3 Day Three: The Future, or (maybe) The Era of Legitimacy 3 Day Three: The Future, or (maybe) The Era of Legitimacy
10/21
In this coming era of legitimacy we might focus less on theories of rights or studies of concrete harms, and instead emphasize process – the creation of frameworks for adjudicating disputes which can achieve buy-in from all relevant participants. Facebook has already taken a step in this direction with the external review board, a sort of “Supreme Court” which might externalize some of the most important content policy decisions that the platform makes. According to Kadri and Klonick, Facebook already relies on many of the tools used by courts to resolve tensions between regulating harmful speech and preserving free expression. What can we learn from Facebook’s earlier efforts to create a governance system based around user votes? Do these developments bring us closer to early internet proponents’ vision of cyberspace institutions that run by the rule of the people? And what are the promises and pitfalls of Facebook’s other proposed solution to its moderation woes: using algorithms to clean up content?
NOTE: In a conversation with Axios, Kevin Martin, Facebook’s vice president of U.S. public policy (...and former FCC chairman), favorably compared the proposed Facebook External Oversight Board to the Motion Picture Association of America’s system for rating movies or the Financial Industry Regulation Authority, a private organization that supervises the securities industry. How well do these comparisons stack up? Before class, please identify another model of governance from another industry that could inform how platforms ought to moderate content, and come prepared to (briefly and informally) explain this comparison.
3.1. “Mark Zuckerberg wants to democratize Facebook — here’s what happened when he tried” by Adi Robertson, The Verge (2018).
In February 2009, Facebook announced a new site governance system built around user votes. Why do you think this democratic model – under which some difficult policy decisions would have been made by users – might have been attractive to Facebook’s leadership? If it had been fully realized, what limitations might it have had?
3.2. Draft Charter of Facebook Oversight Board
3.3. “Facebook v. Sullivan” by Thomas Kadri and Kate Klonick, Southern California Law Review (Forthcoming 2019). Read sections II and III.
This article argues that platforms must create institutions like the Supreme Court to provide transparent decisions and constant rationales, and must create constitution-like charters to guide the independent institutions that would oversee them. Should we seek the constitutionalization of social media platforms? Does the creation of independent institutions solve Facebook’s content moderation problems?
3.4. “Google’s comment-ranking system will be a hit with the alt-right” by Violet Blue, Engadget (2017).
Google’s Perspective API, which uses machine learning to moderate comment sections, promised to avoid bias by focusing on deleting “toxic” comments that might deter people from a conversation. However, sentences such as “I am a black sex worker” or “people will die if they kill Obamacare” were identified as highly toxic and suitable for deletion. What does this tell us about the promise of using AI to moderate content?
3.5. “Just in CASE: Legislation Proposing Small-Claims Court for Copyright Disputes Headed to Congress” by Lauren Heperi, BYU Copyright (2019).
One other possible avenue for governing the decisions of online platforms would be to set up a small claims court analogous to the CASE Act. What role should the government play in moderation decisions?