Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Beyond Community Notes: A Framework for Understanding and Building Crowdsourced Context Systems for Social Media

Travis Lloyd, Tung Nguyen, Karen Levy, Mor Naaman

TL;DR

The paper defines Crowdsourced Context Systems (CCS) as platform-native, crowdsourced context labels that accompany posts to aid interpretation and potentially complement traditional fact-checking. It combines a systematic literature review of Twitter-style CCS research with an empirical analysis of multiple platform implementations to produce a two-part framework: a generalized CCS model and a six-aspect design space (participation, inputs, curation, presentation, platform treatment, transparency). The authors synthesize findings on system impact, contributor behavior, design variations, and user perceptions, and articulate normative guidelines centered on user informedness, power distribution, and fairness. They argue CCS are promising but not a wholesale replacement for content moderation, and advocate for human-centered research to optimize their design, transparency, and integration within broader moderation strategies.

Abstract

Social media platforms are increasingly developing features that display crowdsourced context alongside posts, modeled after X's Community Notes. These systems, which we term Crowdsourced Context Systems (CCS), have the potential to reshape our information ecosystem as major platforms embrace them as alternatives to top-down fact-checking. To deeply understand the features and implications of such systems, we perform a systematic literature review of existing CCS research and analyze several real-world CCS implementations. Based on our analysis, we develop a framework with two distinct components. First, we present a theoretical model to help conceptualize and define CCS. Second, we identify a design space encompassing six key aspects of CCS: participation, inputs, curation, presentation, platform treatment, and transparency. We discuss key normative implications of different CCS design and implementation choices. Our paper integrates these theoretical, design, and ethical perspectives to establish a foundation for future human-centered research on Crowdsourced Context Systems.

Beyond Community Notes: A Framework for Understanding and Building Crowdsourced Context Systems for Social Media

TL;DR

The paper defines Crowdsourced Context Systems (CCS) as platform-native, crowdsourced context labels that accompany posts to aid interpretation and potentially complement traditional fact-checking. It combines a systematic literature review of Twitter-style CCS research with an empirical analysis of multiple platform implementations to produce a two-part framework: a generalized CCS model and a six-aspect design space (participation, inputs, curation, presentation, platform treatment, transparency). The authors synthesize findings on system impact, contributor behavior, design variations, and user perceptions, and articulate normative guidelines centered on user informedness, power distribution, and fairness. They argue CCS are promising but not a wholesale replacement for content moderation, and advocate for human-centered research to optimize their design, transparency, and integration within broader moderation strategies.

Abstract

Social media platforms are increasingly developing features that display crowdsourced context alongside posts, modeled after X's Community Notes. These systems, which we term Crowdsourced Context Systems (CCS), have the potential to reshape our information ecosystem as major platforms embrace them as alternatives to top-down fact-checking. To deeply understand the features and implications of such systems, we perform a systematic literature review of existing CCS research and analyze several real-world CCS implementations. Based on our analysis, we develop a framework with two distinct components. First, we present a theoretical model to help conceptualize and define CCS. Second, we identify a design space encompassing six key aspects of CCS: participation, inputs, curation, presentation, platform treatment, and transparency. We discuss key normative implications of different CCS design and implementation choices. Our paper integrates these theoretical, design, and ethical perspectives to establish a foundation for future human-centered research on Crowdsourced Context Systems.

Paper Structure

This paper contains 24 sections, 3 figures, 7 tables.

Table of Contents

  1. Introduction
  2. Background
  3. Related Work
  4. Generalized Model of Crowdsourced Context Systems for Social Media
  5. Methods
  6. Systematic Literature Review
  7. Design Feature Identification
  8. What Does the Research Community Know About Twitter-Style CCS?
  9. A Design Space for CCS
  10. Participation
  11. Inputs
  12. Curation
  13. Presentation
  14. Platform Treatment
  15. Transparency
  16. ...and 9 more sections

Figures (3)

  • Figure 1: Screenshots of a Community Note displayed on Instagram (left) and X (right).
  • Figure 2: Model of a Crowdsourced Context System. Users create posts, context contributors create notes about those posts, and ratings contributors create ratings for those notes. Based on these ratings, the system selects "helpful" notes, which are displayed alongside the post for users to see.
  • Figure 3: PRISMA flow diagram pagePRISMA2020Statement2021 for our systematic literature review, showing both the database search and the backwards reference search. We analyzed a final set of papers that combined the results of these two searches (n=56).