Bot-based collective blocklists in Twitter: the counterpublic moderation of harassment in a networked public space

Abstract

This article introduces and discusses bot-based collective blocklists (or blockbots) in Twitter, which have been developed by volunteers to combat harassment in the social networking site. Blockbots support the curation of a shared blocklist of accounts, where subscribers to a blockbot will not receive any notifications or messages from those on the blocklist. Blockbots support counterpublic communities, helping people moderate their own experiences of a site. This article provides an introduction and overview of blockbots and the issues that they raise about networked publics and platform governance, extending an intersecting literature on online harassment, platform governance, and the politics of algorithms. Such projects involve a far more reflective, intentional, transparent, collaborative, and decentralized way of using algorithmic systems to respond to issues like harassment. I argue that blockbots are not just technical solutions but social ones as well, a notable exception to common technologically determinist solutions that often push responsibility for issues like harassment to the individual user. Beyond the case of Twitter, blockbots call our attention to collective, bottom-up modes of computationally assisted moderation that can be deployed by counterpublic groups who want to participate in networked publics where hegemonic and exclusionary practices are increasingly prevalent.

The view from the other side: The border between controversial speech and harassment on Kotaku in Action

“…A key approach to managing the problem of online harassment is by developing moderation and blocking mechanisms (Crawford and Gillespie, 2016;Lampe and Resnick, 2004;Geiger, 2016). Our findings add nuance to our understanding of the challenges of this undertaking.…”

Trajectories of Blocked Community Members: Redemption, Recidivism and Departure

“…Online moderation largely addresses the problem of antisocial behavior, which occurs in the form of harassment [50], cyberbullying [47], and general aggression [23]. Approaches to moderating such content include decentralized, community-driven methods [14], as well as top-down methods relying on designated community managers or moderators [33]. Prior research in this area ranges from understanding the actors involved in antisocial behavior [6,28,36,42,51] to analyzing its effects [39] to tools for identifying such behavior [38,53], and even forecasting future instances [31,55].…”

When Users Control the Algorithms

“…There are further examples of this in the scholarly literature. Stuart Geiger details the use of the Twitter API to build automated 'blocklists' -when one member blocks an account it is automatically blocked for all other members of the group [20]. This mass-blocking feature is one that Twitter has, to date, declined to implement.…”