Menu

A contemporary issues primer occasioned by Data & Society’s Who Controls the Public Sphere in an Era of Algorithms? workshop.

In this primer, D&S research analyst Robyn Caplan and D&S Founder danah boyd articulate emerging concerns and tensions coming to the fore as platforms like Google, Facebook, Twitter, and Weibo have overtaken traditional media forms, becoming the main way that news and information of cultural, economic, social and political significance is being produced, and disseminated. As social media and messaging apps enable the sharing of news information and serve as sites for public discussion and discourse about cultural and political events, the mechanisms and processes underlying this networked infrastructure, particularly big data, algorithms, and the companies controlling these information flows, are having a profound affect on the structure and formation of public and political life.

The authors raise and explore six concerns about the role of algorithms in shaping the public sphere:

  • Algorithms can be used to affect election outcomes and can be biased in favor of political parties.
  • Algorithms are editors that actively shape what content is made visible, but are not treated as such.
  • Algorithms can be used by states to achieve domestic and foreign policy aims.
  • Automation and bots are being used by state and non-state actors to game algorithms and sway public opinion.
  • The journalism industry and the role of the “fourth estate” have been affected by the logic of algorithms, and content is no longer serving reflexive, democratic aims.
  • Algorithms are being designed without consideration of how user feedback inserts biases into the system.

The authors also grapple with five different classes of tensions underpinning these various concerns and raise serious questions about what ideal we should be seeking:

  • Universality, Diversity, Personalization
  • A Change in Gatekeepers?
  • A Collapse/Re-emergence of Boundaries and Borders
  • Power and Accountability
  • Visibility, Accessibility, and Analysis

Finally, six proposed remedies and solutions to algorithmic shaping of the public sphere are considered and problematized. With each potential solution, the competing value systems and interests that feed into the design of technologies is highlighted:

  • Proactive Transparency
  • Reverse Engineering, Technical and Investigative Mechanisms
  • Design/Engineering Solutions
  • Computational/Algorithmic Literacy
  • Governance and Public Interest Frameworks
  • Decentralization in Markets and Technology

All systems of power are manipulated and there is little doubt that public spheres constructed through network technologies and algorithms can be manipulated, both by the architects of those systems and by those who find techniques to shape information flows. Yet, it is important to acknowledge that previous genres of media have been manipulated and that access to the public sphere has never been universal or even. As we seek to redress concerns raised by technical systems and work towards a more ideal form, it is essential to recognize the biases and assumptions that underpin any ideal and critically interrogate who benefits and who does not. No intervention is without externalities.

These varying tensions raise significant questions about who controls – and should control – the public sphere in an era of algorithms, but seeking solutions to existing concerns requires unpacking what values, peoples, and voices should have power.