Engineering patterns for Trust and Safety on social media platforms: A case study of Mastodon and Diaspora

Context:
Trust & Safety (T&S) Engineering is an emerging area of software engineering that mitigates the risks of harmful interactions in online platforms. Numerous studies have explored T&S risks on social media platforms, taxonomizing threats and investigating individual issues. However, there is limited empirical knowledge about engineering efforts to promote T&S.
Methods:
This study examines T&S risks and the engineering patterns to resolve them. We conducted a case study of the two largest decentralized SMPs: Mastodon and Diaspora. These SMPs are open-source, so we analyzed T&S discussions within 60 GitHub issues. We analyzed T&S discussions that took place in their online repository and extracted T&S risks, T&S engineering patterns, and resolution rationales considered by the engineers. We integrate our findings by mapping T&S engineering patterns onto a general model of SMPs, to give SMP engineers a systematic understanding of their T&S risk treatment options.
Results:
T&S issues are a challenge throughout the feature set and lifespan of an SMP. A taxonomy of 12 solution patterns are developed, paving the way for academia and industry to standardize Trust & Safety solutions. We conclude with future directions to study and improve T&S Engineering, spanning software design, decision-making, and validation. We conclude with future directions to study and improve T&S Engineering, spanning software design, decision-making, and validation.