Work, Play and World of Warcraft

Bonnie Nardi (UC Irvine Department of Informatics) is an anthropologist interested in expressive uses of the Internet. Her current research involves an ethnographic study of World of Warcraft. Her talk today, “Work, Play and World of Warcraft,” deals with the game World of Warcraft as an environment for play with non-play activities. Bonnie discussed “play as human activity that is freely chosen, exists within its own identifiable boundary, can be repeated, and makes the player feel good in some way.”

WoW is played by 8 million people, with most users in China. To play, you create a character by selecting a race and class, such as “dwarf warrior,” and then take that character through adventures to advance in levels (up to 70). It is based on the fantasy genre popularized by J.R.R. Tolkein. The game encourages group formation and other socialization between players through their characters. WoW is more than an exploration of Azeroth and Outland; it also has many other activities that support those adventures. Nardi has been part of an immersive participant-observation study in and out of WoW since December 2005.

“My daughter says I’m a little bit addicted,” Nardi shared.

The ethnography—which includs analysis of game logs, interviews, guild activity, and screen-capture images—attempts to study what constitutes play in online games. The conclusions thus far include as definition that play:

• exists within its own boundary, separate from the ordinary world;
• repeats customary activities;
• is voluntary;
• makes the player feel good in some way.

Play differs from work given a definite sense of separation between these two activities. In conversations taken from guild website posts, references to employment are met with mock ignorance (“What is this ‘work’ you speak of?”). Players also give very precise times for ending game sessions, illustratiung the sad sense of transition into another activity. The reasons Nardi gives for this sense of play being appealing are Autonomy, Community and Enchantment.

Autonomy is described as voluntary action, something that can be done at one’s own pace with a constant sense of accomplishment. There is value embedded in the experience that makes it difficult to throw away. (One departing guild member, for instance, opted to give three 60-level characters away to others in the guild rather than auction them off on eBay … a sacrifice of about $3500.) The sense of community is strong in WoW, compared to other 3D games. One guild member noted an active desire to not purchase new versions of other applications due to the fact that accumulation of gear and abilities is something that can help other characters. “The multiplayer experience really is the reason I can’t stop playing this game.” Nardi’s measurements for enchantment are more loose, as she hasn’t yet found a good way to characterize it beyond the fantasy nature of the graphics.

Settings for Application of 3D Games
According to Nardi, there is a definitie interest in applying 3D virtual world environments to work, school and military objectives. Whether this can be a useful endeavor is under scrutiny. Two basic conditions for play are likely violated out of the box: participation may not be voluntary, and it is inherently difficult to separate the game from ordinary activity. If games are not about play, can they be as effective?

There have been some past experiences in this area. For decades, schools and military have incorporated games into normal operations with partial success. In a 1991 publication, Ambrose Clegg reviewed hundreds of social studies games, producing no evidence that they were any better than traditional teaching methods. In a 2003 study, Simon Egenfeldt-Nielsen reported much the same. James Gee, Constance Steinkuehler and Kurt Squire from Wisconsin argue that there is educational value in games, but most schools are not set up to invest sufficient time for students to learn how to play them effectively. Nardi also stresses the need for controlled studies to generate some meaningful research for these insights.

Nardi’s partner in collaboration, researcher Vickie O’Day, once described a text-based MOO used in a 6th-grade classroom that targeted disadvantaged children with “extreme needs for attention.” The game was used in class, but also at lunch, after school, and during library sessions. The students liked it, but there were no measurable effects from participation in the world. Pat Seed (UCI) used Civilization and Age of Empires to teach history. Her students also liked it, but the lessons didn’t meet her goals. Seed revised the course to have students design and build a game. They were constructed mostly as card and board games, rather than computer-mediated, but the activity was very successful. Students worked hard, shared their games, and came to understand the selective interpretive use of the facts in history.

The military claims success with simulations for the purpose of training. The Defense Advanced Research Projects Agency (DARPA) is exploring 3D virtual worlds for modeling terrorist activity, insurgency, and “non-kinetic warfare” (i.e. resettling refugees). There are few studies about such games in the work environment. There is interest (Intel, HP and IBM) in platforms for building 3D virtual worlds to support remote work, and text-based MOOs and MUDs have been tried with limited success. Second Life has attracted businesses such as Reuters, Toyota, Sun, IBM, Sony and several libraries. Even the Swedish Embassy has opened up a virtual shop.

It may prove more feasible to implement game play within work and military contexts. Their schedules are more flexible and both have more money to invest into development. The benefits are also potentially easier to measure. Such spaces are attractive because of the ability to visualize content and share both objects and space. In face-to-face communication, bonding is facilitated by tangible events, like touching, eating and drinking, and informal conversation. Those things are either ubiquitous or well simulated to create a richer environment for building social capital. Text-based worlds seem to require too much of participants—writing about space and objects instead of moving or seeing to communicate.

Nardi’s future work will include applications built in Second Life. She also has a small grant to go to spend time in Internet cafes in China to gain a cross-cultural examination of games. “Everything I say now is based on North American experience,” she said.

Some other comments from the Q&A:

• There are different kinds of modes of play. In Nardi’s guild, she can only kill computer-generated monsters. Others have to deal with killing other players. When being griefed, one has to rely on her guild for help.
• When does play become work? There are repetetive activities in the game that feel boring and are labeled work, but Nardi feels there really is a separation between work and play. Most play to get away from work.
• Autonomy is more than just self-awareness and voluntary activity. However, the real power is in creating the character the way you want it to be. People reconfigure characters from time to time. There is a tremendous sense of freedom in doing that, especially since you can’t do it at work or school. Nardi acknowledged that she needs to develop better what is meant by autonomy.
• What about social education, where you learn to share? Although Nardi appears skeptical of educational possibilities for video games, there is lot of opportunity to learn social skills and to improve writing and reading. Unfortunately, we don’t have and educational system that can judge the merits of such student activity. There is a lot of potential, but not so much in the current system.
• We can implement games in school without their being voluntary. Students may be engaging in a game, but it isn’t play in the sense that they have a choice about participation.
• Why are these games interesting to study, beyond the prediction they will become more prevelant? Games are places where people grow attachments. Things are being learned in these environments by people who will be future workers.
• In casual interviews of women who play the game, it seems very few women play melee, where combat is in close quarters. This suggests genders play differently. Nardi cautioned against this somewhat by pointing out that while very few women choose melee, men choose all classes. More research is needed on why this appears to be the case.
• It was also suggested there is a split by age. Alliance is mostly younger, Hoard is mostly older (according to a resource called Blizzard?). Nardi doesn’t have a way to do large surveys, however, so her research is mostly small groups and neglects that kind of assessment.