Content area
The Internet has been instrumental in connecting groups of people with similar interests and viewpoints, enabling Web communities to have a voice through platforms designed for social interaction and engagement. While the Web allows us to create communities that are focused on, e.g., specific characters in a given TV show or video game, it has also enabled those with less socially acceptable interests to bridge the geographic distance that would have kept them from interacting in the past. This downside of communicating with likeminded people, regardless of their geographic location, has made the real-world impact of online harassment and extremism an emerging threat to modern society. The necessity of understanding the formation, evolution, and expression of the communities that are part of this puzzle is growing each day, since there are practically no boundaries for creating new communities.
In this dissertation, I aim to broaden our understanding of niche Web community behavior. First, I examine how niche Web communities coordinate real-world disruption through the phenomenon of zoombombing. By analyzing social media posts that share links to online meetings, I identify patterns in how these links are distributed and discussed on platforms like 4chan and Twitter. My analysis reveals that many of these posts target educational institutions and often include conspiratorial themes. This work demonstrates how behaviors originating in overlooked online spaces can lead to direct social harm beyond the Web. My second research project explores the diversity in online women’s ideological spaces using a multi-dimensional approach. I perform a large-scale, data-driven analysis of over 6M Reddit comments and submissions from 14 subreddits. I elicit a diverse taxonomy of online women’s ideological spaces, ranging from counterparts to the so-called Manosphere to Gender-Critical Feminism. I then perform content analysis, finding meaningful differences across topics and communities. Finally, I shed light on two platforms, ovarit.com and thepinkpill.co, where two toxic communities of online women’s ideological spaces (Gender-Critical Feminism and Femcels) migrated after their ban on Reddit.
In my third research project, I perform a first of its kind large-scale measurement study exploring left-wing extremism. I focus on “tankies,” a left-wing community that first arose in the 1950s in support of hardline actions of the USSR and has evolved to support what they call “Actually Existing Socialist” countries, e.g., CCP-run China, the USSR, and North Korea. I collect and analyze 1.3M posts from 53K authors from tankie subreddits, and explore the position of tankies within the broader far-left community on Reddit. Among other things, I find that tankies are clearly on the periphery of the larger far-left community. When examining the contents of posts, I find misalignments and conceptual homomorphisms that confirm the description of tankies in the theoretical work. I also discover that tankies focus more on state-level political events rather than social issues, and that their toxicity increases following deplatforming. Finally, I studied an underexplored domain. Motivated by the growing impact of podcasts on political discourse, as seen with figures like Joe Rogan and Andrew Tate, I explore the political biases and content strategies used in video podcasts. I conduct an analysis of over 13K podcast videos from both YouTube and Rumble, focusing on their political content and the dynamics of their audiences. While YouTube hosts a broad range of content, Rumble has increasingly become a platform of choice for creators who position themselves against mainstream media and moderation norms. This has led to the emergence of a politically charged and ideologically skewed podcast ecosystem on Rumble, making it a niche Web community in its own right. Using advanced speech-to-text transcription, topic modeling, and contrastive learning techniques, I explore three critical aspects: 1) the presence of political bias in podcast channels, 2) the nature of content that correlates with podcast views, and 3) the usage of visual elements in these podcasts. My findings reveal a distinct right-wing orientation in Rumble’s podcasts, contrasting with YouTube’s more diverse and apolitical content.
Together, these studies offer a data-driven, multi-perspective examination of niche Web community behavior. This dissertation highlights how fringe Web communities express, reinforce, and in some cases act on extreme beliefs. My findings underscore the importance of studying underexplored areas of the Web to better understand how online subcultures shape discourse, challenge content moderation, and generate social impact beyond the digital environment.