Copyright 1995 by Jen Clodius. This is a draft, attempting to outline just what it is I'm trying to write about in my exploration of the DragonMud community. Originally a rough draft for a grant proposal, this is beginning to show its age.
The Internet democratizes information; the Internet will destroy people's ability to think critically. The Internet enhances communication; you can't tell what people really mean without non-verbal cues. Gender lines are blurred on the Internet; there aren't enough women on the net to overcome its male orientation. Pedophiles and pornographers lurk on the net; distant couples establish and maintain long-term relationships. People reach out to help a net-pal in distress; the Internet is a cold, lonely place that will turn us into a generation of social isolates who avoid personal contact.
These are just some of the contradictions in what has been written recently about the Internet, however little has been written about how people actually interact within this international network of computers. To date, no ethnography has been written about a specific "community of interest" that has formed on the net [1].
I have been conducting preliminary fieldwork for the past three years on DragonMud, the oldest continuously operating Multi-User Domain [2] on the international network of computers known collectively as the Internet. A "Multi-User Domain", or MUD, is a kind of computer program that allows numerous people to log into a single computer and use the same program simultaneously. Communications on a MUD operate differently than other forms of computer-mediated communication; email is one-to-one and asynchronous, bulletin boards, news groups, and mailing lists are one-to-many and asynchronous. Unlike "chat" systems, a MUD creates a sense of "space" which allows player-movement between "rooms". Users have the ability to change and modify the database. The conversation "scrolls up on the participant's screen looking like a play script" (Riner and Clodius 1995, 99), or as "a communication soup in real time, with a flavor of improvisatory theater" (Rheingold 1993, 149).
DragonMud's inhabitants, the spaces in which they dwell, and the objects they create and manipulate all exist within a realm comprised solely of text (Clodius 1994, 12). Participation in a MU* [3] is recreational; people log on by choice (Huizinga 1949, 7). Encounters between specific individuals may be happenstance, but everyone who logs on does so expecting to interact with other people on some level. Time zones and geographic boundaries are meaningless on the Internet, and people communicate and form friendships across a multiplicity of boundaries on DragonMud, broadening their perspectives and expanding their horizons. Naibu recently commented, "I'm not all that antisocial... it's just that in my life, everyone I know is just like me! This mud has made me able to meet people from backgrounds that I never would have met otherwise." It is how these friendships are formed through the medium of the Internet, how a sense of "community" is fostered, and how time spent logged on to DragonMud4 [4] affects and informs players' lives away from the keyboard that I propose to study.
In my research, I focus on the following areas:
- the social structure of DragonMud, how and when players interact with
each other, political hierarchies, and the granting of authority,
- the creation and use of "space", how geographic tropes are consciously
used to create a sense of "place" in a community that has no physical
geography, and how players move within that space [5],
- the creation and maintenance of a sense of "community", how
individuals are socialized into it, how the rules by which the community
lives are negotiated, and how the rules for interaction have changed over
time,
- gender roles (and role reversal), the subtle (and sometimes not-so-
subtle) social cues that betray players experimenting with cross-
gendered characters, how DragonMudders create a "safe" space for women, and
how communication is democratized across genders,
- "invisible colleges" and information as a medium of exchange,
- the orality of text, the density of text as a medium, and how
players convey non-verbal communicative cues,
- the use of DragonMud as a Goffmanesque "backstage" (1959), where
players consciously practice attributes that they wish to transfer into
their lives away from the keyboard,
- and the creation and presentation of "self" and the concomitant
issues of anonymity.
The ramshackle international conglomeration of computers, modems, and bailing wire collectively known as "the Internet" is the largely unintentional outcome of several events. In 1962 Paul Baran, of Rand Corporation and under contract from the US Air Force, published "On Distributed Communications Networks", suggesting that a decentralized communication system would be less vulnerable to attack than would one with a central command and control point. In 1969, J.C.R. Licklider, Robert Taylor, and other researchers at Bolt Beranek and Newman, Inc. got funding from ARPA (the Department of Defense's Advanced Research Project Agency), and put together ARPANET to link four sites[6] using disparate computer hardware and operating systems. By 1971, there were almost two dozen sites connected. ARPANET continued to grow, and in 1975 was transferred to the Defense Communication Agency.
While email and mailing lists became common on ARPANET, that information was not available to academics at institutions without ARPA contracts. In 1979, three graduate students at University of North Carolina and Duke University developed Usenet, the "poor man's ARPANET" [7]. Gradually gateways were opened between ARPANET and Usenet allowing the exchange of information.
During the early 1980s BITNET (Because It's Time Network), FidoNet (consisting, then, primarily of IMP PCs and compatibles), and several other small networks were organized, and in the late 1980s, NSFnet was developed to connect the National Science Foundation's supercomputers. Eventually, protocols were written and reformatted to allow all of these networks to communicate with each other, including networks overseas. The resulting conglomeration is what we now call "the Internet" [8] (Gibbs and Smith 1993; Hauben and Hauben 1995; Kantrowitz and Rogers 1994; Raymond 1993).
MUD (short for Multi-User Dungeon, originally) was the name of a game written by Roy Trubshaw and Richard Bartle when they were students at Essex University in the UK. Initially written in MACRO-10, the machine code for the university's DEC-10, Bartle describes it as being "little more than a series of inter-connected locations where you could move and chat" (1990, 7). Rewriting the entire code twice, it eventually developed into a multi-user adventure game available to subscribers to commercial networks in England.
MUDs (which came to stand, generically, for Multi-User Domain) took on a different aspect in the US, much to Bartle's dismay. He stated (using the term MUA (Multi-User Adventure), lest readers confuse the generic code with his particular game) that
at present, the best games are the top-notch UK professional MUAs, but with the huge number of US academics presently engaged in MUA activity, it is only a matter of time before players over there start writing their own versions and marketing them commercially. Unless the UK can maintain the lead that history has given it, these American MUAs will doubtless come to dominate the scene over the coming years (1990, 7).
The first American version of the game code was TinyMUD, written in 1989 by James Aspnes, then a graduate student at Carnegie-Mellon University. Bartle especially objects to the TinyMUD code, stating
TinyMUD is not so much a MUA as a forum for conversation where participants have pinned short pieces of prose on the wall for the benefit of anyone with the inclination to read them. If this kind of MUA gets a strong hold in the USA, it could set the industry back several years (1990, 80).
It is, however, the flexibility of TinyMUD code, and the ease of real-time communications it affords, that attracts most of its users. Text is an information-dense medium; the ability to use the same language forms we use in everyday speech with minimal interference from the underlying code makes this medium a powerful narrative tool. As Howard Rheingold, one of the few people who have actually written about virtual communities, writes,
narrative is the stuff of which MUDworlds are made. Everyone and everything and every place has a story. Every object in a MUD, from your character's identity to the chair your character is sitting in, has a written description that is revealed when you choose to look at the object (1993, 155).
Currently, there are several hundred MU*s being operated on the Internet. MU*s are, simply, text-based real-time multi-person interactions occurring within nodes of computers. Generally, MU*s have some sort of over-arching "theme", ranging from fantasy to futuristic. Some are rather large, with more than 100 people logged in at any time from anywhere in the world; others are restricted to participants at a specific campus. Some are combat-oriented, others are known as being "intellectual" MU*s.
In the same way that there is a recognized difference between "VR" (virtual reality) and "RL" (real life [9]), mudders distinguish between the "character" or "avatar" they role-play and themselves; the human being is the "player" or "operator". This nomenclature is not firmly established, though the concept is universally accepted. The "style", attitudes, physical description, quirks and idiosyncrasies of the character are all generated by the player. While some people play roles, generally "characters" who have been around for a while pretty closely represent their "players" - that is, "players" "play" themselves.
When John Crane (whose tangled initials JO.hn P. C.rane are the basis for the name by which he is known on DragonMud, "Jopsy") was a student at Northern Arizona University (NAU), he wrote an interactive, user-alterable game called "The Playground" in the late 1980s. Written in Fortran77, PL6, and SPITBOL!, it ran on NAU's Honeywell DPS-6. He chose that particular machine, he explains, because it was the most popular mainframe on the campus, giving him the widest possible cross-section of players. Crane states that wrote this implementation because he was frustrated with the lack of elegance in the University BBS code, the maintenance for which he was responsible. Originally designed as a 10x10 grid through which players could move, The Playground grew a then-monstrous 800 objects (Crane 1989, 1; personal communication 1/5/94).
When the Honeywell mainframe was scheduled to be retired, Crane started to rewrite the code in C, intending to transfer it to NAU's UNIX system. When James Aspnes released his TinyMUD code at Carnegie-Mellon, Crane realized that most of what he was trying to do was included in Aspnes's code. Since then, that code has been heavily modified by Crane to increase the ease with which players use it, to increase player privacy, and to add features to make it more enjoyable (personal communication 1/5/94).
DragonMud is the oldest continuously-running implementation of the TinyMUD family [10]. Originally put up in December 1989, it became accessible to non-local players when NAU went on the Internet in March of 1990. While many people initially log on for the recreational aspects of DragonMud, those who continue to log on are integrated into the community. Many MU*s are predicated on playing a fantasy "role"; DragonMud is not. Players are informed when they log in that "In DragonMud, mudders play themselves in 18th century London." or, as The Dragon puts it, "We don't have strange people in a familiar place, we have real people in a strange place." The sense that the people with whom one is interacting are "real" (though, of course, there are issues of anonymity on the net) facilitates the formation of a true "imagined community", where people come together to support and care for each other (Anderson 1983).
Players generally begin to form friendships by asking other players about solutions to quests [11], then gradually become involved in conversations. For many, DragonMud becomes a "third place", a place they can just hang out apart from work (or school) and home (Oldenburg 1991, 14). Discussions cover a wide range; a single recent evening's topics included Chaucer, whether Christianity allows for a "loving" god, how to write binary locks on a room's entrance, a comparison of prices and services offered by commercial Internet providers, a player's "bitter" break-up with his fiancee, what an Eigenvector is, and a rather heated debate about whether or not the federal budget should be balanced. Because people log on to DragonMud from all over the world, it is noteworthy, but not unusual, to find a conversation occurring in Town Square in a language other than English.
DragonMud has a current population of over 5000 players, with a "core" community of between 700 and 900 people [12]. Many are college students, but, compared to other MU*s, an unusually high percentage are post-docs and professional people (O'Brien 1992) [13]. The youngest regular player (who logs in through his professorial father's university account) is 13, the oldest (the author of a programming language) is in his mid-70s. These core players define themselves as a "community", or even as "family". They share a common history, and assume a common future. Within the community there are social and political hierarchies, systems of exchange (albeit non-monetary), fictive kinships, ritual processes, factions and friendships [14]. As Randall put it, "We have everything a RL community has except agriculture... but we grow ideas instead."
When an individual initially logs onto a MU*, the first thing s/he does is
"create" themselves.
** self-creation - "More persons in more parts of the world consider a
wider set
of 'possible' lives than they ever did before" (Appadurai 1991, 197).
** self-creation - "Twentieth-century identities no longer presuppose
continuous
cultures or traditions. Everywhere individuals and groups improvise local
performances from (re)collected pasts, drawing on foreign media, symbols,
and languages" (Clifford 1988, 14).
** socialization - "'Invented tradition' is taken to mean a set of
practices,
normally governed by overtly or tacitly accepted rules and of a ritual or
symbolic nature, which seek to inculcate certain values and norms of
behaviour
by repetition, which automatically implies continuity with the past"
(Hobsbawm
and Ranger 1983, 1)
** justification - "These complex, partly imagined lives must now form the
bedrock of ethnography, at least of the sort of ethnography that wishes to
retain a special voice in a transnational, deterritorialized world"
(Appadurai
1991, 199).
Because DragonMud has no physical geography, how players create and
manipulate
concepts of space and place is particularly fascinating. Rooms are
described in
text that gives a sense of "being" someplace. Players treat public and
private
space differently.
** need PoMo geographers
My preliminary data has been collected through several means: direct interaction within the community on an almost-daily basis, on-line group interviews where I pose a question and then attempt to facilitate discussion, Emailed survey instruments, and face-to-face taped interviews. Additionally, data has been collected at the twice-yearly gatherings of the DragonMud community, originally in Flagstaff AZ, and now held in San Diego CA over Memorial Day and Labor Day weekends.
While I intend to continue using these methods, I also intend to expand my research to discover how participation within the DragonMud community affects and informs players' day-to-day lives. To this end, I plan to spend a year traveling to a variety of cities and living with players, participating and observing their lives away from the keyboard. Players in Boston MA, on Long Island NY, in Washington DC, Raleigh and Chapel Hill NC, Atlanta GA, San Diego CA, the San Francisco Bay area, Seattle WA, and Flagstaff AZ, have already expressed an interest in participating in this portion of my research. Not only will actually living with players facilitate the study of the VR-RL overlap, but a comparison of taped face-to-face interviews and on-line interviews demonstrates that information gathered in face-to-face situations in considerably more elaborated. Typed interactions tend to be terse; players are more verbose when they speak.
I frequently conduct on-line group interviews, which provide a wealth of information. People respond to each other's comments, agreeing and disagreeing, usually (but not always) finding consensus -- but it is very difficult to keep conversations like this on-track. Side issues are raised, parenthetical comments become new subjects of discussion, people leapfrog back to previous topics. Computer-mediated interactions are changing how we communicate, just as the telephone did. This new technology, however, raises some fundamental questions about what we mean by "community" -- and even how we define "friend". MU*s are "living laboratories for studying the first-level impacts of virtual communities" (Rheingold 1993, 146).
Studying a group of geographically distant people that define themselves as being a "community" provides some unique anthropological opportunities. In the "living laboratory" of DragonMud, the very rules by which the community operates are still in flux and being renegotiated. Behavior that was acceptable five years ago is no longer tolerated as male dominance is lessened. Administrators are less capricious in their treatment of "mortal" players as the community grows and Geertzian models "of" and "for" reality come into play (1973). Classic Turnerian ritual tropes (1969) are employed when players are elevated into administrative positions. Mauss's theories of reciprocity become problemitized in a community without corporeal existence. Following Huizinga (1949), the element of "play" within DragonMud acts as a method of socializing new players into the community.
As I pointed out in the keynote speech at a conference [15] on using MU*s as "learning spaces", within the DragonMud community people are meeting and becoming friends with a broad variety of others with divergent and different views -- people whose lives are completely different from their own. People are exposed to new ideas, to new information, to alternate points of view. Sometimes, even if they continue to disagree, people begin to understand each other's viewpoint (Clodius, in press). Yet, as different and diverse as the individuals are, they still define themselves as being a "community". It is how this community forms, represents, and reproduces itself that I propose to study.
Moreover, I am apparently uniquely qualified to accomplish this project [16]. I write as a "native" anthropologist as I have been a "resident" of DragonMud for almost five years, pre-dating my conscious efforts to actually study the community.
Working on the Internet presents some special challenges. What level of anonymity do I provide for people who are already using aliases, but whose real-life identity is common knowledge within the community? What is public discussion and what is privileged information, when everything is being transmitted over public telephone lines?
Footnotes
[1] On-line communities are, however, beginning to enter academic and public consciousness and publications. See, for example, Dibble (Village Voice 1993), Escobar (Current Anthropology 1994), Germain (Time 1993), Hafner (Newsweek 1994), Kelly and Rheingold (Wired 1993), Leslie (The Atlantic 1993), Quittner (Wired 1993), Riner and Clodius (Anthropology & Education Quarterly 1995) and Schwartz (The Washington Post 1994). However, these examples are all about isolated incidents, are written from a "top down" perspective, or are written by non-"residents"; none of them is the product of long-term participant observation.
[2] DragonMud was founded in December, 1989.
[3] Several variations of the original TinyMUD code now exist, including MUSH, MUCK, and MOO. They are collectively known as MU*s, with the asterisk representing the UNIX "wild card" letter(s).
[4] Incidentally, it should be noted that the only "dragon" in DragonMud is The Dragon, player #1. As has been noted by several players over the years, "DragonMud" would be more appropriately named "The Dragon's Mud".
[5] A map of tinyLondon (available to anyone who picks one up in Town Square) looks like this:
~~~~
~~~ ||
(O)########=====###||################(O) TS - Town Square
# ~~~ || # ST - The Carriage Station
# SH===NW ~~~ N2 WA===EA==NEC # BZ - Bazaar
# || ~~~ || || # DO - Docks
# NF SW ~~~ N1 NR # SH - Soho
# || || ~~~ || || # TF - Trafalgar Sqr
(M)=W3===W2===W1===TS===E1===E2===E3====
# || || ~~~ ST||BZ || #
# SF BS ~~~ S1 `-+-. -. SR #
# || || ~~~ || -| `-+-' || # nw n ne
# CS===TF=~~DO===S2----+- `- SEC # \ | /
# ~~~ || alleys # w - * - e
(O)########======####################(O) / | \
~~~ sw s se
~~~~~~~
(note: the ~~ represent the Thames, ## indicate town walls, == and || are streets, other letters indicated space-names: SEC is the SouthEast Courtyard, NF is North Firemarch Road, N2 is North Narthat Street, and so on.
[6] UCLA, UC Santa Barbara, Stanford Research Institute, and the University of Utah
[7] Hauben and Hauben cite a student's comment, "I don't remember when the phrase was coined. But to me it expressed what was going on. We (or at least I) had little idea of what was really going on on the Apranet, but we knew we were excluded. Even if we had been allowed to join, there was no way of coming up with the money. It was commonly accepted at the time that to join the Arpanet took political connections and $100,000. I don't know if that assumption was true, but we were so far from having either connections or $$ that we didn't even try. The 'Poor man's Arpanet' was our way of joining the CS community (Computer Science -ed), and we made a deliberate attempt to extend it to other not-well-endowed members of the community. (1995, 49)"
[8] Parenthetically, if one were to ask an Internet user whose use pre-dates the coining of the phrase, they will explain that the Internet IS the "Information Superhighway" -- and that this "superhighway" comes without speed limits or lane lines, and occasionally, is unpaved. The fact that there is no one governing body has, for years, invoked a spirit of cooperation among users of the Internet, though I hear a growing number of complaints about the "uninitiated" users from commercial networks like America OnLine and CompuServe.
[9] Even "real life" is problematized on DragonMud, as many players would contend that what happens on-line is every bit as "real" as what happens off-line. The generally accepted definition of "RL", then, is merely time spent away from the keyboard.
[10] Of the seven TinyMUDs that Bartle ranked "first rate" in 1990, DragonMud is the only one still up and running (1990, 102).
[11] "Quests" are the puzzles and text-adventures that form a large portion of the DragonMud landscape; solving specific quests earns rank and eventually, permissions to change the database by creating objects and potentially adding one's own quest to the realm.
[12] DragonMud's social system is hierarchical, comprising of "mortals", who control themselves and the objects that they have created, "wizards", who act as "junior administrators" and can see, but not affect, most things regardless of ownership, and "gods" or "senior administrators", who can see and control anything.
[13] I've had to rely on players' self-reported ages, genders, and professions in my preliminary research; coming up with actual statistics of how the population breaks out in these areas won't be possible without face-to-face interviews.
[14] Some players are also quite overt about their use of DragonMud as a Goffmanesque "backstage", using it to practice characteristics they want to carry over into their lives away from the keyboard. (See Riner and Clodius, for further discussion.)
[15] MUDshop II: Learning Spaces, Sept 6-8 1995, San Diego, CA
[16] Amy Bruckman, a PhD candidate at MIT's Media Lab, is writing about grade-school educational applications of a MOO, and I am aware of several graduate students working on MA theses about other aspects. I am not aware of any other anthropology graduate student writing an ethnography of a specific MU*, however.