Download PDF.
The Portland Pattern Repository (PPR) went public in 1995. It was the first instance of a type of collaborative Web site that would come to be known as a Wiki or CoWeb (short for collaborative web). A Wiki site is distinguished from an ordinary Web site by the fact that any visitor to the site can edit its pages using an ordinary web browser. The PPR has since grown to over 25,000 unique pages as of November 2003, collectively authored by over 3500 known authors and an unknown number of anonymous authors. It is widely read by software developers in the patten-based programming and Extreme Programming communities, although recently it is more often consulted as a definitive source of information on Wiki itself and collaborative software in general.
The content of the PPR may be of little interest to those outside the software development community, but the way in which the content was generated is of widespread interest. The original Wiki software has spawned dozens of clones in several different programming languages, and Wikis are used on the Internet and on intranets by individuals, special-interest groups, academic communities, and corporate teams for a variety of purposes from project planning to blogging. It is impossible to come to any sort of understanding about “how Wikis are used,” because the the specific ways in which people collaborate (or fail to collaborate) using Wikis vary from setting to setting. Thus this paper will focus solely on the original PPR and the ways in which it is used.
The PPR enables cooperation among widely dispersed members of the software development community who have the common goal of achieving consensus on certain issues of importance to that community. These issues often involve new approaches or techniques that are not widely accepted in the software development community as a whole. A common impression of consensus on these matters is thus important to users of the PPR as reassurance that these approaches and techniques are valid. Due to the open architecture of the Wiki and the importance placed on openness in “Wiki culture,” control of authors and readers is difficult, so etiquette and decorum become issues of primary importance. Community members also take advantage of the Wiki setting to control who participates in the community without violating the core value of “open access.”
Several concepts defined by Erving Goffman will help to explain the behavior of the PPR community. First is the idea of “team performances.” Goffman is concerned with the ways in which we communicate information about ourselves to others, not only through the literal meaning of the words we say or write, but through the impressions we give to others with our non-verbal behavior. He refers to the former as “expressions given” and the latter as “expressions given off.” Goffman uses metaphors from the theater to organize his study of “expressions given off” and suggests that managing these involves a sort of performance by the communicator. He goes on to note that performances of this type often involve multiple people working together to foster a commonly desired impression. In these performances individuals work together as would the cast of a play. This paper will argue that the users of the PPR are engaged in a team performance of this kind.
Continuing the theatrical metaphor, Goffman distinguishes between “front stage” and “backstage” regions, front stage being the place in which performers carefully regulate their behavior to convey the intended impressions, and back stage being the place where performers behave in a more familiar manner, confident that the audience cannot see them. He argues that rules regarding politeness and decorum serve as standards for regulating front stage behavior in order to ensure a successful performance. This paper will show that the rules of style and etiquette espoused on the PPR are indeed used in this way, although front stage and backstage regions are not as easily defined as they are in physical space.
Finally, Goffman argues that just as in the theater, “settings” and “stage props” such as certain furniture, decoration, layout of rooms, clothing and tools are essential to the success of a performance. The setting serves to set the expectations audience members have about the kind of performance they will see. This paper will suggest that the Wiki setting, though not as richly textured as a physical space, nevertheless is important to PPR users not only for the impressions it creates, but for the role it plays in controlling who become members of the PPR “audience” and “cast.”
Goffman recognizes that the theatrical metaphor has its limitations. In particular, determining who are the performers and who is the audience is not always straightforward. This is particularly problematic for asynchronous computer-mediated communication mediums like Wikis. One might attempt to call “central users” the performers and “peripheral users” the audience. But it may be more useful to view the Wiki as Goffman views the funerals conducted for mental hospital patients with no known kin. Although there is no separate audience there to witness the performance, the performers put on the show anyway to reassure themselves that they are maintaining civilized standards. The PPR can be viewed as a case of performers performing for themselves to reassure themselves that their community is maintaining a coherent consensus.
Most of the participants in the PPR are proponents of software development methodologies which face considerable resistance in the workplace. Many pages in the PPR focus on how to “sell” these methodologies to coworkers and managers. An impression of consensus among practitioners is important for validating these participants’ beliefs that their methodologies are sound. Thus there is a constant insistence that “Wiki pages represent nothing but discussion and consensus.” Though Goffman points to the importance of a “working consensus” in a successful performance, it is important to note that in the PPR consensus is not just a means by which impressions are successfully conveyed but is itself the desired impression.
One feature of a team performance is that any one member can disrupt the performance and thus disillusion the audience. Goffman suggests that this actually contributes to team cohesion: “Each teammate is forced to rely on the good conduct and behavior of his fellows, and they, in turn, are forced to rely on him.” This is particular true in the PPR, where any casual browser of the site can “join the team” and begin deleting or vandalizing pages. Users claim that this vulnerability results in “a high level of mutual trust… between users.” Visitors who add content to the site become stakeholders and have an interest in preventing disruptions of this sort.
Disruptions can be minimized by the use of rules of etiquette. Goffman, quoting Hughes, refers to etiquette as “a body of ritual which grows up informally to preserve… the common front.” Preservation of the common front is evidently of great importance to the PPR community, given the number of pages with titles such as “Wiki Social Norms,” “Welcome to Wiki Please Be Polite,” “Good Wiki Citizen,” and “Good Style.” Leuf and Cunningham go as far as to say that “the totally open Wiki concept is based on the idea that people can be polite and well mannered.” Among the most commonly emphasized norms are the use of real names, proper style, and conventions regarding deletion and editing.
PPR users are strongly encouraged to use their real names and to sign their contributions to the Wiki. By doing so, a user creates a “home page” on the Wiki with back-links to all pages to which he or she has contributed. This allows other users to form impressions of that user based on his or her past activity. Most PPR users recognize this function of their home pages and add supplementary information such as pictures, short resumes or biographies, and links to blogs or other home pages. An individual’s home page is his “appearance” on the Wiki and thus represents one of the few opportunities for individual impression management. This is an example of how one team member may “perform” for his fellow teammates.
Writing style is also regulated by social norms: “There is… a certain pressure to conform to existing styles.” There are two predominant styles of writing on the PPR: thread style and document style. Thread style is distinguished by long pages of often heated discussion. Contributions are written in the first person and are always signed. Threaded pages are an example of what Goffman calls “an interaction… purposely set up… for voicing differences in opinion.” He notes that such interactions still have carefully agreed upon rules of etiquette regarding how the differences will be voiced, and the PPR is no exception. “How to Write and Edit Thread Mode” lists several rules of order for threaded discussions on the Wiki, including “use separate posts for separate subjects,” “quote and respond,” and “use horizontal lines to delimit threads.”
Because threaded discussions often highlight a lack of consensus, they are not well-liked by most PPR members. Threaded pages are often subject to “refactoring,” editing aimed at pulling consensus from a messy discussion. Converting thread-style pages to document-style pages is considered the ultimate refactoring. One PPR page claims that “when it’s done well and in context it provides one of the highest quality forms of information found on Wiki.” Document-style pages are written in the third person and are unsigned, in keeping with the desired impression that these pages represent “the voice of the community.”
According to Goffman, rules of etiquette are intended to regulate “front stage” behavior. Thus one might logically ask, “Where is the backstage?” This is a difficult question. Despite Ward Cunningham’s claim that “a Wiki is like an open cocktail party,” unlike real cocktail parties Wikis do not have clear clusters of interaction in which we can discern “front stage” or “backstage” behavior. The best answer might be that “back stage” is off the Wiki. This is supported by the fact that Wiki users are encouraged to move discussions in which consensus seems very unlikely to other mediums such as email, where they won’t disturb the illusion of consensus-building.
Although rules of etiquette are the primary tools PPR members use to manage impressions of the community, setting plays a role as well. Many users believe that the appearance and structure of the Wiki itself enforce good team behavior. As one user puts it, “Wiki is a pain to use and it’s ugly. This encourages you to simplify your thoughts and focus on communicating them clearly.” Moreover, although there are no formal access controls in place that might allow team members to selectively allow new members to join, the Wiki software helps to ensure that only the desired type of people will end up participating: “It’s an intelligence test of sorts to be able to edit a Wiki page… [Wiki users] are by nature a pedantic, ornery, and unreasonable bunch.” This echoes Goffman’s assertion that “only individuals of a certain kind are likely to be found in a given social setting.”
Goffman’s notion of team performance and the quasi-theatrical devices which are used to sustain it are very useful for analyzing the behavior of the PPR community. Although Goffman was primarily concerned with synchronous, face-to-face, embodied communication, his ideas translate rather well to the asynchronous, disembodied space of the Wiki. Much of the seemingly obsessive behavior of the Wiki users makes a lot more sense when viewed as an attempt to project an impression of consensus among members of a totally open and relatively diverse community. But it would be a mistake to take this analysis to mean that the PPR community is engaged in self-delusion. Actual consensus on the issues at stake is truly important to the participants in this community—otherwise they wouldn’t spend so much time there. But in order to keep this real struggle toward consensus going, they must act as if consensus has already been achieved. In the words of Goffman, they must become “practiced in the ways of the stage.”
