A Genealogy of Virtual Worlds
I must ask you now to join me in a detour from my account of life on LambdaMOO while we consider just what sort of map a place like LambdaMOO might be, and how it got that way. I must ask you, in other words, to delve with me into a brief genealogical history of the MOO, beginning roughly in time immemorial.
The vastness of the time frame is inevitable, I'm afraid, for any historically complete taxonomy of the human innovations ancestral to LambdaMOO must really start where humanity itself did: at that elusive evolutionary moment when the strictly private act of imagination blossomed into the preeminently social one of representation, and the machinery of culture was born. Language, narrative, ritual -- all of these are engines for the creation of virtual realities, and always were, for always they have served first and foremost to allow two or more minds to occupy the same imaginary space. And always that imaginary space has stood as a challenge to technology, or maybe a plea: to make it more vivid, more substantial, to give it a life of its own. Primitive inscription was the earliest device to answer the call; painted cave walls and graven clay tablets lent images and words for the first time a kind of autonomous existence, independent of the bodies whose fleeting speech and gestures had hitherto bound them. But the drive to perfect the technology of representation hardly stopped there, needless to say, and it's nothing less than the entire history of this drive to perfection that comprises the proper genealogy of VR -- the full record of every technique ever devised for making the shared illusion of representation come more convincingly alive, from the venerable conventions of perspective drawing and of the realist novel to the latter-day wizardries that have given us photography, television, Disneyland, and 3-D, smellovisual, surround-sound cinema.
It is possible, however, and in the end probably more enlightening, to tell a less ambitious story about the lineage of LambdaMOO. For just as nothing puts us humans more precisely in our place amid the abundant and interconnected branches of life's family tree than the observation that we are descended from apes, so too the MOO's place in the epic evolutionary history of the virtual is perhaps best grasped by considering the relatively simple fact of its descent from maps.
Whence maps themselves arose, I couldn't rightly say. As for their present-day status as a pet metaphor of certain delirious strains of postmodernism (according to which the image of a huge map overgrowing and ultimately replacing the territory it charts -- Jean Baudrillard's "finest allegory of simulation" -- condenses everything you need to know and dread about the decay of the real in contemporary culture), I assure you it isn't theoretical modishness that leads me to locate the origins of the MOO in the invention of cartography. Any close encounter with a map is all it takes, really, to sense the embryonic MOO-space embedded within it. Just look at a map yourself for a while and try, as you look, to resist the urge to imagine yourself transplanted into the tiny territory spread out before you, riding the tip of your own colossal index finger down toy rivers and over minute mountain ranges, hopping flealike from city to city as your giant gaze flits across the chartscape. More than most other traditional ways of representing the world, maps conjure a vision of representation itself as a space the viewer might enter into bodily, a construct not merely to be comprehended but to be navigated as well. They invite interaction, and of course they frustrate it too: their smooth surfaces remain impenetrable, like shop windows, inspiring in the most avid map-gazers a yearning that has less to do perhaps with simple wanderlust than with an ancient dream of literal travel into the regions of the figurative.
Small wonder, then, that the earliest appearances of maps seem to have been followed not long after by the first attempts to shatter their surfaces and place the viewer (or a serviceable representative thereof) inside them. Board games is what we would call those attempts today, but that shouldn't keep us from recognizing them as crude realizations of the map's implicit interactivity. Nor should it dissuade us from suspecting that the impulses behind their invention were far from trifling. After all, the oldest game of all -- the casting of lots -- began as a device for divining the will and wisdom of gods, and the history of games in general remained entwined for millennia with that of religious and magical ceremony. Is it such a stretch, then, to speculate that the oldest of board games -- which seems to have been a prehistoric, northeast-Asian sort of Parcheesi in which tiny horsemen raced each other around a circular chart not dissimilar in design to the earliest maps of the world -- enacted for its players a voyage through the shadow world of the imagination, the world that gods dwelt in and that the still-novel technologies of representation brought to life?
Not that the game wasn't also, undoubtedly, something very much like fun. But even fun has its serious dimensions, and in the case of map games (to coin a term distinguishing these board games from those like Scrabble, for instance, that don't in effect represent any sort of navigable territory) the fun to be had has always to an exceptional degree depended on and referred back to the dread seriousness of fate. The ancient racing games evolved quickly into games of battle like checkers and chess, and much later into economic contests like Monopoly and the Game of Life, but what has remained a constant in their appeal is that they quite literally map the real world of day-to-day and ultimately life-and-death existence onto the timeless and ultimately inconsequential realm of the imagined. They promise a temporary escape from the inescapability of history (whether personal or global) into a place where history is just a simulacrum built of rules, turns, strategies, and dice rolls, a weightless flow in which no outcome is so fatal that it can't be rewritten the next game around. After all these years, in other words, the map games continue to show their religious roots, since even our simple, secular delight in these rough-hewn virtual worlds turns out to be, in a sense, just another way of wrapping our hearts and minds around religion's primal conundrum: the cosmic raw deal that gave us each just one life to live.
Still, secular delight is also, in another sense, simply its own reward, and if the tension between reality and unreality was always the source of the map-gamer's delight, then it stood to reason from the outset that a heightening of that tension would increasingly be sought by players as the games evolved. With other sorts of games of course, gambling has long been the preferred means of flavoring the airy fluff of play with the rugged feel of real-life results, but tellingly enough, this quick-and-dirty injection of genuine fate never became much of a fixture of map games. Instead, starting with the archaic precursors of chess, they have more often borrowed from real life not its consequences but its complexity. With the arrival of chess itself in courtly sixth century India -- and perhaps too with the later development of the arguably cartographic East Asian game of Go -- the map game attained a degree of tactical intricacy that remained unsurpassed for hundreds of years, suggesting perhaps that for the time being the form's evolving complexity had actually outpaced that of social reality.
By the middle of this century, however, reality was catching up with a vengeance, and for the first time map games of a significantly woolier design than chess's began to appear. Inspired, no doubt, by the increasingly media-blitzed busyness of the postwar information landscape -- and nurtured, obviously, by the sudden abundance of leisure time in postwar consumerist societies -- these new games carried out their inherited role of simulating history with an unprecedented and often overwhelming attention to detail. Their earliest exemplars were the monumental war games produced since the 1950s by the Avalon Hill company (and still clinging tenaciously to their sizable niche market to this day): played on towel-sized, geographically precise maps of combat sites like Gettysburg, Stalingrad, or Waterloo, encrusted with arcane rules and timetables designed to model actual conditions of battle, and littered with hundreds of miniature playing pieces all subtly different from one another in their designated abilities, the games demanded a certain obsessive fortitude just to get through the instructions, let alone to commit to the hours, days, or even weeks a single game might take to play.
But even these tabletop sagas proved to be light diversions compared to the groundbreaking genre that emerged from their midst in 1973, when two veteran wargamers named Gary Gygax and Dave Arneson introduced a new game they called Dungeons and Dragons. Abandoning the typical military-historical setting in favor of a mythical age peopled by wizards, dwarves, elves, and other Tolkienesque entities, D&D (as millions of aficionados would later routinely abbreviate it) took wargaming into a whole new conceptual world as well, turning it into an endeavor so involved and involving that it became, in some ways, difficult to recognize as any sort of game at all.
The most obvious of D&D's novelties, perhaps, was its near-total indifference to what had until then supplied the formal cornerstone of virtually every game in existence -- direct competition between players. Collapsing the wargamer's swarming battlefield of units into a single heroic character loaded with dozens of precisely defined attributes, skills, and possessions, the rules didn't prohibit player-characters from fighting against each other, but they made it much more interesting for them to band together instead and set off on lengthy, shared adventures. These adventures were designed and refereed by a godlike metaplayer known as the dungeon master, who threw potentially lethal monsters and other dangers at the players and awarded ever-more-impressive attributes, skills, and possessions to the survivors in accordance with a mind-numbingly complicated set of rules. Roughly speaking, then, there was a point or two to it all, but winning wasn't one of them. In fact, nobody ever clearly won the game, and for that matter no game ever clearly ended: players simply battled on from adventure to adventure until their character was killed, at which point they felt a little sad, maybe, and then created a new character, so that in principle games might go on for as long as anyone cared to play them. In practice, they often lasted years.
Such elaborately structured open-endedness brought map-gaming closer than ever, of course, to the free- form complexity of real life itself, and this was no small contribution to the evolutionary history of virtual worlds. But in the end, D&D's truly pivotal role in that history should really be credited to a subtler breakthrough: its slight yet radical redesign of the millennia-old relationship between the board-game player and the board. Dungeons and Dragons succeeded as no game ever had at slaking the ancient desire of the map-gazer to enter the map, and it did so, paradoxically enough, by simply taking the map away. Drawn up fresh by the dungeon master with every new adventure, the D&D map remained hidden from the players at all times, its features revealed only as the players encountered them in the course of adventuring, and even then only by the DM's spoken descriptions. Gone was the omniscient, bird's-eye perspective that had always undercut map-gaming's illusion of immersion, and in its absence game-play took on a near-hallucinatory quality so integral to the experience that the official Players' Handbook now actually begins with vaguely shamanistic tips on how best to achieve it:
"As [the dungeon master] describes your surroundings, try to picture them mentally," advises the manual, walking novices through a hypothetical labyrinthine dungeon. "Close your eyes and construct the walls of the maze around yourself. Imagine the hobgoblin as [the dungeon master] describes it whooping and gamboling down the corridor toward you. Now imagine how you would react in that situation and tell [the DM] what you are going to do."
What had happened, in effect, was that the cloaking of the map had also hidden the player's token self, the game-piece, thereby compelling the player to put himself psychically in its place. As a result, D&D players weren't merely represented by their richly detailed characters -- they were identified with them, in a relationship so distinctively intimate that in time it came to be recognized as the definitive feature of both D&D and its scores of eventual imitators, which to this day are known generically as role-playing games. As apt as the name is, however, it doesn't do justice to the breadth of the innovation, for the same mechanics that made D&D's style of role-play so vivid also made D&D more than just a new kind of game. They made it, frankly, a whole new mode of representation -- an undomesticated crossbreed, combining the structured interactivity of the map game with the psychological density of literary fiction, yet eluding the ability of either medium to fully embody it. Indeed, the grab-bag of primitive media actually used in playing Dungeons and Dragons -- pencil and paper for making maps, dice for resolving combat situations and character details, and the spoken word for just about everything else -- tended to give the impression that the technology hadn't yet been invented that could single-handedly manage the unwieldy hybridity of the new form.
The impression was a false one, however. The technology had been invented, three decades earlier in fact, when a small army of North American engineers perfected a species of overgrown calculator known as the all-purpose digital computer -- and in the process inaugurated what might reasonably be considered the single most revolutionary moment in the history of representation since the emergence of language. Even before the computer existed as functional hardware, the theoretical work of mathematician Alan Turing had established that the device was no mere number-cruncher, but rather the ultimate representational Swiss Army knife, a universal simulator capable in principle of symbolically recreating the dynamics of any real-world process it was possible to imagine. Like the map game, then, only on a much grander scale, the computer was a tool for creating artificial history, and by the time Dungeons and Dragons appeared, computer scientists had long been peering into their machines to watch such complicated and consequential events as rocket flights, managerial decisions, and World War III unfold in the weightless, adjustable atmosphere of digital make-believe.
In comparison, obviously, the simulation of an adventurous romp through faerie posed scarcely a challenge to the technology, and given the abundance of free time, enthusiasm, and sword-and-sorcery geeks among the junior code-slingers of the day, it was really only a matter of time before someone did the requisite programming. In the event, it was three years after D&D hit the stores that a pair of evidently underworked Palo Alto hackers by the names of Will Crowther and Don Woods wrote the world's first computer-based role-playing game, an instant classic known variously as ADVENT, the Colossal Cave Adventure, or simply Adventure.
Formally speaking, there was little about the game that any D&D player would find surprising. The principal setting was the bowels of a cavern crowded with dwarves, dragons, and magic treasures, and though the position of dungeon master was gone, the DM's basic functions were performed transparently enough by the game's underlying code. Written descriptions appeared onscreen in elegantly sparse but otherwise entirely standard DM-speak ("YOU ARE IN A MAZE OF TWISTY LITTLE PASSAGES, ALL ALIKE," "YOU ARE IN A VALLEY IN THE FOREST BESIDE A STREAM TUMBLING ALONG A ROCKY BED"), and adventurers typed in stripped-down versions of typical D&D player- statements ("GO SOUTH," "DROP SWORD," "KILL DRAGON") to which the program gave equally typical responses ("KILL THE DRAGON WITH WHAT, YOUR BARE HANDS?"). Yet from a psychological perspective, Adventure's automation of the dungeon master was clearly no trivial modification. For as a direct result, the rules that defined the game world suddenly felt a good deal more like those that defined the physical world. No longer dependent on a human referee's always revocable agreement to abide by them, the binary-encoded laws of Adventure were maintained instead by the same sort of logical machinery that had always enforced the laws of nature: a nonnegotiable procession of unthinking causes and inevitable effects. Any moderately skilled programmer could always stop the game and rewrite its rules, of course, but for anyone in the midst of exploring it, the world of Adventure was as hard-wired as gravity, and almost as convincing.
One particularly lifelike element no one would find in that world, however, was other people. Quite unlike Dungeons and Dragons, you see, Adventure was a solitary entertainment, pitting a lone player against the creatures of code that dwelled in the software recesses of the Colossal Cave. It was also a high-quality, addictive entertainment, to be sure, and wildly popular in computer labs throughout the English-speaking world. Yet anyone who came to the game seeking role-play at its richest was bound to sense something missing -- and once again, as with the earlier leap from D&D to Adventure itself, it was really just a matter of time before some inspired young programmer took on the task of completing the picture.
But time was aided, too, in this case, by historical coincidence, for it happened that the high-techies of Adventure's early years were just starting to get used to a fairly radical notion about the computer: namely, that it was an ideal tool for connecting its users not only to complex, abstract realms of logic and data, but to one another as well. The technology of computer-mediated communications had been in its infancy at the start of the '70s -- when the first nodes of what would later become the Internet were sprouting in Pentagon-fertilized fields of academe -- but it grew steadily, and in the final year of the decade its coming of age was signaled by a cluster of landmark developments. The earliest computer bulletin boards had been wired into the phone system by pioneering PC hobbyists the year before; the first commercial online services opened for business not long after; the first Usenet newsgroups began to circulate, stirring up a hint of the vast global storm system of discussion they would eventually grow into; and last but assuredly not least, code-smiths Roy Trubshaw and Richard Bartle, both of them undergrads at Britain's University of Essex, spent much of the year putting final touches on a program that would at last fulfill the promise of computerized role-play, allowing two or more geographically distant players to enter the game at once.
Its name was MUD. The MU stood for "multiuser," and much later, as you may recall, the D would commonly be taken to stand for "dimension," but at the time what it really stood for was "Dungeon," in homage to a popular Adventure knock-off known by that name. The game space itself likewise leaned heavily on Adventure-inspired conventions, as the generically evocative look and feel of the opening description made plain:
Naturally, The Land was filled with the usual automated hobgoblins and hidden treasures as well. But there, precisely, MUD's debts to its predecessors ended, because of course The Land was also filled with real live people, and their presence introduced new elements of surprise and camaraderie into computer- adventuring's clockwork worlds. And these elements, in turn, raised the attraction of those worlds to an apparently irresistible level. By early 1980, the DEC-10 mainframe on which the new game was installed had been opened up to log-ins from beyond the university, and it wasn't long before the machine was swamped with an influx of players so hooked that a near-total ban on outside MUD connections (permitted by school authorities only between two and six o'clock in the morning) did little to discourage them. "Even at those hours," Richard Bartle later recalled, "the game was always full to capacity."
Such a hit was bound to spread, of course. Requests for copies of the game's core operating system started coming in from around the world, and Bartle honored them, exporting MUD code to Norway, Sweden, the U.S., Australia (where in time the games' network-clogging proliferation would lead to an official, continentwide prohibition of them). Inevitably, other hackers took to revamping and reinventing the program -- streamlining its inner workings, adding to the diversity and realism of its features. And wherever new variants appeared, new worlds were built around them, often retaining the stock sword-and- sorcery thematics of the original MUD, but increasingly veering off into realms of almost fetishistic specificity. Devotees of Anne McCaffrey's dragon-happy fantasy novels stepped into great scaly text- bodies to roam detailed recreations of the books' faraway planets; Star Trek fans built vast working models of the Enterprise and sailed them off through MUD-space; college students erected simulations of their schools and spent nights slashing giddily away at monstrous, digital parodies of their professors.
Hundreds and thousands of person-hours went into the collective design of these games, and many more went into the often-passionate playing of them -- and all the while the culture at large obliviously looked elsewhere for visions of the mind-bending dream-tech of artificial worlds it was beginning to sense computers had in them. Millions got their first glimpses of the dream in early-'80s science fictions set amid the gleaming, corporate geometries of a place most memorably referred to (by novelist William Gibson) as cyberspace, and millions more saw it later in breathless media accounts of goggles-and-gloves contraptions being patched together by starry-eyed Silicon Valley capitalists, yet few people outside the MUDing crowd seemed to realize that a global VR industry of sorts was already cranking out one lucidly believable digital microcosm after another, more or less just for the fun of it.
And even among the MUDers, it's safe to say, not many saw with clarity just what an oddly substantive sort of fun their pastime was on its way to becoming. Right up to the end of the '80s, after all, all MUDs were still at least ostensibly nothing more than games. Granted, they were impressively elaborate games -- no less free-wheeling and engrossing than the pencil-and-dice role-playing epics they descended from -- but they were games nonetheless, with specific adventures to be pursued, puzzles to be solved, and typically, hierarchies of points-based levels to be ascended (leading ultimately to wizard grade and the right to build new regions and adventures into the game). Even so, however, MUDers had long noted the marked tendency of the game space to become a social space as well. Players not infrequently stepped outside the game without leaving the MUD, going "OOC" (or out of character) to hang out amid the passing adventurers, to haggle over administration of the game and its resources, to deepen the genuine friendships and authentic antipathies formed in the midst of play. Something very much like real community was coalescing at the edges of all that make-believe, in other words, and though such virtual communities were hardly rare in the online world, nowhere did they enjoy as richly nuanced and concretely grounded a setting as amid the gesturally expressive make-believe bodies and psychically immersive make-believe landscapes of which MUDs were constructed.
Despite their principal deployment as games, then, MUDs were more than just incidentally serviceable as a medium for broader forms of social intercourse. They were in fact ideally suited for the role. And it may be that a recognition of that fact was what led, late in the summer of 1989, to the final significant turn in the technological path to LambdaMOO. Or it may not be. James Aspnes, the Carnegie-Mellon grad student who took that turn by creating TinyMUD, the first of what would eventually be referred to as the "social MUDs," certainly didn't seem to think he was inventing anything but a more fluid adventuring environment. "I wanted the game to be open-ended," Aspnes wrote later, explaining his decision to leave the conventional framework of player-rankings and fixed goals out of his new MUD. And open-ended the MUD indeed turned out to be, though hardly in the familiar, structured manner made standard long before by Dungeons and Dragons. The truth was, TinyMUD really had no structure at all -- it was more or less literally whatever its players wanted it to be. With building privileges no longer limited to a wizard class, the topology of the MUD quickly came to reflect the diverse whims and backgrounds of the inhabitants, with virtual Taiwans popping up next to virtual Cambridges, and Wesleyan University steam tunnels leading to the buildings of a University of Florida campus. In time there was even a full-scale replica of Adventure to be found somewhere on the grounds, though it's unlikely many TinyMUDers ever sought it out. For it was clear enough by then that, whatever James Aspnes's original intentions may have been, people didn't really come to TinyMUD to play games.
What they did come for wasn't exactly easy to pin down, but neither was it all that hard to understand. They came to create, for one thing -- to build spaces and construct identities. They came, too, to explore the sprawling results of all that creation. But mainly they came for the simple reason that other people came as well. They were there to talk, to tell jokes, to make love and fall in it, to bitch and bicker and backstab. They were there, in short, to make human contact, which by a hardly remarkable coincidence seems also to be what most people are on this planet for. Even less remarkable, then, are the facts that TinyMUD, which its creator had expected to "last for a month before everybody got bored with it," instead grew fat and thrived in various incarnations for years; or the fact that it inspired a miniboom in the construction of MUDs generally and social MUDs in particular; or the fact that its success almost instantly began to attract the attention of scholars and professional media developers, intrigued by the now amply demonstrated depth and versatility of MUDs and eager to explore their limits.
And what of the fact that the earliest of such high-minded investigations was initiated by a thirty-year-old Xerox researcher called Pavel Curtis? Surely, in the context of the grand evolutionary narrative we've been tracing, that particular point of information is among the least remarkable of all. But as it is the point upon which the entire narrative converges, let it be noted: that on the morning of the day before Halloween, in the year 1990, Pavel Curtis issued the command that for the very first time summoned into existence LambdaMOO, a social MUD in the classic mold, with little at that point to distinguish it from the general run of TinyMUD's progeny aside from its exceptionally powerful set of world-constructing tools (built into the original MOO code by its author, Stephen White) and the fact that a major multinational corporation would be keeping a close watch, through Curtis, on the world LambdaMOO's players constructed with those tools.
Of course, given the relatively hands-off nature of the experiment, even the latter distinction didn't ultimately make much of a difference to life within the MOO. Nor might it have meant much outside the MOO either, had the multinational corporation in question been a different one. But inasmuch as Curtis worked for the same Xerox think tank that had essentially dreamed up the personal computer from scratch a decade and a half before (only to watch helplessly as Xerox marketers dropped the ball and a tiny startup by the humiliatingly cutesy name of Apple carried it into the endzone), his employer-sanctioned interest in MUDs more than whisperingly suggested that they might contain the seeds of the next revolution in the nature of the human-computer interface.
Thus, where TinyMUD had cleared the way for research into MUDs as a serious technosocial phenomenon, LambdaMOO ushered the new field in with a loud and legitimating fanfare. Before long, ethnographers, sociologists, and literary theoreticians were poking their heads into the nearest MUD for an often illuminating and invariably gratifying glimpse (here was a world, after all, in which the social construction of reality wasn't a matter merely of academic dogma but of basic physics), and the Net was peppered with research-oriented MUDs that went beyond LambdaMOO's ant-farm experimentalism into areas of ever-more pragmatic application. There were MUDs designed to teach kids about science and programming while they played, local-area MUDs where teams of office workers gathered to coordinate ongoing projects, a MUD where far-flung astronomers came to trade observations amid the whirling orbs of a virtual solar system, and even, perhaps inevitably, a MUD reserved for media researchers who felt like getting together to schmooze about, well, MUDs mostly.
So that by the summer afternoon of 1994 on which I showed up at LambdaMOO to wrestle with the curious case of the dislocated television set, the world I happened to be coming home to was but a single and not especially singular member of an increasingly diverse ecology of such worlds. The three or four hundred MUDs now up and running embodied a range of applications stretching from the still very popular hardcore adventure games through the more broadly focused social and research MUDs and on out to the first limited prototypes of schemes in which the entire Net might someday be blanketed by one big MUD, its code distributed across all the world's computers and its sprawling terrain providing context for every type of digital interaction conceivable. More and more, as well, the tens of thousands who inhabited these worlds were dividing into loose and loosely antagonistic subcultures reflective of their divergent interests, with habitués of the social MUDs sometimes jocularly disparaging the adventure worlds as so much "hack-and-slash" childishness, and adventurers in turn dismissing the social worlds as "chat systems with furniture."
Despite the growing differences between MUDs, however, it was an underlying unity that still ultimately defined them. For just as there had never been any MUD so steeped in playful make-believe that it wasn't also fertile ground for serious emotional connections among its players, likewise there was yet no MUD so dedicated to serious purposes that it could do without the elements of playful make-believe that made it function. All MUDs, that is to say, existed in a conceptual twilight zone between the games from which they had evolved and the real-life social meshes they had come to resemble, and at bottom it was in this irreducible ambiguity -- rather than in any of the increasingly various uses to which MUDs were being put -- that their deepest significance lay. They constituted neither an escape from historical existence nor simply an electronic extension of it, but rather a constantly disputed borderland between the two --between history and its simulation, between fate and fiction, between the irrevocable twists and turns of life and the endlessly revisable possibilities of play.
If I make any great claims for the curiousness of LambdaMOO, therefore, understand that they are really only claims on behalf of MUDs in general, and also, perhaps, on behalf of what can really only be called the human condition. Like all MUDs, you see, LambdaMOO was still essentially a map, and like all MUDs it mapped a place as yet uncharted by conventional cartographic means: the strange, half-real terrain occupied by the human animal ever since it started surrounding itself with words, pictures, symbols, and other shadows of things not present to the human body. It's a place we're all well-acquainted with, of course, since we live in it from the moment we begin to talk till the moment we have nothing left to say. But have you never noticed how seductively exotic even the most familiar ground can come to look, when it is looked at in the tiny abstractions of a map?
The above article is an extract from a book by Julian Dibbell, "My Tiny Life: Crime and Passion in a Virtual World".