High-tempo collaboration

Exploring sudden and emergent information behaviors

My first research focus explores the role of what I call high-tempo online collaborations [1]. Much of the research about online community processes overlooks the role that disruptions play in structuring social interactions, sustaining user engagement, or disseminating innovations and norms. “Bursts” are characterized by a sudden onset, significant change in intensity, and temporary duration of social behavior and often follow breaking news, collective action, and policy changes. In the aftermath of these events, there are bursts of collective attention as people seek out information to consume to help them make sense of current events. But bursts are also instigators of unusual collaborations to produce information characterized by temporary but high tempo activity, a convergence of a large number of contributors, and the regeneration of previous coordination practices. The mediation of bursts of social behavior like high-tempo online collaborations through information technologies capturing detailed digital traces enables researchers like me to understand the antecedents, processes, and consequences of these disruptions. The coalescence of temporary collaborations following disruptive events on platforms like Wikipedia, Twitter, and Reddit vividly illustrates the origins of social structure and these exogenously forced events provide ideal opportunities for naturalistic causal inference.

Wikipedia’s breaking news collaborations

The primary empirical sites for my research career since graduate school have been Wikipedia’s breaking news collaborations. Wikipedia was founded in January 2001 and has distinguished itself since at least the September 11, 2001 attacks by its contributors’ ability to rapidly synthesize information about breaking news events into new encyclopedia articles [2]. My research has primarily (but not exclusively [3]) focused on collaborations on the English Wikipedia and characterized the “bursts” of revisions and pageviews in the aftermath of disasters [1,4,5], around elections [6,7], following VIP deaths [8], and during social movements [9]. Methods drawn from network science, natural language processing, statistics, and digital trace ethnography were the primary empirical tools used to understand the structure and dynamics of these emergent collaborations. These research findings emphasize the importance of improvising coordination and conflict resolution techniques [4], regenerating prior practices and collaborative structures and adopting differentiated social roles [10,5], and managing flows of pro-social collaboration and re-appraisal consumption [8,9,6]. Wikipedia remains one of my primary settings for studying and theorizing about how disruptions illuminate the dynamics of social structures.

Software patches in online games

Software patches in online gaming communities are unique setting to study the coupling of disruptions and social behavior because of their profound importance to the player community. I won a two-year NSF EAGER grant in 2018 to study high-tempo organizational behavior within a popular genre of online games called multiplayer online battle arenas (MOBAs). This research was augmented by computing awards from Microsoft and Oracle. Early versions of this research were presented at a network science conference and manuscripts from her dissertation are currently under review and preparation for submission. My interest in using social behavior within online games as an empirical setting to study processes like high-tempo collaborations started a graduate student [11,12,13,14,15] and continued through my post-doctoral appointment exploring social dilemmas in team formation [16] and the role of non-verbal communication on performance [17].

Cross-cultural conflict documentation

Wikipedia’s documentation of conflicts is a particularly exciting setting to study the coupling of disruptions and social behavior. Its multiple language editions enables research designs around how the same disruption unfolds across communities with different norms owing to social mechanisms like culture and geography. In collaboration with Emily Porter and Peaks Krafft, we explored the variation in image use on Wikipedia articles about conflicts across languages [3]. Using thematic coding, network analysis, case studies, and machine learning methods we identified distinctive structures and patterns of image use for culturally-proximate languages versus outsider languages. Professor Alexandra Siegel and I are co-advising a Ph.D. student, Laurie Jones to explore how conflicts in the Arab world differ across Wikipedia’s language editions.

New Work: Community spillovers

The consequences of high-tempo collaborations are not limited to an article about a current event but also ripple out to modify the content on other pages, the behavior of participating editors, and activity throughout the project. These are community spillovers where the disruptions of a current event have both “good” and “bad” consequences for the rest of the platform. All these changes are examples of how high-tempo collaborations are embedded within a larger ecology of roles, practices, and values that have been largely ignored by empirical scholarship, including my own. The temporarily engagements with new collaborators, exhilarating pace of changes, immediate validation of contributions, and remixing and diffusion of innovative practices all could be important mechanisms for retaining editors and improving the health of the broader project. But coordination costs, burnout, content complexity, conflict, distraction, preferential attachment, and other dynamics of editors’ self-organization in these collaborations could also compromise the health of the broader project. There has neither been a systematic enumeration of these competing mechanisms nor an accounting of their influence on fundamental metrics of success. Simply put: despite their scale, importance, and ubiquity, very little is known about the consequences of high-tempo collaborations on the success of articles, editors, and projects.

References

  1. Hot Off the Wiki: Structures and Dynamics of Wikipedia’s Coverage of Breaking News Events
    Keegan, Brian C., Gergle, Darren, and Contractor, Noshir
    American Behavioral Scientist 2013
  2. A History of Newswork on Wikipedia
    Keegan, Brian C.
    In Proceedings of the Ninth Annual International Symposium on Wikis and Open Collaboration 2013
  3. Visual Narratives and Collective Memory across Peer-Produced Accounts of Contested Sociopolitical Events
    Porter, Emily, Krafft, P. M., and Keegan, Brian C.
    ACM Transactions on Social Computing Feb 2020
  4. Hot off the Wiki: Dynamics, Practices, and Structures in Wikipedia’s Coverage of the Tōhoku Catastrophes
    Keegan, Brian C., Gergle, Darren, and Contractor, Noshir
    In Proceedings of the 7th International Symposium on Wikis and Open Collaboration Feb 2011
  5. Staying in the Loop: Structure and Dynamics of Wikipedia’s Breaking News Collaborations
    Keegan, Brian C., Gergle, Darren, and Contractor, Noshir
    In Proceedings of the Eighth Annual International Symposium on Wikis and Open Collaboration Feb 2012
  6. The Dynamics of Peer-Produced Political Information During the 2016 U.S. Presidential Campaign
    Keegan, Brian C.
    Proceedings of ACM Human-Computer Interaction Nov 2019
  7. Wiki-worthy: collective judgment of candidate notability
    Margolin, Drew B., Goodman, Sasha,  Keegan, Brian C., Lin, Yu-Ru, and Lazer, David
    Information, Communication & Society Nov 2016
  8. ’Is’ to ’Was’: Coordination and Commemoration in Posthumous Activity on Wikipedia Biographies
    Keegan, Brian C., and Brubaker, Jed R.
    In Proceedings of the 18th ACM Conference on Computer-Supported Cooperative Work & Social Computing Nov 2015
  9. Black Lives Matter in Wikipedia: Collective Memory and Collaboration around Online Social Movements
    Twyman, Marlon,  Keegan, Brian C., and Shaw, Aaron
    In Proceedings of the 2017 ACM Conference on Computer Supported Cooperative Work and Social Computing Nov 2017
  10. Do Editors or Articles Drive Collaboration? Multilevel Statistical Network Analysis of Wikipedia Coauthorship
    Keegan, Brian C., Gergle, Darren, and Contractor, Noshir
    In Proceedings of the ACM 2012 Conference on Computer Supported Cooperative Work Nov 2012
  11. As real as real? Macroeconomic behavior in a large-scale virtual world
    Castronova, Edward, Williams, Dmitri, Shen, Cuihua, Ratan, Rabindra, Xiong, Li, Huang, Yun, and Keegan, Brian C.
    New Media & Society Nov 2009
  12. The Formation of Task-Oriented Groups: Exploring Combat Activities in Online Games
    Huang, Yun, Zhu, Mengxiao, Wang, Jing, Pathak, Nishith, Shen, Cuihua,  Keegan, Brian C., Williams, Dmitri, and Contractor, Noshir
    In 2009 International Conference on Computational Science and Engineering Nov 2009
  13. Mining for Gold Farmers: Automatic Detection of Deviant Players in MMOGs
    Ahmad, Muhammad Aurangzeb,  Keegan, Brian C., Srivastava, Jaideep, Williams, Dmitri, and Contractor, Noshir
    In 2009 International Conference on Computational Science and Engineering Nov 2009
  14. Dark Gold: Statistical Properties of Clandestine Networks in Massively Multiplayer Online Games
    Keegan, Brian C., Ahmed, Muhammad Aurangzeb, Williams, Dmitri, Srivastava, Jaideep, and Contractor, Noshir
    In 2010 IEEE Second International Conference on Social Computing Nov 2010
  15. Sic Transit Gloria Mundi Virtuali? Promise and Peril in the Computational Social Science of Clandestine Organizing
    Keegan, Brian C., Ahmed, Muhammad Aurangzeb, Williams, Dmitri, Srivastava, Jaideep, and Contractor, Noshir
    In Proceedings of the 3rd International Web Science Conference Nov 2011
  16. The Proficiency-Congruency Dilemma: Virtual Team Design and Performance in Multiplayer Online Games
    Kim, Jooyeon,  Keegan, Brian C., Park, Sungjoon, and Oh, Alice
    In Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems Nov 2016
  17. Ping to Win? Non-Verbal Communication and Team Performance in Competitive Online Multiplayer Games
    Leavitt, Alex,  Keegan, Brian C., and Clark, Joshua
    In Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems Nov 2016