User ban controversy reveals Bluesky’s decentralized aspiration isn’t reality

This just reads like not a response to the problem, but instead a description of technical implementations.

Concretely, nobody is going to subscribe to anything else in bsky but the default stack, anything else is unlikely to ever gain much adoption, and thus federation is basically meaningless and a technical implementation detail for how they wanted to manage their stuff.

Moderation labels are applied at the client level, not at appview level. Forks of the official client (e.g. deer.social) allow you to completely remove the official moderation labeller.

The users could reject exchanges and trade entirely on-chain, but that's expensive/complicated/risky.

It is a serious concern for cryptocurrency that most users don't even get the touted benefits because of reliance on exchanges.

It undermines the value of decentralization itself given intermediaries tend to pop up in all of these various platforms and become the dominant way to use them.

Nothing. That's the point. This wasn't a contingent outcome; it was inevitable due to the nature of the difficulty of accessing it for anyone except the nerdiest 0.1% of the population. So any mass adoption by its nature tends towards a centralized popular method of accessing it.

Correction: businesses are naturally going to end up prioritizing the desires of the majority of their customers

And for social media that isn't their userbase.

common people arent really using twitter or bluesky that much anyways. its just the crazies that are on there.

> It always comes back to the userbase. I don't know how many times technologists need to learn the lesson that normal people simply don't care about ideological technical principles like decentralization and often actually prefer the benefits of centralized systems like ease of use and typically stronger moderation.

The choice need not be limited to the familiar corporate hellscape vs decentralized usability nightmare dichotomy. Middle grounds can exist if we want them to.

I've seen a lot of general support for the criticisms and concepts described in this article:

https://www.noemamag.com/the-last-days-of-social-media/

Anyone who builds what they describe there can expect it to take off faster than ever.

That is true, but due to the smaller size and larger number of instances, it's easier to find a place where you won't get randomly banned without an admin taking the time to talk to you about what's wrong. My instance has about 1K active people on it and I've always been able to have a dialogue with the moderators and sysop. When things have gone wrong I had the opportunity to correct them.

I think there's a great value to the "small community" ethos that the fediverse supports much better than bluesky.

It would be strictly lesser if there were a significant free choice of App Views. BlackSky doesn't have one. NorthSky doesn't have one. You can currently be a BlackSky user, but if BlueSky bans you, your posts become invisible.

ActivityPub's lack of portable identity is a pretty serious problem, but the fact that significant portions of ATProto still rely on centralized infrastructure with no credible path to decentralization is pretty clearly worse[1].

[1] There are only two credible relays, and only one credible app view, and building/running either requires hundreds of times more capital than spinning up a Mastodon instance.

100% it doesn't make any sense to expect different behavior out of the same format just because you make slightly different promises about what the rules will be

Bluesky has been drama central since the beginning, consisting mostly of people who thought Twitter wasn't censoring enough (or censoring the wrong people), the free speech crowd came later and, well, tested the waters and found transphobic speech was in fact not free, and that despite distributed promises, the town wasn't big enough for the two parties to coexist

>I’d argue that people lose the moment they sign up for an account, deluding themselves into believing that the problem with the last nth iteration of the same thing isn’t them.

Indeed. As you said, it's the people, not the technical details of the "protocol" or "platform". My "favorite" Mastadon example: <https://news.ycombinator.com/item?id=34748195>

Sure there's bad uses, but putting yourself out there, on the public record, and cherishing those others around us doing the same is, in my view, divine.

I can't imagine the mindset that wouldn't want to be capturing some of the amazing wonderful world about them & the thoughts in their head & sharing them with others. I find these views about walking away from putting yourself online, seeing only the harm, as being deeply nihilistic & running away from clear amazing basically spiritual human value.

which is leaving out the full context, they specificly cited the alt-text of the image, not the image itself.

The email, verbatim read:

> A reply with an image; alt text reads:

> 'Charlie Kirk sitting in a white T-shirt that says "Freedom." A negative consequence follows!

[0]: https://bsky.app/profile/aliafonzy.blacksky.app/post/3m2jm7u...

That part actually still works, it's just that BlueSky imposes a moderation list that can't be disabled (which to me is the problem) in their aggregator. In theory this is fine because you can run your own with different rules but alternative communities don't. It seems very unlike BlueSky's general "you can turn off content filters" ethos to do this, I wonder why they don't just have everyone default subscribed to their list.

Speech is not violence, guilt by association is undemocratic, and this hypothesis of de-platforming as a tactic to limit uncouth ideas was thoroughly tested over the last ~15 years and demonstrably shown to be false: Trump, Alex Jones, and many others were banned across platforms. One of these people now sits in the White House, in part because of backlash to the deplatforming of him and others with similar politics.

Speech is not violence.

What is this for kind of logic. So if you have a store, and you happen to have a client that smashes someone else's head in your shop, then that means in your logic that you as a shop owner are guilty of violence.

But I guess you say you should know the reputation of every person coming into your shop, and if their reputation is deemed inappriopriate by a certain group, they should not be allowed into the shop to prevent them from harassing any other customers.

But how are you gonna regulate that? Who is gonna decide who is inappropriate and who isn't?

I think we already have a fairly well organized system for that: law & order. If someone breaks laws, they are punished for it. So if someone is violent, whether it's inside or outside a shop, they can be punished for it.

And you as a shop owner don't have to also individually take the effort to investigate and punish the individual. Although if you want to you have the freedom to; it's your shop in the end.

This makes some sense if Vance was a minor, fringe figure. But he was on a ticket voted for by ~50% of US voters. This is effectively saying that the goal is to have a social media site where half the country is not welcome.

The problem with that is two-fold. One, it neuters any political impact - you're effectively driving away the very voters you need to convince. And two, it creates an echo chamber that distorts reality because everywhere you look people are agreeing with you. Then 2028 rolls around and you're shocked that "the bad guys" won again.

Yes, blocking is fully supported. The goal isn't to ignore, it's to silence/de-platform.

I thought that was 2021. Did it happen earlier too?

Aheming about "the right" aside... there exists people across the political spectrum that think the way they do because of a variety of reasons. Some because they read the literature and arrived at the conclusion, some because that is the economically or socially advantage position to be in, and some because that is just their political position tautologically.

I think the core Bluesky team has been discussing ways to limit toxicity so I definitely think this is on their radar. Substack also leans into the idea that their platform manages social media toxicity better than other platforms, at least in their marketing copy. I think in 2025+ toxicity is a major dimension to evaluate social platforms on.

There is, it's almost every successful advertisement-driven social media site. Youtube, reddit, facebook, etc. have all adapted to remove negative and controversial comments and leave positive/advertiser-friendly comments.

This is the culture we have tried to foster over at Discuit (small open source feed-style social media site).

Granted, we had a huge Imgur user migration, so that's the current flavor of the content, but the OG userbase was aiming for a culture that is perfectly defined by "thoughtful positivity."

> I do wish there was a community that encouraged thoughtful positivity over negativity

Yeah, it's called the real world. We have millions if not billions of years of experience in dealing with in-person differences, tentative contacts with members of neighbouring tribes and smoothing out conflicts. One will find the risk of getting smacked over the face for being an absolute idiot to be a very effective motivator to remain on our best behaviour with strangers.

On the internet these days, you can only choose between an echo chamber, or all-out culture war.

> In the AT protocol, your identity isn't tied to a server.

It kinda is. Your identity is in the hands of whoever controls the PLC Directory. You can argue this is better than having your account and identity on the same place, but it still a third party that you depends on.

I know, that's a terrible image, and I should be the last person suggesting such an image.

I was mentioning one use case that's a showstopper that much of HN would appreciate, and which forces a lot of the people who attract other people to be on the proprietary platforms.

Of course a risk of making Fediverse become more popular is ruining it for those people who want very small-neighborhood community. And who don't mind that, on any given topic, they're basically stuck with the handful of people who happened to be there, not those who would be most interested.

I'd still like to make Fediverse more welcoming to more people, to get more people on it.

I have an optimistic belief that a more popular Fediverse could be better than any of the commercial alternatives. Though it would become more like them in some ways, as it grows.

Because with mastodon, you can just run your own server, and continue using the network.

Between my single-user-instance and Facebook's billion-user Threads, ActivityPub is genuinely decentralised.

The irony of that example is that it is not the same as BlueSky here. There were huge parts of the FediVerse entirely untouched by what happened at Mastodon.LOL .

It already is and has been for a long time. The kinds of content you have to moderate are not things anyone wants to look at. The worst ones are far worse than you can imagine and the average ones are nudes of people you don't want to look at all day.

This is actually today's controversy on Bluesky because the #1 attribute of its power users is they're terrified of "AI" and the idea that "companies will steal their posts to generate AI slop", which means they think the ML moderation is stealing their posts.

Oh, but the ML can't be decentralized because the training datasets are illegal.

The big example that comes to my mind is Matrix, where most homeservers use Mjolnir to apply centralized public blocklists of other servers/people they don't like.

So if for example #archlinux disagrees with your opinion and they decide to ban you for it, you are now banned from many other unrelated channels.

I have also seen subreddits that auto-ban users that have ever posted in specific other (unrelated) subreddits.

But don't you have to have some sort of ranking system for discoverability?

Mastodon is pretty much anti-that. No algorithm. If you don't like something, you just don't follow it. Nothing defaults to the instance/global timeline either and you're free to mute anything.

Very much a conservative.

His most controversial post on bluesky that I saw, he commented on Obama saying that perhaps they should just shutdown the government and with the saved tax $, they can afford to buy healthcare.

The deal would be counted in equity

This is so funny that with a self-hosted PDS you can just tell BlueSky "trust me bro." What an absolute joke these age verification laws are.

We were content with "Are you 21 Y/n" for sites featuring alcohol for years, why did we have to go and mess it up?

I believe a PBC designation is somewhat meaningless; OpenAI was a PBC until a few months ago.

PBCs are absolutely corporations. They incorporate and that makes them corporations. It's in the name for Pete's sake.

He kind of did this by making it so you can see the posts of people who block you. Of course, he did this because he's so annoying everyone blocked him.

You seem to have missed that Musk nurturing a cult of mentally ill edgelords is a feature, not a bug. Why would he ever want to shield his users from the made-up ragebait that so successfully radicalized so many of them into joining MAGA?

Ahhh, makes sense. I vaguely thought they already had an app-view, though looks like they will soon!

According to the post, it's a usability issue of the open source app layer and not some failure of moderation principles. "Blacksky is dependent on Bluesky’s application server to give users a fast experience, which also means that it is dependent on Bluesky’s labeling system and its moderation choices." Also, Singal's name is mentioned multiple times throughout the the article in irrelevant contexts. I still see the agenda here and I've seen this tactic before.

I can see that. I am even guilty of using cat a bit too much.

It seems a bit random though. Are mastodon people big on trigger warnings? I never signed up for it. Heck, I never signed up for twitter or bluesky either. For mastodon I considered it, but telling people "ha, ha, I am writing toots" was a little too silly, even for me.

I guess? They just seemed to reference some story or situation I am not familiar with

>They're currently working on inventing new slurs for AI chatbots which are mostly remarkably offensive.

That sounds awesome. Thanks for the tip. I'm gonna join this effort immediately.

Slurs for AI? Maybe take to heart the idea that the Humanities are useful in that you don't end up having thoughts like this if you've studied them.

Neckbeards, neckbeards whatcha gonna do?

Whatcha gonna do when we come for you?

Your criticisms have some validity, but email is in a sense objectively great in that it's the thing that "won."

I believe there's a universe of difference between ONE provider and a few megacorps plus weirdos like me who pay for my own domain; I actually think that would ALSO be a good model for twitter-esque microblogging the way it pretty much works for email.

Because Bluesky actively sells itself on the whole "take your stuff with you thing?"

At least with Masto, you know the possibility of being shutdown if you don't trust your provider, which is the same as email?

Meh. Handle a little spiciness and participate in the actual topic of discussion, or don't.

Holy. Moly. I had no idea that this was another goal. That's ridiculous to try to do at the protocol level.

What I'm realizing, perhaps as an old timer -- projects like this seem to utterly forget the concept of "layers," despite, you know that being the whole internet, OSI and whatnot.

I remember having the idea that even this exact use case, air-quotes "twitter", could be done by simply layering something over email, e.g. you shoot an email to "posty-instance@myserver" or whatever and then some central-ish thing posts that to a public feed or whatever.

So much unnecessary reinvention.

To take the example of trans: yes. Multiple different groups have genuine concerns about trans issues. For example, you have trans people wanting to be respected and not victimized, you have women who fear men being in their spaces, you have parents who fear their children being harmed and encouraged to transition, you have female athletes concerned about unfair competition from people who went through male puberty, etc... None of these groups innately hate each other or are bigoted, they just have different opinions on what the best society should be like as shaped by their own experiences.

And all of these groups have a point. Obviously you do get examples of trans people being victimized, you do get examples of children regretting transitioning, you do get examples of unfair athletic competition, etc... It is a complete conspiracy theory to label all of the groups as made of hateful fake news spreading bigots. They are all just human beings, muddling through the world!

> he doesn't go out of his way to avoid these controversies.

Science journalist covering the science of perhaps the most salient social issue of our time?

Don't get me wrong -- there are other really important stories which aren't being covered as well as they should be, but Singal's beat would seem to be at least as important as... most anything linked from HN on a daily basis.

> People who vocally wish he was less influential would do well to remember how much juice they've been giving him.

I totally agree. By turning Singal into a boogeyman he almost certainly isn't, people are only feeding his social and journalistic capital. Singal is just someone who disagrees with you (in that classically liberal 90s American sense). Efforts to cancel or ban him have made some look like they don't have actual arguments to contribute.

That's how Singal makes those with opposing views look ridiculous.

> he's one of journalism's answers to the concept of edgelordery

either the term "edgelord" means something very different from what it did 10ish years ago or you have a very misinformed understanding about how timid Jesse approaches online conflict. I haven't seen him initiate any form of edgy engagement in forever .. maybe ever.

> And pretending that (in the United States of America of all places) teens can just walk up to their doctors super easily and get surgery done somehow.

You should see some of the background evidence on this. NY Times did a podcast series not long ago, which covered the controversy, called The Protocol: https://podcasts.apple.com/us/podcast/the-protocol/id1817731...

It was about how the so called Dutch Protocol re: gender affirming care first came to the US, and how long-ish psychological evaluation times were often truncated because of how far patients had to travel in the US. Then the lid flew off when such care moved to California.

> Singal is a fellow traveler to people who defend conversion therapy

Yikes. One doesn't usually expect people to use the actual language of McCarthy-ites.

> So who is Singal out there fighting for?

The kids? It's really not crazy to believe that the children, perhaps not given adequate care, could simply be wrong, or even socially influenced (like everything else in a teenager's life).

> He's going out there writing this kind of stuff so consistently that he's known for it. And I have a very hard time understanding who he is the voice for, except for people who are just against trans people existing in public society.

It's a new treatment and how it is practiced here in the US is the very Q he is examining. I'm not saying he's right, but I do feel better if there is someone with a critical eye looking at new treatment methods.

What is Singal misrepresenting? His writings on the subject mostly serve to highlight just how poor the evidence around gender affirming pediatric care is. And he's hardly alone in this conclusion. The UK, Sweden, Italy, Denmark, Norway, and Finland have all stopped routine prescription of hormone blockers to children. Most countries never allowed it in the first place.

> Well, no, he did clearly break the TOS back when he originally joined.

Care to explain? The links in the article re: potential violations are mostly BS.

My understanding is their TOS was unclear and they clarified it after the outrage, but their moderation policy didn't actually change. They're not going ban people that break the TOS outside the S, because that's practically unenforceable.

> would lean to a ban of probably 85% of the US

A decentralized system would allow for that to happen tbh. That 85% can exist in their bubble but other actors who see them as dangerous and unsafe should have the means to mute/disconnect.

You'll need to explain this a little bit more, because the TC article seems to indicate the issue is that he is accused of targeted harassment. I doubt you could design a poll to make people vote 85% in favor of targeted harassment.

Even better: the only “evidence” of Singal’s “anti-trans” views are that his work has been quoted by anti-trans politicians and activists. This is an absolute ridiculous bar to have. Anyone who follows him would be hard-pressed to describe him as “anti-trans” unless you think anything less than a full throated endorsement of self ID and medical transition interventions for minors is “anti-trans”.

There was an article about this issue on techcrunch yesterday, and it linked to a long discussion on Bluesky about whether "clanker" was a slur that should be banned. JC. What a waste of time by some people.

>if applied to everyone, would lean to a ban of probably 85%

I'm not American but looking in from the outside i can't figure out anything 85% of you would agree on there right now.

The _actual_ trouble with Bluesky is that it's just a retread of Twitter with only a few minor tweaks, right down to the inevitable moderation and governance problems from shoving every user into a single shared social space and having a single centralized point of failure.

It's not 2012 anymore, and the modern mainstream social media ecosystem has turned into an utter disaster area. If you're going to do social media in the 2020's, you need something better, not the same tired ideas and empty promises about "We'll do it _right_ this time around, honest."

Mastodon, for all its faults, at the very least was truly and demonstrably decentralized.

Facebook and Twitter no longer bother with moderation, do you think the experience on those platforms is better or worse than 5 years ago?

It used to be common sense to immediately ban creeps of all stripes, especially the obvious ones. Singal certainly qualifies. Putting aside the super annoying 'just asking questions' vitriol that he publishes to national papers, his pdf-file chat log stuff alone would warrant an instant perma from me without a second thought.

Speaking to your broader point about the 'death' of a platform. The people that made bsky what it is now (good and ill) are precisely the people you are blaming for its downfall, which is weird. Normally when you run a business you want your users to remain so that you might profit.

Relatedly, I'm very very tired of the 4chan/crypto/ai gas-leak that has enshittified everything, aren't you?

I've seen bsky users chat casually about their rpe and death-threat ratio before and after leaving twitter, and for that alone, I would choose the 'threat lite' platform for as long as it remained so.

> "The question is not whether those views are right, it's whether those are mainstream."

I don't agree that this is the question, nor do I agree with the your unsupported number. This isn't an election, and popularism is a coward's appeal. Was the Gaza genocide not a genocide until the polls caught up with what we could all see was happening?

I don't even think you believe what your wrote. If the 'views' in question are truly shared by 85%!* of a population that never agrees on anything, then surely there's no problem with sharing them on this forum? A guarantee of 85% positive karma is awaiting you if you just speak your truth. It's the Trump era, and you can say the 'r-word' and the 't-slur' now. What was actually holding you back?

> Bluesky has a problem of its user base demanding purity, and it will 100% be the death of it.

Yup.

As someone who describes themselves as Leftist, I wanted to enjoy Blue Sky, but the purity tests are insane.

To give a concrete example, I was called a Nazi for owning a Tesla Model 3. The fact that I bought the car 6 years ago, long before Elon Musk made his hard-right turn, was irrelevant. They literally expected me to sell the car and take a huge financial loss (Since I need a car and would then need to buy a new one, and the trade-in value of my M3P is shit) just to virtue signal.

That's an argument that proves a lot, given that it kind of implies that a quorum of customers can simply vote other customers off the island.

This can backfire. eg Twitch moderators.

Discretion should be rarely used. For everything else, create a set of rules and stick to them.

Good on Jay Graber that she's sticking to a reasoned set of principles instead of bowing down to the braying mob.

He /doesn't use the service at all/. I agree that is a way of not breaking its rules.

But I guess users also expect that there's freedom of opinions on the platform within the set rules. If users are merely being banned based on their opinion, that doesn't sound like a healthy environment.

I think there’s a lot of misconceptions about atproto; unlike ActivityPub where your home server runs both the application as well as stores your data, atproto is build in the concept that “applications” like BlueSky store your data inside your PDS (Personal Data Store), which may or may not be hosted somewhere else.

While anybody can host their own PDS, the public bsky.app instance can and will “block” users by preventing their login and not pulling data from the PDS of the banned user or showing it in feeds.

Since, to date, the BlueSky “AppView” (the service backend itself that handles aggregating data, generating feeds, etc.) continues to be closed, being banned from the public instance is effectively being banned from the network. The data model (lexicon) is well documented and somebody else is free to write their own, but, for now, BlueSky is just as centralized as other platforms even if you can store your data elsewhere.

Just because something is common and widespread doesn't mean it should continue to be common and widespread, though it will continue to happen due to human nature. And yet, people striving to be constructive and positive won't celebrate the death of a stranger. "Do unto others as you would have them do unto you" is a cultural lodestar found globally for a very good reason: cheering the death of others may lead to others hoping to cheer your death, and that potential is enough to significantly curtail offers of constructive and positive assistance from the victims to the perpetrators, leading to a gradual social degradation within the perpetrators. Certainly remember and even memorialize a person's death, but the exaltation of a person's death is a sure path to cultural collapse.

Now, that's assuming people are one unified group. In reality, most people are forced into an "in" group or an "out" group. The "in" group exalts the death of the "out" group member, so the "out" group members must respond in kind. That eventually leads to the degradation of both groups, leaving the "above" and "beyond" groups with the remnants. In turn, the destructive and negative conflict continues.