• Rentlar
    link
    fedilink
    arrow-up
    26
    ·
    2 years ago

    Right in the article they are discussing adding ads and premium subscriptions features in the future.

    I can already see the timeline for BlueSky over the next few years:

    • Receive startup venture capital
    • Provide a useful, user-friendly service
    • Grow a userbase <-- (you are here)
    • Add a few more useful features, continue growing
    • Dominate the market, run at a loss
    • Add less useful features like ads, premium features, subscriptions
    • People complain, but most are too hooked in to leave
    • The useful features added earlier start getting removed, content gets sanitized <-- (Imgur, Reddit is here)
    • Big fundraising campaign, IPO, or bought out by big corp or the melon husk of the day.
    • Content quality goes way downhill, actively making life harder for users, volunteers and staff, bots rampant, more stupid features, monetization monetization monetization <-- (Twitter is here)
    • People leave for another useful, user-friendly service
    • CalcProgrammer1@lemmy.ml
      link
      fedilink
      arrow-up
      19
      ·
      2 years ago

      Any service backed by a business person will follow this timeline. This is why we need to move to self-hosted, federated, and community based platforms. Capitalism ruins everything it touches.

    • JillyB@lemmy.ml
      link
      fedilink
      arrow-up
      4
      ·
      2 years ago

      In the article, they outline how Blue sky is resistant to that path through the PBLLC structure and ability to move your account when you don’t like the web site.

      Time will tell if this model works in practice but it could be interesting as an alternative to Twitter without the usability issues normally associated with federation.

      • Rentlar
        link
        fedilink
        arrow-up
        2
        ·
        2 years ago

        I won’t knock BlueSky for trying. I think it will go well to replace Twitter, and run strong for some years. I’m still a bit on the skeptical side, especially until we see the ActivityTracker protocol on a non-BlueSky server in action, that BlueSky won’t be bought out and puppeted by some wealthy group.

        • JillyB@lemmy.ml
          link
          fedilink
          arrow-up
          5
          ·
          2 years ago

          I can see a potential path for it’s downfall.

          1. A competing service starts using AT and starts to get popular.
          2. BS starts losing money and slowly steers AT protocol to be more friendly to themselves and less friendly to competition.
          3. Competition forks AT, fragmenting the communities and the whole thing implodes.
          • Rentlar
            link
            fedilink
            arrow-up
            2
            ·
            2 years ago

            That is also part of people’s worry if one instance gets way too big on Lemmy, Mastodon etc. There is potential to change their federation scheme, closing themselves off from outside servers through unique features made incompatible to federate, eventually returning to a centralized model of its own.

      • robador51@lemmy.ml
        link
        fedilink
        arrow-up
        1
        ·
        2 years ago

        If the API standard is dictated by a for profit, there’s always the chance that at some point it’ll incorporate unfavourable features or impose limitations. To me Chrome/Chromium comes to mind, or even ‘embrace, extend, extinguish’. I’m not familiar with pbllc though, it might be different, but I’m sceptical all the same.

    • bionicjoey
      link
      fedilink
      arrow-up
      3
      ·
      2 years ago

      I wonder if because of decentralization the fediverse apps like Lemmy are insulated from going more than two more bullet points further past the “you are here”

      • Rentlar
        link
        fedilink
        arrow-up
        4
        ·
        2 years ago

        That is the hope… Lemmy and other Fediverse servers will still have challenges ahead of it in the future, including combatting botspam, targeted attacks to force servers offline (through DDoS, threats of lawsuit, etc.).

        • CleoTheWizard@beehaw.org
          link
          fedilink
          arrow-up
          4
          ·
          2 years ago

          I’m no expert but I’ll go through it. DDoS can be mitigated, but it’s a scaling issue. Usually it’s a problem for medium sized services. Lawsuits will fall mostly on individual server owners. And even then, the links are being hosted elsewhere. Botspam is a huge issue. I expect that to be one main problem. And also monetization. That will also be a big problem.