Michael Gilbert

Editing beyond articles: diversity & dynamics of teamwork in open collaborations

We report a study of Wikipedia in which we use a mixed-methods approach to understand how participation in specialized workgroups called WikiProjects has changed over the life of the encyclopedia. While previous work has analyzed the work of WikiProjects in supporting the development of articles within particular subject domains, the collaborative role of WikiProjects that do not fit this conventional mold has not been empirically examined. We combine content analysis, interviews and analysis of edit logs to identify and characterize these alternative WikiProjects and the work they do. Our findings suggest that WikiProject participation reflects community concerns and shifts in the communitys conception of valued work over the past six years. We discuss implications for other open collaborations that need flexible, adaptable coordination mechanisms to support a range of content creation, curation and community maintenance tasks.

Project talk: coordination work and group membership in WikiProjects

WikiProjects have contributed to Wikipedias success in important ways, yet the range of work that WikiProjects perform and the way they coordinate that work remains largely unexplored. In this study, we perform a content analysis of 788 work-related discussions from the talk pages of 138 WikiProjects in order to understand the role WikiProjects play in collaborative work on Wikipedia. We find that the editors use WikiProjects to coordinate a wide variety of work activities beyond content production and that non-members play an active role in that work. Our research suggests that WikiProject collaboration is less structured and more open than that of many virtual teams and that WikiProjects may function more like FLOSS projects than traditional groups.

A content analysis of wikiproject discussions: toward a typology of coordination language used by virtual teams

Understanding the role of explicit coordination in virtual teams allows for a more meaningful understanding of how people work together online. We describe a new content analysis for classifying discussions within Wikipedia WikiProject - voluntary, self-directed teams of editors - present preliminary findings, and discuss potential applications and future research directions.

Indicoder: an extensible system for online annotation

Online annotations provide an effective way for distributed individuals to better understand and categorize online content, both from the perspective of distilling information presented into more easily interpretable forms and by supporting content analysis to tag individual statements with their intended meaning. This poster presents Indicoder, an application to support in-place content analysis, allowing users to both annotate online corpora and providing a means of tracking those annotations over time so that living documents such as Wikipedia articles and online sources can be analyzed in their authentic contexts.

Context from the human perspective: dispositions and practice

As the notion of ubiquitous computing moves from abstract ideal to collective reality, a number of opportunities emerge that have the potential to dramatically change the devices that populate our daily lives and the patterns of interaction that those devices engender. To ensure that ubiquitous technologies enable useful interactions, we need to incorporate a more expressive notion of context, including how it is enacted, how it is identified, and how it is understood and utilized. This paper presents such an approach, highlighting the ways in which context can be defined in practice through the identification of the human dispositions that shape our everyday activities. These dispositions provide a shared vocabulary for technologies and its users to more effectively mediate meaningful interactions.

Towards a class-based model of information organization in Wikipedia

As complexity increases in commons-based peer production communities, the means of organizing and facilitating collective action must also mature to ensure the ongoing health and active maintenance of those communities [1]. This study examines the types of structured data that exist in Wikipedia, introduces an argument for an extension to the types of structured and semi-structured data within Wikipedia supported by that descriptive analysis; and presents an implementation of that extension that supports instantiations of semi-structured content that facilitate both human and tool-mediated interactions with Wikipedia data. This extension offers a novel means of structuring data to support the ongoing health and maintenance of online communities like the community of editors that maintain and develop Wikipedia.