Skip Navigation
Asklemmy @lemmy.ml
Stomata @sh.itjust.works

What's wrong with bluesky?

Seriously i have zero idea what is going on with bluesky. I never used it. Why are people saying it's centralised? I also heard that a lot of people are joining it.

148 comments

  • https://lemmy.ca/comment/12906744

    I talked about it in this comment, which should hopefully still be recent enough to be accurate

    It's still too soon to tell what they will do. It's totally possible that they will take the necessary steps to be properly decentralized by transferring control of the registry + protocol to an independent non profit.

    Right now I feel that they don't have much of an incentive to do that, since the vast majority of their users won't care.

    I would love to be proven wrong

    • As a follow-up, if you have people on Bluesky you want to follow, go for it :) Community is important

      There is also a mastodon bluesky bridge that some people use to access both

  • There's nothing from a user experience currently that makes bluesky bad, it's just that since it doesn't seem to actually support decentralization, there's nothing to stop it from eventually getting just as bad as twitter over time due to profit incentive. Misskey/mastodin are the only microblogging platforms that are truly immune from corporate manipulation and enshittification, which would mean it's a long term solution (that while imperfect, can only get better).

  • One of the best reasons to want ActivityPub (or similar software) to become the primary way that social media sites are populated with information is that it divorces the particular front end you use from the content that is displayed. Meaning that if, in the future, someone writes a new front end that is better/faster/whatever it doesn't have to (most likely fail to) fight the network effect to have enough content to be worth using. So you don't have a David vs Goliath situation for every new, innovative social media site to get off the ground. Never mind Mastodon or Lemmy or Misskey or Mbin. Maybe ten years down the line there are a host of newer and better fediverse sites that are usable right off the bat because they have the same content available that these current sites have. Look at what a trial it's been to get any new social media site off the ground (Bluesky included). It's in every user's interest to remove individual sites' ability to squash competition via the network effect.

    Bluesky's model of decentralization does not allow for this so far as I know.

  • If you move from twitter thinking it'll not end up like twitter you're wrong. It'll go through the same growing pains process and you'll end up right back where you started with nothing to show for it.

  • I enjoy it, but I am fully aware that history could repeat itself, and I am ready to pack up and move if/when that time comes. For now, it's big enough that I can follow communities I enjoy being a part of without worrying about the constant influx of racists/fascists.

    Those people are of course present, but they're easy to block and move on.

  • People dislike it because it's not federated, but hot take: federation doesn't solve enshittification. It just devolves everything into little shitty internet fiefdoms. It doesn't do anything to prevent the inherent problems that arise as a result of having everyone freeball a random moderation structure, where they can outsource their agency to some guy they don't know, with the illusion that there's some clear set of rules or useful tools that exist somewhere off in the distance, being used by the "correct" actors and moderators. Which in turn means that everything becomes vulnerable to any abuse of the static, singular, broad rules, inside of these walled gardens that people are basically locked into.

    You get bait, you get ragebait, people taking advantage of the singular "algorithms" in order to game the system for maximum attention, and you incentivize that behavior because you make it way too easy to engage in. You get people paying to get on the front page of reddit, and you get eglin air force base being the most reddit addicted town. People think that AI abuse is some recent phenomenon, but it's not, bots have been on the internet forever, and people have been incentivized to engage in bot-like behaviors forever. Eventually you get a huge, hollow system, where everything has the guise of legitimate human interaction at the surface level, but is really just subject to this huge system of incentives and planned interactions which people are made subconscious of.

    You'd really need the ability to have account migration for a better decentralized network, and you'd probably actually just need self-hosting for everyone. You'd probably want blocklists to easily propagate around (+2 for bluesky), and you'd probably actually want those to have easily copied and pasted rules that could be shared between users to prevent spam and make it so abuse is less common and easily prevented before it happens.

    Which is what the usenet already had/has. It's just that the common consensus (which I believe to be false), is that the usenet is too hard to use, and requires demands too much intellectually from its users. If you decide to take this philosophy to the extreme, you end up with something like tiktok, where the idea is that people use their premade google account, scroll downwards forever, and that's it.

    I wouldn't mistake this as being some sort of like, natural occurrence, though, that's an intentional decision, made by businessmen, that want to maximize sales through an in-app store and control a massive cultural space. That's a specific decision that they've made, and they've tuned their platforms to take advantage of people's worst instincts in order to perpetuate that. Often with the assistance and explicit consent of governments which want these platforms to be used to track everything.

    They pour money into that system, it's an explicit decision they're making to push that onto people as a result of current economic and political structures, and it's due to those structures that they have that power to be able to do that, and due to those structures that these shit systems succeed, keep being cycled out in boom and bust cycles, over the better systems that people create.

    • but hot take: federation doesn’t solve enshittification. It just devolves everything into little shitty internet fiefdoms.

      Enshittification, by definition, is a result of profit seeking, especially from venture-capital funded projects.

      Shitty internet fiefdoms are shitty, but it's got nothing to do with enshittification.

      • Yeah, the broader point I'm making is that the federation doesn't solve the entire encompassing system in which this all exists.

        Federated projects both have their own problems in those shitty little fiefdoms, as said, and are probably never going to succeed in this broader economic context where huge, profit seeking, venture capital funded market actors are able to spin up a new twitter ripoff in no time at all. This is while similar market actors in the form of spam farms, bot farms, adversarial influencers looking to make a quick buck, and moderators themselves, have incentives to game whatever systems are in place on any platform, not just the large ones. This then increases the strain on smaller projects, and decreases their ability to actually be sustainable long-term, especially in comparison to these huge market actors and their platforms.

        The systems that are gamed, in the modern internet, are cordoned off and channeled by a bunch of moderators that we all trust to kind of do the work for the rest of us, apply the rules, use the tools to their discretion. Federation just makes it so you can jump from one moderated section to the other, one administrated section to the other, while on the same "platform". But it doesn't solve the inherent problems at play here, where moderators and higher level administrators are incentivized to make their platforms shittier with the invitation of advertisers, the invitation of more bad faith posters which can increase engagement, the adoption of shorter form, less substantive content, things like that. Those drive up traffic, and make more money, money they can use to then make their platforms "better", or basically, to eat up more of the market share. Eventually you play the short term gains game long enough, and then your platform's growth sputters out, and then venture capital dries up, and then you end up making the moderation more lax as a last resort, and then nazis come flooding in. Then the platform either dies, or mutates into a horrible shambling corpse.

        Even if you were to cut out all of that as a possibility, say, by trying to make your stuff copyleft, then you just cut out the route towards short term growth for anyone using your particular platform, and then you'll just get outdone by all of the other market actors which lean into that short term growth, while still filling your platform's niche, while using none of the specific parts of your platform.

        It's basically not going to succeed as an approach because it, as we keep learning on the internet over and over and over and over, it exists in a broader material context, the context of the market.

  • Nothing is wrong with it. It is just much earlier in the timeline of becoming twitter/xitter eventually. Maybe it'll take longer this time, maybe the change will be more subtle.

148 comments