Fuse your illusion

In virtual reality, to describe something is to create it

It's hard to get excited about the promise of virtual reality as understood in the mass media. There have been a seemingly endless parade of Hollywood movies such as The Lawnmower Man and breathless science specials trumpeting how `a shared place in cyberspace' will change our lives. For most people, virtual reality means wearing a headset and cumbersome body-tracking equipment. The slow, jagged graphics and clunky headsets developed so far have not been particularly inspiring as a new medium.

Nevertheless, there do exist fully functioning virtual realities in which people both work and play. These systems belong the vast family of text-based virtual realities, more familiarly known as MUDs, MUSHs, MOOs, and so on, available usually through the Internet. These worlds have already achieved true virtual reality by approaching the problem from the opposite direction. Rather than make expensive three-dimensional graphics interactive, text VR systems start with interactivity and leave the graphics out. The frontier of text-based VR, although still primitive, is in some ways more advanced than its graphical counterpart. Text VR is being extended beyond leisure activity to uses in business and research. It is in these spaces that we can also gain a hint as to how virtual spaces will change communication and identity.

Entering a text-based VR system feels nothing like any other sort of computer interface. In text VR, appearance, environment, location, cause and effect, communication and more are all mediated through descriptive text. For example, in order to appear as Thor, the god of thunder, all the user has to do is type describe me as followed by appropriately Nordic features. If another character meets Thor, she might look at Thor and read that description anew. Such places have geography and history. A user can continuously present different appearances, a different genders, or different personalities. Users can meet other people, converse with dozens at a time or secretly whisper dulcet poetry to one person in a roomful of people, create fantastical objects -- in short, anything within the abilities of the system and the descriptive power of language. All this takes place in an environment which itself could be part of a larger theme, such as science fiction, a historical period, or something more eclectic. It's something like a novel with a command-line interface.

This medium has a power which for some is instantly compelling. A magazine writer, previously familiar with only discussion-based systems such as the Well, described his immediately enthusiastic reaction to a popular system known as FurryMUCK:

I'm a lurker on places like the Well ; I like to read but I almost never post. I can't lurk here, because I have an almost physical presence in a place that itself has dimension, color, size and scope.

I want to belong to this community and learn cool spells. I want to be a virtual Yenta. 1

There were no spaceships to be shot, no graphics at all, just descriptions of localities and prompts asking players where they wished to go or what they wanted to do next. [...] The charm of the game lay in the illusion it gave players of being inside the game universe. It engaged the imagination in a way that no game had done before.2

In a world of text, to describe something is to create it. In their rapid evolution, text VR has acquired a whole range of features for communicating with other players, the creation of new objects in the environment, and reacting to the environment. Users' creative behaviour falls into three main categories: communicating, describing, and programming.

In the social MUD, all manner of improbable communication is allowed since the purpose is unstructured. "hivemind leaps onstage!" could accentuate a point or entertain those in the room based on the context. It is an example of emoting or posing. In Textuality in Cyberspace, Jeffrey R. Young notes that characters are not speaking but writing to one another. There are some advantages, since the written message can be reflected upon for more than the spoken.4 Even in real life, there is a blurry line between writing and speaking, but in text VR there is no distinction whatsoever. It's typical to weave in third-person descriptions of your own reactions with first-person dialogue while you send a second-person message to someone else, such as "You cause hivemind to fall down laughing." It is easily possible to hold two or three conversations at the same time, since people type in short bursts and it is relatively easy to separate out different voices even in a crowded room. Also, only on a text VR system can you entertain guests in your living room while having sex with someone else by mental telepathy. Users distinguish themselves and gain power by the quality of their prose and the werewithal to handle numerous threads of discourse.

As for the second category, although a wide range of description is possible, users' creative behaviour often falls into certain genres. Nobody ever builds a car, despite its high status in real life. It would be functionally useless, since users can often instantly teleport anywhere in most text VR systems. More importantly, anyone can describe a BMW. Descriptions are usually intended to convey information about the person who created them, either by their choice of subject or the quality of the prose. Conformity is pointless; users stand out by diverging from the commonplace. Thus experienced users tend not to describe themselves as physically perfect according to mainstream aesthetics, despite what many mass media reports have suggested. In fact, virtual life is more of a cacophony of divergent, postmodern identification.

A third way of leaving an imprint in the virtual world is to become a programmer. Most systems have a higher class of individuals who are allowed to modify or add to the environment. Among the possibilities for representation are objects with programmed behaviour, like a camera left behind in a room to spy on others, in the form of communicated 'pictures' to its absent owner. Some program what amounts to extended, complex theatrical scenes which can be called upon at the appropriate moment -- such as having an annoying person removed by Klingon warriors.

In short, in a social text VR system, creative user activity is (not surprisingly) entirely focused on information about and representation of virtual identities. There is an ongoing debate whether text VR is a game with lifelike qualities or an extension of real life with gamelike qualities. I believe from how the users really interact and use the system that the latter is true. This statement is controversial, even to many experienced text VR users, but the examples below help substantiate this claim.

The users of computer-mediated communications have developed an elaborate array of ways to be friendly and are often over-effusive and demonstrative of affection in cyberspace, much more than the North American real-life norm. This helps gloss over the imperfections and underlying constructions in computer-mediated communication. Nevertheless, some can abuse the system and the very nature of virtual reality for violent purposes. I must preface this discussion with by noting that violence is hardly the most important, most prevalent, or even the most interesting aspect of virtual reality. However, it is when these cozy social environments break down that their elements become more visible.

Many people have difficulty imagining how someone could be violated on the Internet. Some may wonder why the victims did not simply switch their computers off and walk away. The famous "cyberspace rape" on LambdaMOO was done, technically speaking, by creating puppets who could partially imitate the parties concerned. The victims did not lose control of their characters -- that would be much more difficult to accomplish. Instead, the virtual rapist merely set up a situation where others in the same room as the victims would see descriptions of the violence as well, as feigned pleased reactions from the victims. Unlike a real-life rape, the perpetrator had to perform his action in public for it to have the desired effect.

Although such puppets can be used in a playful way to poke fun at others, losing control of one's virtual character or one's virtual environment is always unsettling. At the other extreme, players of combat systems can drop into extreme depression when their characters `die', since their access to the system has been cancelled, along with the items, power and fame they had acquired and the friendships they made through that identity. Although it took place on a computer screen, the trauma that the victims of the original LambdaMOO cyberspace rape described was real.

In Computers as Theatre5, Brenda Laurel argues that interfaces are not simple divisions between the computer's circuitry and the human's consciousness. Instead, both parties are representing things to each other, and the interface constitutes the sum total of those representations. Thus, both computer and person "act" together on a stage in a spotlight of representation. The computer's digital switches and the human's thoughts are unseen, like manipulations behind the stage or the faces of the audience.

Laurel stresses that with computers "the representation is all there is". The implications of this theoretical statement are enormous, and help explain some of the aspects of text VR systems. There is no such thing as "real life" for a character except for when the person behind it chooses to include such aspects. Control over the representation of one's character constitutes the virtual body. When others invade that control in a malicious way the analogies to violence are well-founded. These transgressions become possible in virtual spaces because representations, or identities, are always negotiated between the person and the environment, even in real life.

Adding the social realities of text VR systems, and how transgressors can destroy that sense of community (which is, after all, built entirely through communication and shared representation) it becomes easy to understand the shocked reaction to such crimes against identity.

However, by far the most written-about aspect of text VR is the ability to take on multiple identities. In her M.A. thesis, "Identity Workshop",6 MIT's Amy Bruckman described text VR systems as positive places where different components of one's personality could be aired. Gender and appearance are frequent channels for the creation of a fantasy persona, which can offer release for the "real" person.

As much of the population of text VR consists of college-age males, the issue of mastery as described by Sherry Turkle7 becomes an issue. Many of the game systems have levels to rank the players, with the ultimate goal of becoming a `wizard'. To `level', slang for advancing a level, one must accumulate points through feats of daring-do or money-making schemes.8 The worlds are often based on fan universes such as Star Trek, mining an existing repertoire of images for a shared context. The relatively safe and controllable worlds inside these systems are a powerful draw for many -- some play more than forty hours a week. Many are driven by ambition to control the virtual world; others see it as a refuge where their best friends are. Some have suggested that MUD stands for Multiple Undergraduate Destroyer,9 a reference to the many courses failed due to MUD playing.

Pavel Curtis, a pioneer of text VR systems and operator of LambdaMOO at Xerox PARC, sees the addiction to virtual spaces as an ongoing problem which society will have to come to terms with. He notes, however, that this form of addiction is precisely the opposite of most others. Instead of shrinking into a solipsistic world of drugs, for example, the virtual-space addict interacts compulsively with other people.10 Bruckman points out, however, that the "don't-these-people-have-lives" argument makes perhaps improper value judgments on what constitutes wasted time and proper social interactions.11

Some researchers believe that the best model for computer-mediated communications of the future is, or will be based on, text VR. For example, after an exhaustive search of communications technologies, USENIX, an association of computer professionals, decided on a private MUD system as a cheap and powerful solution for their conferencing needs. According to the group, the MUD "solved all [their] communications problems and had some unexpected benefits".

The largest unexpected benefit of the MUD is that it created a social environment that didn't exist. People have real conversations on the MUD with other participants. These are typically, but not always, about systems-related issues. Because of this, members of the group find out about projects that they're not involved in. The new undergraduate volunteers come to know senior members of the staff that they would otherwise not recognize. The MUD has become a social place for the systems group that is populated even when everyone is logged in from home. It may be said that this takes away from work hours, or cuts down on effectivity while in the office. But, over time, we have seen our systems group, people who had simply worked on related jobs, grow into a real team. This change is due partly to our shared social context.12

Xerox PARC's Jupiter Project, headed by Curtis, recently published a manifesto describing directions for virtual reality interfaces to the Internet.15 Through a fictional narrative, the document expresses an optimistic, almost millenial vision of renewed community through the creation of public electronic spaces. Curtis' LambdaMOO could be considered a preview of this future electronic democracy, where the structure of society is determined by petitions, arbitrative hearings, and an Architectural Review Board. There are slow inchings as well towards more formal constitutions, including a Bill of Rights. However, it's not obvious that the residents of LambdaMOO are any more involved in the political process than in real life, other than by voting on the five or six petitions that arise in any given month. Also, given the text-only, sometimes ponderous environment embodied in text VR right now, some might find the Jupiter Project's vision implausible.

However, it hardly needs pointing out that Xerox PARC's last contribution to computer interfaces was the now ubiquitous window-icon-mouse combination. Before long, many ordinary computer users may see their singular identities scattered between 'real life' and virtual spaces.

2 "Collaborative Networked Communication: MUDs as Systems Tools", USENIX Association: Proceedings of the Seventh Systems Administration Conference (LISA VII). pages 1-8, November 1993, Monterey, CA.

3 MUD, Object-Oriented. "Object-oriented" refers to a sophisticated array of objects that can be called upon to construct new ones.

4 Jeffrey R. Young, Textuality in Cyberspace: MUDs and written experience" May 1994.

6 Amy Bruckman, "Identity Workshop:Emergent Social and Psychological Phenomena in Text-Based Virtual Reality", Master's thesis, MIT Media Laboratory, 1992

7 Turkle, Sherry (1984). The Second Self: Computers and the Human Spirit. New York: Simon & Schuster.

8 Most MUDs have a currency system to enable players to exchange goods and services. Inflation is a major worry in these systems. Wizards maintaining this currency have powers similar to a central bank with monetary policy, and some MUDs have been used as economic simulators.

9 Elizabeth Reid, "Cultural formations in text-based virtual realities", Master of Arts thesis, Cultural Studies Program, Department of English, University of Melbourne, January 1994

10 Reid, op.cit.

11 Howard Rheingold, The Virtual Community: Homesteading on the Electronic Frontier, Addison-Wesley, 1993, 168

12 USENIX Association, op.cit.

13 Rheingold, 172

14 Amy Bruckman and Mitchel Resnick, "Virtual Professional Community: Results from the MediaMOO Project", draft presented to the Third International Conference on Cyberspace in Austin, Texas, 1993

15 Jupiter Project Team, "Not a Highway, but a Place: Joint Activity on the Net", Xerox Palo Alto Research Center, January 1995