{Raman19} Aravindh Raman, Sagar Joglekar, Emiliano De Cristofaro, Nishanth Sastry, Gareth Tyson. Challenges in the Decentralised Web: The Mastodon Case. Proc. of IMC, 2019. We found that Mastodon’s design decision of giving everyone the ability to establish an independent instance of their own has led to an active ecosystem, with instances covering a wide variety of topics. However, a common theme in our work has been the discovery of apparent forms of centralisation within Mastodon. For example, 10% of instances host almost half of the users, and certain categories exhibit remarkable reliance on a small set of instances. | The Largest Connected Component (LCC) in the social follower graph reduces from 92% of all users to 46% by outages in five ASes.
{Polinski24} Michael Polinski, Richard Jo, Kevin McAfee, and Fabián E. Bustamante. The Centralization of a Decentralized Video Platform - A First Characterization Of PeerTube. SIGCOMM Comput. Commun. Rev., 2024, 54(4):25–35. Our findings reveal concerning trends toward centralization that echo patterns observed in other contexts, exacerbated by the limited degree of content replication. PeerTube instances are mostly located in North America and Western Europe, with about 70% hosted in Germany, the USA, and France, and over 50% hosted on the top 7 ***ASes.
{Balduf24} Leonhard Balduf, Saidu Sokoto, Onur Ascigil, Gareth Tyson, Björn Scheuermann, Maciej Korczyński, Ignacio Castro, and Michał Król. Looking AT the Blue Skies of Bluesky. Proc. of ACM IMC, 2024. first large-scale analysis of Bluesky, a prominent decentralized microblogging platform [Project|Tool]
{Kleppmann24} Martin Kleppmann, Paul Frazee, Jake Gold, Jay Graber, Daniel Holmgren, Devin Ivy, Jeromy Johnson, Bryan Newbold, and Jaz Volpert. Bluesky and the AT Protocol: Usable Decentralized Social Media. Proc. of ACM CoNEXT DIN Workshop, 2024.
Social Structure & Dynamics
{Zignani18} Matteo Zignani, Sabrina Gaito, Gian Paolo Rossi. Follow the “Mastodon”: Structure and Evolution of a Decentralized Online Social Network. Proc. of ICWSM, 2018. In this paper we present a dataset containing both the networkof the “follow” relationships and its growth in terms of newconnections and users, all which we obtained by mining the decentralized online social network named Mastodon. The dataset is combined with usage statistics and meta-data (geo-graphical location and allowed topics) about the servers comprising the platform's architecture.
{Zignani19} Matteo Zignani, Christian Quadri, Sabrina Gaito, Hocine Cherifi, and Gian Paolo Rossi. The Footprints of a “Mastodon”: How a Decentralized Architecture Influences Online Social Relationships. Proc. of IEEE INFOCOM Workshops, 2019.
{Cava21} Lucio La Cava, Sergio Greco and Andrea Tagarelli. Understanding the growth of the Fediverse through the lens of Mastodon. Applied Network Science, 2021, 6:64. Our first contribution is the building of an up-to-date and highly representative dataset of Mastodon. Upon this new data, we have defined a network model over Mastodon instances and exploited it to investigate three major aspects: the structural features of the Mastodon network of instances from a macroscopic as well as a mesoscopic perspective, to unveil the distinguishing traits of the underlying federative mechanism; the backbone of the network, to discover the essential interrelations between the instances; and the growth of Mastodon, to understand how the shape of the instance network has evolved during the last few years, also when broading the scope to account for instances belonging to other platforms.
{Cava22} Lucio La Cava, Sergio Greco, Andrea Tagarelli. Information consumption and boundary spanning in Decentralized Online Social Networks: The case of Mastodon users. Online Social Networks and Media, 2022, 30:100220. From a mesoscopic perspective, based on Louvain, Leiden, and Infomap community detection methods, the user networks exhibit a moderately high modularity (around 0.7) and a high number of communities; this trait, which indicate the existence of small densely connected groups of users tailored to specific shared interests, appears to be consistent with the spontaneous connectivity trend in Mastodon.
{Sabo24} Eduard Sabo, Tim Gesthuizen, Kelvin J. A. Bouma, Dimka Karastoyanova, Mirela Riveni. An analysis of mastodon adoption dynamics based on instance types. Social Network Analysis and Mining, 2024, 14:184. we conduct a time-based dynamics analysis of various Mastodon instances, from popular ones to country-specific servers | Upon evaluating the communities within multiple mastodon instances, we found that the accounts from a single instance interacted with accounts on more than 18.000 unique instances.
{Failla24} Andrea Failla, Giulio Rossetti. “I’m in the Bluesky Tonight”: Insights from a year worth of social data. PLoS ONE, 2024, 19(11): e0310330. present a large, high-coverage dataset of social interactions and user-generated content from Bluesky Social to address this pressing issue
{Jeong24} Ujun Jeong, Bohan Jiang, Zhen Tan, H. Russell Bernard and Huan Liu. Descriptor: A Temporal Multi-network Dataset of Social Interactions in Bluesky Social (BlueTempNet). IEEE Data Descriptions, 2024, 1:71-79.
{Balduf25} Leonhard Balduf, Saidu Sokoto, Onur Ascigil, Gareth Tyson, Ignacio Castro, Andrea Baronchelli, George Pavlou, Bjorn Scheuermann and Michal Krol. Bootstrapping Social Networks: Lessons from Bluesky Starter Packs. Proc. of AAAI ICWSM, 2025. we asses whether starter packs have been indeed helpful in supporting Bluesky growth
{Brauweiler25} Michael Brauweiler, Meher Chaitanya Pindiprolu, Jürgen Pfeffer, and Ulrik Brandes. Decentralized Discourse: Interaction Dynamics on Mastodon. Proc. of ACM WebSci, 2025. explore the communication dynamics between Mastodon instances and the distribution of user communities, emphasizing their effects on cross-server communication – a critical scalability issue as the Fediverse continues to expand
UGCs & Applications
{Cerisara18} Christophe Cerisara, Somayeh Jafaritazehjani, Adedayo Oluokun, Hoa T. Le. Multi-task dialog act and sentiment recognition on Mastodon. Proc. of COLING, 2018. [Corpus and Code] We manually annotate both dialogues and sentiments on this corpus, and train a multi-task hierarchical recurrent network on joint sentiment and dialog act recognition.
{Radivojevic24} Kristina Radivojevic, Nicholas Clark, Paul Brenner. LLMs Among Us: Generative AI Participating in Digital Discourse. Proc. of AAAI Spring Symposium Series, 2024. We developed the “LLMs Among Us” experimentalframework on top of the Mastodon social media platform forbot and human participants to communicate without knowingthe ratio or nature of bot and human participants
{Gassmann24} Luke Gassmann, Ryan McConville, Matthew Edwards. Leading the Mastodon Herd: Analysing the Traits of Influential Leaders on a Decentralised Social Media Platform. Proc. of IEEE BigData, 2024. Our analysis finds a strong correlation between influence and negative traits at every network resolution, with positive and neutral traits in some cases being negatively correlated with influence
{Hironaka24} Shiori Hironaka, Mitsuo Yoshida, Kazuyuki Shudo. Comparing User Activity on X and Mastodon. Proc. of IEEE BigData, 2024. reports on the differences in user activity between Twitter and Mastodon, a prominent example of decentralized social media
{Liu25} Yuhan Liu, Emmy Song, Owen Xingjian Zhang, Jewel Merriman, Lei Zhang, Andrés Monroy-Hernández. Understanding Decentralized Social Feed Curation on Mastodon. Proc. of ACM CSCW, 2025. presents findings from a two-part interview study with 21 Mastodon users, exploring how they perceive, interact with, and manage their current feeds, and how we can better empower users to personalize their feeds on Mastodon
{Guan25} Maggie Yongqi Guan, Yaman Yu, and Kanye Ye Wang. 2025. Using Affordance to Understand Usability of Web3 Social Media. Proc. of ACM CHI, 2025,
Migration from Twitter / between instances / Cross-Site Linking
{Kupferschmidt22} Kai Kupferschmidt. As Musk reshapes Twitter, academics ponder taking flight. Science, 2022, 378(6620):583-584.
{He23} Jiahui He, Haris Bin Zia, Ignacio Castro, Aravindh Raman, Nishanth Sastry and Gareth Tyson. Flocking to Mastodon: Tracking the Great Twitter Migration. Proc. of ACM IMC, 2023. content of tweets as well as the number of URLs, the number of likes, and the length of tweets are effective metrics for the prediction of user migration
{Cava23} Lucio La Cava, Luca Maria Aiello & Andrea Tagarelli. Drivers of social influence in the Twitter migration to Mastodon. Scientific Reports, 2023, volume 13, Article number: 21626. analyzed the social network and the public conversations of about 75,000 migrated users and observed that the temporal trace of their migrations is compatible with a phenomenon of social influence, as described by a compartmental epidemic model of information diffusion
{Jeong24a} Ujun Jeong, Paras Sheth, Anique Tahir, Faisal Alatawi, H. Russell Bernard, Huan Liu. Exploring Platform Migration Patterns between Twitter and Mastodon: A User Behavior Study. Proc. of AAAI ICWSM, 2024. elaborate on how we investigate these questions by collecting data over 10,000 users who migrated from Twitter to Mastodon within the first ten weeks following the ownership change of Twitter
{Jeong24b} Ujun Jeong, Ayushi Nirmal, Kritshekhar Jha, Susan Xu Tang, H. Russell Bernard, Huan Liu. User Migration across Multiple Social Media Platforms. Proc. of SIAM International Conference on Data Mining (SDM), 2024. We curated a dataset of 14,000+ users migrating from Twitter to Bluesky, Threads, and Mastodon, following those platforms’ terms of service. To our knowledge, this is the first study to exam- ine the differences among multiple migrant groups (based on their chosen platforms) and contrast them with non-migrants [Code]
{Zia24a} Haris Bin Zia, Jiahui He, Ignacio Castro and Gareth Tyson. Fediverse Migrations: A Study of User Account Portability on the Mastodon Social Network. Proc. of ACM IMC, 2024. we have explored user switching between Mastodon instances and made several key observations
{Radivojevic25} Kristina Radivojevic, D.J. Adams, Griffin Laszlo, Felixander Kery and Tim Weninger. User migration in the Twitter diaspora. EPJ Data Science, 2025, 14:36. Surprisingly, our findings indicate that users with larger followings on X/Twitter are more likely to migrate
{Ng25} Yee Man Margaret Ng and Rik Ray. The journalists’ exodus: Navigating the transition from Twitter to Mastodon and other alternative platforms. To appear: New Media & Society. the practical reliance on Twitter’s functionalities, audience base, and professional obligations made total abandonment challenging
{Jeong25} Ujun Jeong, Alimohammad Beigi, Anique Tahir, Susan Xu Tang, H. Russell Bernard, Huan Liu. Fediverse Sharing: Cross-Platform Interaction Dynamics between Threads and Mastodon Users. Proc. of ASONAM, 2025. we introduce FediverseSharing, the first dataset capturing interactions between 20,000+ Threads users and 20,000+ Mastodon users over a ten-month period. This dataset serves as a foundation for studying cross-platform interactions and the impact of federation as two separate platforms integrate.
Moderation
{Hassan21} Anaobi Ishaku Hassan, Aravindh Raman, Ignacio Castro, Haris Bin Zia, Emiliano De Cristofaro, Nishanth Sastry, and Gareth Tyson. Exploring Content Moderation in the Decentralised Web: The Pleroma Case. Proc. of CoNEXT, 2021.
{Zia22} Haris Bin Zia, Aravindh Raman, Ignacio Castro, Ishaku Hassan Anaobi, Emiliano De Cristofaro, Nishanth Sastry, and Gareth Tyson. Toxicity in the Decentralized Web and the Potential for Model Sharing. Proceedings of the ACM on Measurement and Analysis of Computing Systems, 2022, 6(2):Article 35. We have characterised the spread of toxicity on the platform, confirming that the federation process allows toxic content to spread between instances. We have further explored the challenges of moderating this process by building per-instance models | federated toots constitute the most significant chunk of toxic content on 26/30 of the instances. | 60% of toxic toots get more than one reblog, compared to only 16% of non-toxic toots. This trend indicates that interest and uptake in toxic material are consistently greater
{Anaobi23} Ishaku Hassan Anaobi, Aravindh Raman, Ignacio Castro, Haris Bin Zia, Damilola Ibosiola, Gareth Tyson. Will Admins Cope? Decentralized Moderation in the Fediverse. Proc. of WWW, 2023. We observe a diversity of administrator strategies, with evidence that administrators on larger instances struggle to find sufficient resources. We then propose a tool, WatchGen, to semi-automate the process.
{Agarwal24} Vibhor Agarwal, Aravindh Raman, Nishanth Sastry, Ahmed M. Abdelmoniem, Gareth Tyson, Ignacio Castro Decentralised Moderation for Interoperable Social Networks: A Conversation-based Approach for Pleroma and the Fediverse. Proc. of AAAI ICWSM, 2024. Thus, each server only has a partial view of an entire conversation because conversations are often federated across servers in a non-synchronized fashion. To address this, we propose a decentralised conversation-aware content moderation approach suitable for the fediverse.
{Bono24} Carlo Bono, Lucio La Cava, Luca Luceri, and Francesco. An Exploration of Decentralized Moderation on Mastodon. Proc. of ACM WebSci, 2024.
{Zhang24} Zhilin Zhang, Jun Zhao, Ge Wang, Samantha-Kaye Johnston, George Chalhoub, Tala Ross, Diyi Liu, Claudine Tinsman, Rui Zhao, Max Van Kleek, Nigel Shadbolt. Trouble in Paradise? Understanding Mastodon Admin’s Motivations, Experiences, and Challenges Running Decentralised Social Media. Proceedings of the ACM on Human-Computer Interaction, 2024, 8(CSCW2):Article No. 520. we conducted semi-structured interviews with 16 Mastodon instance administrators, including those who host instances to support marginalised and stigmatised communities, to understand their motivations and lived experiences of running decentralised social media
{Colglazier24} Carl Colglazier, Nathan TeBlunthuis, Aaron Shaw. The Effects of Group Sanctions on Participation and Toxicity: Quasi-experimental Evidence from the Fediverse. Proc. of AAAI ICWSM, 2024. We investigate the effects of defederation in the context of the Fediverse, a set of decentralized, interconnected social networks with independent governance. Mastodon and Pleroma, the most popular software powering the Fediverse, allow administrators on one server to defederate from another | Accounts on blocked servers reduce their activity, but not accounts on blocking servers. By contrast, we find that defederation has no effects on post toxicity on either the blocked or blocking servers.
{Melder25} Erika Melder, Ada Lerner, and Michael Ann DeVito. “A Blocklist is a Boundary”: Tensions between Community Protection and Mutual Aid on Federated Social Networks. Proc. ACM Hum.-Comput. Interact., 2025, 9(2):Article CSCW021. develop a series of community-sourced suggestions for Fediverse platform designers, blocklist curators, and community staff to help address these conflicts and protect marginalized users
{Cava25} Lucio La Cava , Andrea Tagarelli. Safeguarding Decentralized Social Media: LLM Agents for Automating Community Rule Compliance. Online Social Networks and Media, 2025, 48:100319. By analyzing over 50,000 posts from hundreds of Mastodon servers, we find that AI-agents effectively detect non-compliant content, grasp linguistic subtleties, and adapt to diverse community contexts. Most agents also show high inter-rater reliability and consistency in score justification and suggestions for compliance
Policy, Governancy & Privacy
{Zulli20} Diana Zulli, Miao Liu, Robert Gehl. Rethinking the “social” in “social media”: Insights into topology, abstraction, and scale on the Mastodon social network. New Media & Society, 2020, 22(7):1188–1205.
{Gehl23} Robert W. Gehl & Diana Zulli. The digital covenant: non-centralized platform governance on the mastodon social network, Information, Communication & Society, 2023, 26(16):3275-3291.
{Tosch24} Emma Tosch, Luis Garcia, Cynthia Li, Chris Martens. Privacy Policies on the Fediverse: A Case Study of Mastodon Instances. Proceedings on Privacy Enhancing Technologies, 2024, 4:700-733.
Crawler
{Zia24b} Haris Bin Zia, Ignacio Castro, and Gareth Tyson. Mastodoner: A Command-line Tool and Python Library for Public Data Collection from Mastodon. Proc. of ACM CIKM, 2024. [Code]
{Min25} Shaojie Min, Shaobin Wang, Yaxiao Luo, Min Gao, Qingyuan Gong, Yu Xiao, Yang Chen. FediLive: A Framework for Collecting and Preprocessing Snapshots of Decentralized Online Social Networks. Proc. of the Web Conference (WWW’25), Resource Track, Sydney, Australia, Apr.-May 2025. [Code]
ArXiv
{Zia25} Haris Bin Zia, Aravindh Raman, Ignacio Castro, Gareth Tyson. Collaborative Content Moderation in the Fediverse. [PDF]
{Jeong25} Ujun Jeong, Lynnette Hui Xian Ng, Kathleen M. Carley, Huan Liu. Navigating Decentralized Online Social Networks: An Overview of Technical and Societal Challenges in Architectural Choices. [PDF]