4 Matching Annotations
  1. Jan 2025
  2. social-media-ethics-automation.github.io social-media-ethics-automation.github.io
    1. Steven Tweedie. This disturbing image of a Chinese worker with close to 100 iPhones reveals how App Store rankings can be manipulated. February 2015. URL: https://www.businessinsider.com/photo-shows-how-fake-app-store-rankings-are-made-2015-2 (visited on 2024-03-07).

      This article shows how easy it is to manipulate app store rankings using fake downloads, which connects well to the chapter's discussion about bots. It’s shocking to see a photo of one person with so many iPhones, all just to cheat the system. I wonder if stricter rules or better technology could stop this kind of behavior and make rankings more fair.

    1. Bots, on the other hand, will do actions through social media accounts and can appear to be like any other user. The bot might be the only thing posting to the account, or human users might sometimes use a bot to post for them.

      Bots can act like real users on social media, which is interesting but also a bit scary. I feel like it's hard to tell if you're talking to a real person or just a program. Should there be a rule that bots must say they’re bots? It might help people trust social media more, but maybe fewer people would like the posts.

    1. Now, people who design and program social media platforms can do things to encourage or discourage users adding alt-text to their photos. For example, they could require or forbid users from adding alt-text to their photos, or they could do more subtle versions of those by putting reminders to add alt-text, or making the option of adding alt-text hard to find. Additionally, they could run a machine learning program to try to automatically generate alt-text for photos [b135], though those programs may make mistakes or have biases.

      I think encouraging people to add alt-text is a good idea, but it needs to be easy to use. If the platform hides the option, most users won’t bother. Also, machine-generated alt-text could help, but what if it gives wrong or biased descriptions? It might cause more problems than it solves.

    1. There are absolute moral rules and duties to follow (regardless of the consequences). They can be deduced by reasoning about the objective reality.

      I get the idea of having absolute moral rules, but what if two rules clash? For example, being honest vs. keeping someone safe. How should we decide which one is more important? It feels like sometimes we still need to think about the result.