• remi_pan@sh.itjust.works
    link
    fedilink
    arrow-up
    22
    ·
    1 year ago

    Like the author of this paper said, for me it’s not really about third party apps. The problem is that reddit try to monetize a content that is our collective property.

    • nude@kbin.social
      link
      fedilink
      arrow-up
      3
      ·
      1 year ago

      Is it our property though?

      Intellectually speaking yes, but legally speaking? Probably not. Chances are if its stored on their servers, it belongs to them.

      • ___hulk@kbin.social
        link
        fedilink
        arrow-up
        2
        ·
        1 year ago

        Legally, currently, yeah. Needs to change though. I’m ok with monetizing the presentation and delivery but if you want to use collective property you should still have to make that available as part of the deal.

        • Parallax@kbin.social
          link
          fedilink
          arrow-up
          5
          ·
          1 year ago

          Replacing “comment” with “artwork” kind of helps illustrate it. If we all made tons of artwork for Reddit, then they started gating it behind a paywall and while painters and all the behind-the-scenes painting staff earned nothing – well that’s kinda where we are today.

          • EnglishMobster@kbin.social
            link
            fedilink
            arrow-up
            3
            ·
            1 year ago

            But what you mentioned absolutely happens, though.

            It’s €17 to go to the Louvre. Many of the paintings there are public domain, which logically says they should be free as they have no owners. Yet to see them, most people need to pay €17.

            Those are paintings locked behind a paywall. The pieces may be donated freely by an artist - just as users contribute freely on a website - but the museum still charges for admission.

            So while I’m not defending the practice - and there are many free museums; even the Louvre has ways to get in for free - it’s also not exactly a way to convince others that the practice is inherently bad.

            • Pyr_Pressure@lemmy.ca
              link
              fedilink
              arrow-up
              1
              ·
              1 year ago

              Wouldnt the image of the painting be public domain but the physical piece be different?

              Like you could take a photo of the Mona Lisa and use it however you like, but the physical item itself is private property and access to it can be monetized?

    • Rentlar@beehaw.org
      link
      fedilink
      arrow-up
      2
      ·
      1 year ago

      Yeah, pretty much this. When I post anything on social media I intend to speak publicly to anyone who wishes to listen. I’m not speaking to Reddit, Lemmy, whoever in confidence, so no matter what the ToS says, my intent is any advice or anecdote I give is anyone’s to view and use, not in the content host’s sole ownership.

  • Samuel Proulx@rblind.com
    link
    fedilink
    English
    arrow-up
    12
    ·
    1 year ago

    I also think it is the greater good to let AIs train for free.

    I’m not entirely sure about this. Freely available AI’s, sure, maybe. But corporate closed source AI’s that charge per token? I’m not saying no, but I’d love to hear someone justify the thinking behind why this is “the greater good”.

  • GrandChicken@sopuli.xyz
    link
    fedilink
    arrow-up
    11
    ·
    1 year ago

    I figure its fair to charge for access if you pay the moderators and the contributors.

    Seems like they are taking over the moderator roles now, by force. Maybe the mods will also be the main contributors soon?

  • ZeroEcks@lemmy.blahaj.zone
    link
    fedilink
    arrow-up
    8
    ·
    1 year ago

    I am not sure I agree entirely with this, while I do think there should be third party clients, we have entered the age where allowing API access is giving free reign to very valuable LLMs to train on your data, which also IMO violates your users privacy. I think it’s better to have it be gratis but not libre, perhaps some kind of app approval process or some such.

    • PenguinTD@lemmy.ca
      link
      fedilink
      arrow-up
      2
      ·
      1 year ago

      I think it’s possible to have a middle ground, just by putting properly formed license terms like foss projects. Ie. Specify that the AI/bot must be following certain rules(ie, fetch the comments but not the user IDs), because if we don’t provide data for open and free alternative, there will be no good AI tools for common folks. And the top dogs are all hoarding data with sneaky ToS.

    • yuun@lemmy.one
      link
      fedilink
      arrow-up
      1
      ·
      1 year ago

      That was wild, yeah. Social media sites shouldn’t lock down APIs because it’s all user generated and LLMs should be given free reign to harvest user content for their own proprietary ends?

      I guess I can agree that these two ideas aren’t fully contradictory, but there’s a lot of friction between these two conceptions of user’s rights to their content/data.

    • RandomBit@lemmy.sdf.org
      link
      fedilink
      arrow-up
      1
      ·
      1 year ago

      Nothing can stop screen scraping. A nefarious LLM startup could hire an outside group to screen scrape the data and give it to them for training. Charging exorbitant amounts of money for the API hurts users far more than other companies with a profit motive to get the data.