Networks beyond the links: Methodological challenges in mapping YouTube communities

This paper develops a methodological framework that integrates structural indicators with the symbolic and cultural dimensions of community formation on YouTube, addressing the platform’s limited relational data and conceptual ambiguities. This study bridges the computational approaches of information sciences with the qualitative traditions of communication studies, social sciences, and digital humanities, proposing a hybrid method to identify, size, and interpret communities on the platform. Using the GNU/Linux content creator ecosystem as a case study, this research identifies 62,082 YouTube channels that have published related content over the past 15 years. This large-scale dataset serves to test various filtering and delimitation strategies that combine quantitative network analysis with qualitative reasoning to define community boundaries. Results reveal structural limitations in YouTube’s API — particularly the scarcity and opacity of relational data — and highlights the methodological implications of relying on subscriptions and featured channels as community indicators. This work concludes that integrating cultural interpretation with computational mapping enables the scalable yet reflexive study of online communities, offering a significant contribution to digital methods and online community studies.