There's basically only one instance.
There's only one PLC directory.
There's very few full relays (edit: appviews), none that I'm aware of that don't mirror bluesky censorship/moderation decisions.
There's basically only one instance.
There's only one PLC directory.
There's very few full relays (edit: appviews), none that I'm aware of that don't mirror bluesky censorship/moderation decisions.
- There's no "instances" so I don't know what you mean by this.
- Re: PLC directory, indeed, there is only one of those. I think this is a legit point but it's worth considering the whole point of PLC directory is to be the single minimal stateless open source part that lifts identities out of apps and hostings. PLC governance and maintenance is being spun out into a Swiss organization (https://atproto.com/blog/plc-directory-org). Longer term the idea is for it to have a similar role to ICANN. Here's more on that: https://youtu.be/9z0z-Qu66yM?si=_8Dcw1M3VSKFGZhm&t=493
- Re: full Relays, they're easy/cheap to run, and you can run one yourself if you think the other ones are coordinating with Bluesky and don't trust their decisions. You don't need to depend on something else to do that.
> PLC governance and maintenance is being spun out into a Swiss organization (https://atproto.com/blog/plc-directory-org). Longer term the idea is for it to have a similar role to ICANN.
And since that sounds like a massive centralization problem, how do we have a dozen more of them with independent governance that aren't all controlled by either the same legal entity or by whoever has legal leverage to compel that entity?
Well, I think you also need to consider what PLC is. It’s an open source implementation of an open source spec. The implementation holds zero private state and exposes a verifiable log of operations for audit. There’s ongoing work on mirrors and replicas. Also, its output itself is cryptographically self-verifying.
I get that it’s not ideal but I think it’s worth keeping in mind that there’s not much you can mess up with it other than refusing to update requests. The threat model is very limited and it would immediately be obvious that this is happening, killing the credibility.
I’d also call out that activitypub has the same threat model in the form of ICANN, as it’s also heavily dependent on DNS for identity. I believe these are reasonable trades to make; realistically the alternative is to use a blockchain, which few people are keen to do.
There are many relays, here's a list of 13 but it's not comprehensive: https://tangled.org/firehose.club/community-firehose-list/bl...
Bluesky's moderation actions are generally implemented as moderation labels which take effect at the AppView level. Sometimes they'll take down someone's PDS or censor from their relays, but I don't believe third-party relays are aware of those relay takedowns at all.
You're right, I was probably confusing appviews with relays.
Most decentralized systems I can think of tend to follow a power law distribution (roughly) where one or a few platforms dominate. In the fediverse that's mastodon.social (or maybe Threads), in email that's Gmail, in AP that's Bluesky. I'm not sure this is unique to the AT protocol?
I think the centralization on mastodon.social is a bit over estimated. According to fedidb.com there are over 1.1M active users in the fediverse. Of these mastodon.social has about 273k. That means mastodon.social has about 1/4 of the active users. The rest are scattered over nearly 43k servers. So I'd say the fediverse is pretty decentralized. That would make it pretty resilient.
On fedi there are several distinct pockets apart from the mastodon crowd which although don't dominate in raw headcount have their own flavour and unique view of the network (in 2nd place is "dark fedi", which has more shitposters, hackers and free speech adjacents; there's also a contingent of Japanese users mostly on Misskey instances). What I get out of AT is that this wouldn't really happen, since the content distribution would dominate your view with Bluesky posts regardless of your relationship with the rest of the network.
In AT you also get communities, in the same way that twitter had weird twitter, black twitter, and so on. But because you, by design, don't really see what PDS people use, it's more fluid.
I personally think that this actually leads to healthier communities but that may be a matter of taste!
Yeah, I mean a big reason why Fedi is segmented the way it is, is because the Mastodon crowd has a lot of merited and unmerited prejudices against the other parts of the network, and put a lot of effort into splitting it with centralized blocklists. The W3C working group that works with the AP spec (which refuses to cooperate with credible people representative of projects like Pleroma) is now looking to adopt a centralized moderation model similar to Bluesky's. It's one of the issues with the instance model, even though users have the tools to shape their own view of the network, to not have to see things they don't like, instance admins still get to dictate who they're allowed to talk to and what they see. I don't think that's right, even if the reasons are justifiable.
The guy who runs FSE (one of dark fedi's more notable instances) has written a lot of good blog posts on his side of the technical and social details, his one about running FediList is a good picture of the type of one-sided politics involved (and, if you're into the technical stuff, is rich with that too): https://blog.freespeechextremist.com/blog/about-fedilist.htm...
re: censorship & moderation, the folks @ BlackSky have been active in building out an alternative moderation structure (alongside running PDSes & AppViews).