Web 3.0 and The New Epistemic Threat to Democracy

I recorded the talk I gave last weekend at the American Philosophical Association Central Division. You can listen to it above.

In the talk I discuss ways in which Web 2.0 and Web 3.0 could threaten democracy. The bulk of it focuses on Web 3.0 and defends net neutrality as an appropriate and permissible preventative measure. Given the recent FCC ruling – it seemed like a good idea to post it here. Below is the handout from the talk for anyone who wants to have it in front of them while they listen. Here’s a google doc version of my handout.

  1. Central Aims

    • Articulate ways in which the semantic web (Web 3.0) threatens to put us in a collectively worse epistemic position and further undermine our democracy.
    • Defend net neutrality regulations as an appropriate and permissible preventative measure.
  2. The Evolution of the Internet
    • Web 1.0
    • Web 2.0
    • Web 3.0 or The Semantic Web
  3. The Beginning: Web 1.0
  4. The Web 2.0 Threat
    • Does the blogosphere put us in a worse epistemic position?
    • Posner
      • Error Correction
      • People are generally good at not trusting blogs without validation
  5. The First Web 3.0 Threat: Algorithmic Filtering
  6. The Second Web 3.0 Threat: Non-Net-Neutral Landscape

    • Net Neutrality = the thesis that internet service providers should be neutral and enable access to all content and websites available on the internet. Furthermore, they should not favor, speed-up, block, or slow-down content and services from any website.
    • Targeted information manipulation
    • Minority Voices via Innovative Tech Services
  7. Objections to Net Neutrality Regulation
    1. This is government regulation without good reason. Let the market correct for this.
    2. Reduce Incentive to Innovate (and provide cheap Internet access)
    3. Reduced Incentive to Expand Networks
    4. Increased Costs
      1. Advertising Revenue High, ISP charge to consumer low
      2. Advertising Revenue Low, ISP charge to consumer high
      3. Advertising Revenue Intermediate, ISP charge to consumer intermediate
        (Castles on the Rhine Effect)
    5. It’s Failing in Europe
    6. Chicken Little What-If-ing

 

Share on FacebookTweet about this on TwitterShare on TumblrShare on Google+Share on LinkedInPin on PinterestShare on RedditEmail this to someone