With previous Rexit’s like the API debarcle etc. many users were left looking for an alternative, but with decision fatigue and bad UX etc. most did not find the Fediverse a viable option.

What needs to still improve, how can we be ready this time?

  • badgermurphy@lemmy.world
    link
    fedilink
    English
    arrow-up
    5
    ·
    12 hours ago

    One of my favorite tricks that a friend of mine showed me years ago was this:

    Put a check box or radio button somewhere on the page that will never end up visible to the end user marked with a label like “check here to verify you’re not a not” or “choose your ethnicity from this list or select prefer not to say”, then reject accounts that ever check those boxes, because a human never would. If you occasionally snare a blind person by mistake,they can email to bypass that with a human admin.

    I don’t know if it would trick modern bots, but he said it worked awesome back then.

    • douglasg14b@lemmy.world
      link
      fedilink
      English
      arrow-up
      1
      ·
      53 minutes ago

      It’s largely considered ineffective these days. Detecting elements that don’t affect layout is trivial, or elements that are occluded, transparent…etc

      Capchas are one of the best options. But even then, LLM users bypass those relatively easily, and LLM users are one of the biggest risk areas for astroturfing.