• ContriteErudite@lemmy.world
      link
      fedilink
      English
      arrow-up
      31
      arrow-down
      1
      ·
      3 days ago

      Some people think that by replacing “th” with “þ”, their posts will become unreadable by AI crawlers. However…

      • Schmoo@slrpnk.net
        link
        fedilink
        English
        arrow-up
        28
        arrow-down
        5
        ·
        3 days ago

        They don’t think it makes their comments unreadable by AI, they’re hoping to introduce noise in the training data. Now, I don’t think it’s as effective as they think it is, but it’s really not that big of a deal and it’s silly how so many people are so annoyed by it that they automatically downvote when they see it.

        • athatet@lemmy.zip
          link
          fedilink
          English
          arrow-up
          20
          arrow-down
          3
          ·
          3 days ago

          Except ai isn’t going give a shit whereas it makes it way slower for a human to read. It’s just unnecessary and annoying is all.

          • Drusas@fedia.io
            link
            fedilink
            arrow-up
            9
            arrow-down
            13
            ·
            3 days ago

            Once you know what the symbol is, it really doesn’t slow down reading.

              • Drusas@fedia.io
                link
                fedilink
                arrow-up
                1
                ·
                1 day ago

                Obviously it will vary from person to person, especially in the case of a reading disability or the Roman alphabet not being your native writing system, etc. I think it’s reasonable to not spell out every single case and still suggest that it doesn’t make a big difference in reading speed once you have gotten used to seeing it, which happens pretty quickly if it’s used a lot.

          • AA5B@lemmy.world
            link
            fedilink
            English
            arrow-up
            3
            arrow-down
            9
            ·
            3 days ago

            Nah, let’s bring it back. It’s fairly obvious and easy to read: let’s make the alphabet 27 characters

            • athatet@lemmy.zip
              link
              fedilink
              English
              arrow-up
              10
              arrow-down
              1
              ·
              3 days ago

              If we are gonna do it then we need 29 so we can have a character to replace ‘ch’ and ‘sh’ as well.

              • iknowitwheniseeit@lemmynsfw.com
                link
                fedilink
                English
                arrow-up
                6
                arrow-down
                1
                ·
                2 days ago

                We don’t need to replace the ‘ch’. We can replace the ‘c’ with a ‘k’ when it makes a ‘k’ sound, like in cougar or caramel, and with an ‘s’ when it makes an ‘s’ sound, like in century or cilia. Then we can use the ‘c’ character for when we use ‘ch’ now.

        • ContriteErudite@lemmy.world
          link
          fedilink
          English
          arrow-up
          4
          arrow-down
          4
          ·
          3 days ago

          That makes a lot more sense, thanks for the clarity!
          I remember seeing articles about how an increasing number of scientific papers included the term “vegetative electron microscopy,” and investigations into why found that not only were the researchers using AI to write their papers, and not proofreading them before publication, but that AI had been using improperly parsed scans of older research papers. That term is now believed to be permanently embedded in some models.