APPROACHES TO MANAGING DEVIANT BEHAVIOR IN VIRTUAL COMMUNITIES Amy Bruckman (organizer and panelist) MIT Media Lab. 20 Ames St., MIT E15-315a, Cambridge, MA 02139. E-mail: asb@media.mit.edu Pavel Curtis (panelist) Xerox PARC. 3333 Coyote Hill Road, Palo Alto, CA 94304. E-mail: pavel@PARC.xerox.com Cliff Figallo (panelist) Former Director of The WELL. 33 Roque Moraes Ct. #7, Mill Valley, CA 94941. E-mail: fig@eff.org Brenda Laurel (moderator) Interval Research. 1801 Page Mill Road, Palo Alto, CA 94304. E-mail: laurel@interval.com ABSTRACT It is an unfortunate fact of life that where there are multi-user computer systems, there will be antisocial behavior. On bulletin board systems (BBSs), there are those who persist in being obscene, harassing, and libelous. In virtual worlds such as MUDs, there are problems of theft, vandalism, and virtual rape. Behavior is "deviant" if it is not in accordance with community standards. How are such standards developed? Should standards be established by system administrators and accepted as a condition of participation, or should they be developed by community members? Once a particular person's behavior is deemed unacceptable, what steps should be taken? Should such steps be taken by individuals, such as "filters" or "kill" files on BBSs, and "gagging" or "ignoring" on MUDs? Or should the administrators take action, banning an individual from the system or censoring their postings? What is the appropriate balance between centralized and decentralized solutions? (Figure 1-- omitted in plain text version). Gags and filters are computational solutions to deviant behavior. Are there appropriate social solutions? How effective are approaches like feedback from peers, community forums, and heart-to-heart chats with sympathetic system administrators? Are different approaches effective with communities of different sizes? What is the appropriate balance between social and technological solutions? KEYWORDS: Community, standards, behavior, social versus technological approaches, virtual communities, MUDs, Bulletin Board Systems (BBSs). [Figure omitted in plain text version.] Figure 1: Approaches to Deviant Behavior: Two Continuums POSITION STATEMENTS BRENDA LAUREL, INTERVAL RESEARCH (MODERATOR) BACKGROUND Brenda Laurel is a researcher and writer whose work focuses on human-computer interaction and cultural aspects of technology. She is a Member of the Research Staff at Interval Research Corporation in Palo Alto, California. She is editor of the book, The Art of Human-Computer Interface Design [Addison-Wesley 1990] and author of Computers as Theatre [Addison-Wesley 1991; 2nd edition 1993]. POSITION In rural Nova Scotia, some say, one small community deals with socially unacceptable behavior in a novel way. They put a live lobster on the offender's back. It can only be removed with the help of others. The technique is said to promote intensive individual learning and a high degree of social conformity. It is likely that virtual communities will be at least as diverse-culturally, demographically, ethically, and politically-as actual communities. What are some potential means for virtual communities to deal with "antisocial" behavior? How effective are they? What are the tradeoffs involved in various "solutions"-for example, how do they affect the character of a community and the rights of individuals? In terms of both problems and solutions, how are virtual and actual communities different, and how are they the same? AMY BRUCKMAN, MIT MEDIA LAB BACKGROUND Amy Bruckman is a doctoral candidate at the Media Lab at MIT, where she founded MediaMOO, a text-based virtual reality environment or "MUD" designed to be a professional community for media researchers. Amy received her master's degree from the Media Lab's Interactive Cinema Group in 1991. For her dissertation, she is creating a MUD for kids called MOOSE Crossing. MOOSE Crossing is designed to provide an authentic context for kids to learn reading, writing, and programming. POSITION In computer-based communities, it is tempting to throw technological solutions at social problems. Someone programmed virtual guns? Delete them. Got an obnoxious user? Cancel their account. I will argue that social solutions are often more effective and also help to reinforce a sense of community. I have had success with a psychoanalytic approach to dealing with problem users. Someone who is causing trouble probably wants attention. A heart to heart chat with a sympathetic system administrator can often solve the problem. Technological interventions are rarely more than a band-aid for social problems. However, social solutions require time, effort, and leadership. Being able to take the time to engage each problem user in a dialogue is a luxury that comes from having a small community size. Larger communities necessarily become bureaucracies; in a real sense, they cease to be communities at all. I will propose a model of clusters of small, affiliated communities and sub-communities as a structure for preventing and managing social problems. PAVEL CURTIS, XEROX PARC BACKGROUND Pavel Curtis has been a member of the research community at the Xerox Palo Alto Research Center since 1983, during which time he has worked on programming environments and on other projects mostly related to the design and implementation of programming languages. His current work centers on the Social Virtual Reality project, investigating the implementation, applications, and implications of systems that allow multiple simultaneous users to communicate and interact in pseudo-physical surroundings. He is the founder and chief administrator of LambdaMOO, one of the most popular recreational social virtual realities on the Internet. POSITION For behavior to be deemed "deviant," it must by definition deviate from some accepted norm. Who defines the norms in any given society and how are those norms communicated to newcomers? In LambdaMOO (an online community on the Internet), there are a number of mechanisms through which the community as a whole can decide upon "the rules" and communicate those decisions to all. I'll discuss the origins and evolution of some of those mechanisms. Hand-in-hand with establishing behavioral norms, societies decide how to cope with members or visitors who violate those norms. I will argue that we should distinguish two broad categories of deviants and craft separate policies for dealing with them. Finally, I will describe some of the coping mechanisms suggested by LambdaMOO users, including some that have been implemented and applied in practice. CLIFF FIGALLO, THE WELL BACKGROUND Cliff Figallo was Managing Director of the Whole Earth Lectronic Link (the WELL) during six of its first seven formative years, and has worked as Online Communications Coordinator for the Electronic Frontier Foundation. He is currently consulting in the field of online communications. Before becoming involved in online community activities, Cliff spent twelve years living and working in an intentional "real life" community called the Farm. POSITION On the WELL, system managers sought to nurture a community where free speech was the norm and where all users felt safe to express themselves. System managers took care not to publicly exercise power in ways that might inhibit open group interaction. By encouraging the formation of core groups of users who shared their desire for minimal social disruption, management not only relieved itself of the need to intervene as the authority in minor cases of disruption, but it also gained the socializing influence of a dispersed citizenry actively supporting community standards of behavior and passing them on to new arrivals. Online system managers are easy targets for challengers of authority. If peer pressure can be relied on to quell minor disruptive incidents, management can be more effective as a court of last resort for more incorrigible violators of social norms. Management can also be creative in its treatment of disruptive but non-malevolent users. Expulsion from the system is, like capital punishment in Real Life, the most extreme option and, in these new media, there may well be technical remedies where social ones are lacking. Copyright of the Association for Computing Machinery, 1994. To be presented at CHI'94 in Boston, MA in April 1994.