestalling griefers, as each player could bar any other from their site. The result is distinctly odd: since many players don't own properties, they must meet their avatar's basic needs by entering other player's private houses, using their shower and bathroom, getting food from their fridge, and even sleeping in their beds. This socially anomalous practice is not only tolerated but encouraged, since owners are rewarded for the time other sims spend on their properties. Indeed, as a result, the primary function of a private property is not that of a home planned and decorated to delight its owner and inhabitant. Instead, properties become investments, designed to optimize return on capital by providing the skilling and working equipment, along with food, gyms, motel-style beds, showers, and bathrooms, to attract and retain the maximum number of visitors. As we saw in section two, the incentive structure of this economic model not only produces very boring play but reduces human relationships to one of instrumentality, grinding any real sociality out of the game.

By leaving out the multiple levels of commons recommended by Alexander, the game designers blocked the formation of effective higher-level community structures within the game. It turned out that the hyper-privatized model was not effective in preventing crime and griefing: Evangeline and others set up their bordellos on their own properties, and the mafia leveraged the power of roommates, harassing chat and e-mail, negative reputation links, and a host of other tools within and outside of the game to achieve their ends. Community-based governing structures, in contrast, had a hard time getting off the ground. Consider the case of the Alphaville Government, established by a group of friends in early 2003. This act involved the creation of the avatar Mr-President, played by Arthur Baynes, and the construction of a Capitol in the best neoclassical tradition. Baynes web site at avg.simsgov.com showed an animated graphic of Mr-President waving from the balcony. The focus, then, was on the trappings of power, and since all properties were private, the Capitol, and later the Court, had to belong to an individual player who might at any time decide to quit. The Alphaville Government also aspired to take on the task of maintaining law and order within Alphaville, but they had no legitimate tools with which to enforce city-wide laws and regulations and never became anything like an effective government.

In the fall of 2004, as this manuscript goes to press, there is widespread expectation among the gamers that The Sims Online is about to close. The initial prospect of hundreds of thousands of customers paying monthly fees of ten dollars each had made the idea of porting their most popular game to the Internet financially extremely attractive to EA. Suneel Ratan, its former vice president, said informal projections had run as high as a million subscribers for the online version, implying a regular annual revenue of $120 million [27]. Instead, TSO subscriptions at the first anniversary were estimated to be around 80,000 [20] and by April of 2004, the company reported 57,500 subscribers [38]. If the attrition rate of 5,000 a month is sustained, TSO will dip below 20,000 before the year is out -- the level at which EA closed Earth & Beyond [20].

Several of the key players have taken the consequences. After being kicked out of The Sims Online in late 2003, Peter Ludlow continued for a while to visit Alphaville through other players' accounts, but soon moved on to Second Life, a very differently managed online social space. In July 2004, Jeremy Chase predicted TSO would close after a final Christmas season and reported he had joined the online multiplayer game Star Wars Galaxies. In August, Simoleanman, one of the main currency traders, announced his closing sale, and The Alphaville Government shut down. It is possible that it's still not too late to rescue the game: the example of the innovative rebels has inspired a host of online newspapers, mafias, and government players, and EA has continued to make significant improvements to the game. A crucial innovation in friendship formation has moved the game closer to a psychology of real relationships. Yet these improvements may have arrived too late. On a recent late-night visit to Alphaville I counted a total of 23 players.

In its heyday, The Sims Online did a spectacular job as a dystopian experiment and remains a rich source of opportunities for social research. Rushed into production, it contains layers of questionable design decisions. Its doll-house ancestry, emphasizing sims driven by artificial intelligence, militated in multiple ways against effective player-avatar identification. The incentive system implicit in its economic structure produced mind-numbingly boring play and purely utilitarian avatar relations. Its lack of strategic opportunities spurred creative players to extend their gameplay beyond the company servers and to bend the built-in features of the game to their own purposes. Finally, the absence of land held in common at different levels prevented the natural emergence of effective layers of governance. Yet precisely by inspiring players to warp the rules and to extend the game to the world-wide web, The Sims Online has succeeded in introducing the larger practical and ethical questions of online community governance to the public at large, making us all participants in a virtual utopian experiment.

References

  1. Grossman, Lev (2002). Sim Nation. Time, 25 November. Available <http://www.time.com/time/magazine/archive/preview/0,10987,1101021125-391544,00.html>.

  2. Seguret, Olivier (2004). Putsch et Sims. Liberation, 27 February. Available <http://www.liberation.fr/page.php?Article=181923#>.

  3. Ward, Mark (2003). The dark side of digital utopia. BBC News, 22 December. Available <http://news.bbc.co.uk/go/pr/fr/-/1/hi/technology/3334923.stm>.

  4. Harmon, Amy (2004). A Real-Life Debate on Free Expression in a Cyberspace City. The New York Times, 15 January. Available <http://www.nytimes.com/2004/01/15/technology/15SIMS.html>.

  5. Manjoo, Farhad (2003). Raking muck in "The Sims Online". Salon.com, 12 December. Available <http://www.salon.com/tech/feature/2003/12/12/sims_online_newspaper/index_np.html>

  6. Keighley, Geoff (2002). The Endless Hours of The Sims Online. Gamespot.com, 28 November. Available <http://gamespot.com/gamespot/features/pc/simsonline>

  7. McKenna, Katelyn Y.A. and John A. Bargh (2000). Plan 9 from Cyberspace: The implications of the Internet for personality and social psychology. Personality and Social Psychology Review 4. 1: 57-75.

  8. Lessig, Lawrence (1999). Code and Other Laws of Cyberspace. New York, NY: Basic Books.

  9. Alexander, Christopher W., Sara Ishikawa, and Murray Silverstein (1977). A Pattern Language: Towns, Buildings, Construction. New York, NY: Oxford University Press. pp. 3-4.

  10. Walker, Jesse (2004). Hobbes in Cyberspace: Life in an online game world proves nasty, brutish, and short. Reason, 1 April. Available <http://www.keepmedia.com/ShowItemDetails.do?item_id=387500>

  11. Electronic Arts (2003). Annual report, 31 March. Available <http://ccbn26.mobular.net/ccbn/7/266/277/>

  12. Gordon, James E. (1978). Structures: or, Why Things Don't Fall Down. New York, NY: Penguin. p. 63.

  13. Salen, Katie and Eric Zimmerman (2004). Rules of Play: Game Design Fundamentals. Boston, MA: MIT Press.

  14. Gibson, James J. (1979). The Ecological Approach to Visual Perception. Boston, MA: Houghton Mifflin.

  15. Maslow, Abraham H. (1943). A theory of human motivation. Psychological Review 50: 370-396.

  16. Green, Jeff (2002). The Sims Online: Indulging your inner weirdo. Gamers.com, May 1. Available <http://www.gamers.com/game/1016135/previews?page=1>

  17. van den Berg, Monique (2004). Simmery and sluttery... the mo pie sims. Available <http://www.mopie.com/sims/simdex.html>

  18. Shakespeare, William (2001). The History of King Lear. Ed. Stanley Wells. Oxford, UK: Oxford University Press. IV.1.36-37.

  19. Schroeder, Ralph (ed.) (2002). The Social Life of Avatars: Presence and Interaction in Shared Environments. London, United Kingdom: Springer. pp.7-8.

  20. Woodcock, Bruce Sterling (2004). An Analysis of MMOG Subscription Growth. Version 10.0. September. Available <http://www.mmogchart.com>

  21. Greenfield, Patricia M. (1997). Culture as process: Empirical methods for cultural psychology. In J. W. Berry, Y. Poortinga, & J. Pandey (Eds.), Handbook of cross-cultural psychology: Vol. 1. Theory and method. (pp. 301-346). Boston, MA: Allyn & Bacon.

  22. Steen, Francis F., Patricia M. Greenfield, Mari Sian Davies, and Brendisha Tynes (2005, in press). What went wrong in The Sims Online? Cultural learning and barriers to identification in a MMORPG. In Peter Vorderer and Jennings Bryant (eds.). Playing Computer Games: Motives, Responses, and Consequences. Mahwah, NJ: Erlbaum.

  23. Kirk, Craig (2004). Culling external sensory response: How we feel videogames. Unpublished manuscript, Department of Communication Studies, UCLA.

  24. Brown and Bell, this volume.

  25. White, A. A. (2003). Chatting for Dummies. Game-Revolution.com, January. Available <http://www.game-revolution.com/games/pc/sim/sims_online.htm>

  26. Kollock, Peter (1996). Design principles for online communities. Harvard Conference on the Internet and Society. Available <http://www.sscnet.ucla.edu/soc/faculty/kollock/papers/design.htm>

  27. Ratan, Suneel (2003). Sim flop dogs game developers. Wired, 12 May. Available <http://www.wired.com/news/games/0,2101,58749,00.html>

  28. Hopkinson, Turner 'Bulldog' (2003). The Sims Online Mob Interview. GameGossip.com, 11 June. Available <http://www.gamegossip.com/comment.php?id=208>

  29. Ludlow, Peter (2004). JC Soprano: Now the city is mine! The Alphaville Herald, 14 January. Available <http://www.alphavilleherald.com/archives/000089.html>

  30. Wadhams, Nick (2003). Sex, mob hits: Sims tests virtual morals. Associated Press, 5 July. Available <http://www.cnn.com/2003/TECH/fun.games/07/05/misbehaving.online.ap/index.html>

  31. Sieberg, Daniel (2003). Next@CNN, CNN, 2 August. Transcript available <http://www.cnn.com/TRANSCRIPTS/0308/02/nac.00.html>.

  32. Twist, Jo (2004). Simulating life, love and the Universe. BBC, 17 September. Available <http://news.bbc.co.uk/1/hi/technology/3645552.stm>

  33. Ludlow, Peter (2003). Evangeline: Interview with a child cyber-prostitute. The Alphaville Herald, 8 December. Available <http://www.alphavilleherald.com/archives/000049.html>

  34. Ludlow, Peter (2003). Comment. The Alphaville Herald, 11 November. Available <http://www.alphavilleherald.com/archives/000016.html>

  35. Ludlow, Peter (in preparation). From Sherlock and Buffy to Klingon and Norrathian Platinum Pieces: Pretense, Contextalism, and the Myth of Fiction. Available <http://www.alphavilleherald.com/archives/Fiction.rtf>.

  36. Chase, Jeremy (2004). Bombshell! JC Soprano retires from TSO, but confesses ALL on the way out. The Second Life Herald, 20 July 2004. Available <http://www.alphavilleherald.com/archives/000331.html>.

  37. Cambron, Melanie (2002). A chat with Will Wright. GIGnews.com, April. Available <http://www.gignews.com/goddess_wright.htm>

  38. Glassman, Mark (2004). Braving Bullying Hecklers, Simulants Run for President. The New York Times, 1 April. Available <http://tech2.nytimes.com/mem/technology/techreview.html>.

 

 

top