I'm tired of Bluesky
What the hell are we doing y'all
This wasn't what I had planned for my first blog post, but when inspiration strikes you gotta take it, right?
I have read, primarily in The Verge, that the product of social media is its moderation. Bluesky has a bad product. Even if they say they that their product is actually the AT Protocol, I still think they have a bad product! The protocol is what they have used to justify the lack of private accounts, blocks being public (even if they are otherwise quite strong), every post potentially "escaping containment" beyond just the following feed. For every good feature they implement like being able to detach quote posts, they have another that potentially enables harassment like being unable to remove yourself from a starter pack without completely blocking the creator (and not even soft blocking will do it).
The promise of decentralization and federation to protect against a possible billionaire takeover making the experience terrible is worthless if the experience is already terrible. Composable moderation isn't the answer either, that's just a way to avoid taking responsibility as the team supposedly in charge of Trust and Safety. It's only made worse by the poor discoverability of labelers and block lists, and asks users to put their trust into a random given user. And you can't trust them blindly, bad actors have tried to take advantage of this system already! One would hope that the official Trust and Safety team would help, but it shouldn't be a surprise to anyone that they're less than useless given that the team is lead by Aaron Rodericks, formerly of Twitter. Not exactly a site well known for best in class moderation.
I've buried the lede enough: after stunning inaction against Jesse Singal, apparently Rodericks decided that what required moderation action (a full suspension, no less) was a user creating a bot that just happened to show him liking a porn post on his main account. It's all public of course, thanks to the wonderful AT Protocol! I don't care that he's liking porn posts, or even that it was from some TikTok aggregator account (though that part is embarrassing), I care that this is what caused him to actually get off his ass and use the real moderation tools he has access to. And this isn't on the clock either, he's taking these actions past midnight local time.
Bluesky has shown what their development priorities are. They could have designed the protocol from the ground up with anti-harassment in mind. They could have implemented strong protections into at least the official app (or appview, whatever jargon they want) during the invite only period. They could have let posters choose whether or not posts would show up in "algorithmic" feeds. They could have responded to concerns about the implementation of starter packs by adding ways to remove yourself from them without permanent blocks. They have done none of these things. Instead they have been represented by a cowardly T&S team, they have chased user growth and engagement above almost anything else, and they have taken funding from Blockchain capital. Bluesky is not a platform that can be trusted to prioritize user safety. Maybe they're right and that's just a product that doesn't bring in revenue, but that doesn't mean we have to buy what they're selling us in its place.
I don't think I'm going to leave Bluesky, at least not yet. I love the artists I follow, people do post insightful writing to the platform despite the microblogging limitations, and quite frankly I don't want to jump ship from yet another platform so soon. I do, however, feel more strongly than ever that the right course of action is to curate your own feeds through RSS, make your own website, and not only to not trust large platforms, but distrust them. The nearly exclusive centralization of the web we have seen in Web 2.0 may be the path of least resistance for users, but it is not done in our best interests.
UPDATE:
Juni, the community developer that created the bot, was unbanned after around two hours. This is in spite of Juni appealing the ban and receiving a rejection email only 16 minutes after the original ban. Fastest investigation I've ever seen! Jay Graber, the CEO of Bluesky, responded to a user expressing concern about the ban and the lack of community trust in Trust & Safety. I do not care for her response, it seems to imply that the ban was carried out by a new member of the mod team. It contains no apology, and the only mention of future action (other than finishing onboarding of new mods) is "more options on atproto", which just reads as more composable moderation to me. As far as I can tell, there has been no acknowledgement of the situation from Aaron Rodericks.
None of this changes how I feel, and I stand by everything I wrote in the original post. Aaron is a coward, and Bluesky does not prioritize user safety. They continue to burn goodwill with community developers, the very community they are supposed to be empowering through their work on the AT Protocol. It is also not lost on me that Juni is trans, and the other recent T&S snafu was around inaction around transphobe Jesse Singal. It's impossible to know whether or not a cis community developer would have been banned in this situation, but I don't like the pattern that this falls into. Bluesky has had issues with moderation surrounding vulnerable populations before, and their proposed future solution was once again federation and composable moderation. At least they're consistent.
While I'm glad Juni is rightfully unbanned, I'm still tired of Bluesky and worried about their vision and priorities. Their responses to issues have been inadequate and more importantly their actions have been too. We'll see if any community developers are left to take advantage of the Skyseed VC fund other than grifters, or if they've all been burned one time too many.