There are no instances in ATProto • Hacker News Board

Re: There are no instances in ATProto

#151

Post by embedding-shape » Fri, Jun 19, 2026, 8:47 PM UTC

echelon wrote: post #112

embedding-shape wrote: post #99

Earlier quoted context omitted.

> why would you assume that when someone asks "where are the instances?" they're not using the common mainstream use of the word "instances", like, servers, or running software, or VMs, or containers? Of course depends on the context, but in a lot of discussions about ATProto, ActivityPub, Mastodon and nearby areas, people talk about "instances" as in "ActivityPub instances that host my data and my profile uses its U…

All of this goes away if we just do P2P social media. Swarms of content. Cryptographic identities and content signing/attribution. Cryptographic hashes for content uniqueness/immutability. Immutability in general. Ephemerality (content lives as long as some node cares to retain it, otherwise it gets forgotten). Concrete but extensible ontology for core concepts. You don't need login. You don't need to agree on a comm…

> All of this goes away if we just do P2P social media.

This is the wrong way to see it. There is no "Best and correct" solution, only solutions with different trade-offs. ActivityPub/Mastodon/Federation makes sense in some cases, "pure" direct distributed P2P makes sense in some cases, one central server makes sense in some.

Bluesky/ATProto just made different trade-offs, for different use cases, some of which wouldn't have been possible without the architecture they ended up with, which sibling commentator expanded on exactly what.

Re: There are no instances in ATProto

#152

Post by fabrice_d » Fri, Jun 19, 2026, 9:11 PM UTC

msla wrote: post #138

fabrice_d wrote: post #134

Earlier quoted context omitted.

> You can’t even accurately calculate things like post likes. And maybe that's a good thing.

Technical problems give way to philosophical differences but the over-arching problem is that the people behind ATProto really want to make a social media ecosystem that attracts lots of average people who will refuse to understand that the solution you're giving them can't do things Twitter could do back before Musk bought it. People get angry enough at Bluesky not having an edit button, and it's at least possible t…

You can switch to the AppView from https://mu.social/ and you'll get editing working.

Re: There are no instances in ATProto

#153

Post by dbingham » Fri, Jun 19, 2026, 9:15 PM UTC

notthemessiah wrote: post #96

p4bl0 wrote: post #82

I think the analogy presented here is broken. RSS doesn't depend on Google Reader at all. Even at its prime, RSS depended less on Google Reader than email depends on Gmail now. In ATProto, AppViews heavily depends on Relays to be useful, and Relays are quite expensive to run. Also, the yellow circles which represent blogs in the RSS illustration are really not of the same nature as the same circles which represent po…

> Relays are quite expensive to run While relays are among the more intensive parts of AT Protocol infrastructure, their cost of operation is still something most people can afford: approximately $30/mo now. What is truly expensive and difficult is something that will be immutably so regardless of how centralized or decentralized you are: moderation. The author of this piece wrote about this common misconception abou…

"Most people" cannot afford $30 /month. And of those who can, most of those cannot afford the time and effort required to run a relay.

And I honestly think this is one of the fundamental problems with the push back towards protocols and decentralization. We're overestimating the bandwidth and capabilities of the average user and we haven't fixed the problems that pushed everyone towards centralization in the first place.

Take me, for instance. I am not only capable of running my own Mastodon or Atproto data server+relay -- I'm technically capable of writing my own ActivityPub or Atproto app.

But I'm currently sitting with accounts on bsky.app and mastodon.social -- the biggest most centralized "instances" (yes, I know, but it reasonably describes the problem). This is because I do not have the time or mental bandwidth to even pick an "instance" that would be better suited to me and migrate, let alone run my own.

And this is doubly and triply true for the average person who doesn't have the technical abilities I have.

As a result, both Mastodon and Bluesky are still practically centralized to a large degree. An overwhelming majority (more than 90% last I found data) of Bluesky users are hosted by bsky.app. Similarly on Mastodon, a large plurality of users (~20%) are on Mastodon.social. Mastodon's obviously doing better than Bluesky in this regard, but it also has about a quarter of the overall traction, and I'd honestly put that down to Bluesky's apparent centralization which makes it a lot easier for people to join and wrap their heads around it.

Re: There are no instances in ATProto

#154

Post by Fraterkes » Fri, Jun 19, 2026, 9:18 PM UTC

danabramov wrote: post #145

Groxx wrote: post #130

Earlier quoted context omitted.

it's somewhat similar, yeah. minus the part where bluesky itself is by far the majority host of people's data. that puts it more in the realm of "Office 365 Online + OneDrive storage" than "Word" - a lot of people will lose a lot of data, though something resembling it can be started up again. and people with backups (their own PDS) will just move to OpenOffice for a bit. Blacksky finishing their full forking does fi…

Yeah that’s a fair clarification. At the very least I think Bluesky hosting should start doing something for automatic backups.

I'm sorry, just to clarify: in your scenario where the app/company is suddenly vaporized from the face of the earth, if that happened to Bluesky right now it would effectively mean that >90% of content currently published using Atproto would be lost?

Re: There are no instances in ATProto

#155

Post by shimman » Fri, Jun 19, 2026, 9:25 PM UTC

notthemessiah wrote: post #96

p4bl0 wrote: post #82

I think the analogy presented here is broken. RSS doesn't depend on Google Reader at all. Even at its prime, RSS depended less on Google Reader than email depends on Gmail now. In ATProto, AppViews heavily depends on Relays to be useful, and Relays are quite expensive to run. Also, the yellow circles which represent blogs in the RSS illustration are really not of the same nature as the same circles which represent po…

> Relays are quite expensive to run While relays are among the more intensive parts of AT Protocol infrastructure, their cost of operation is still something most people can afford: approximately $30/mo now. What is truly expensive and difficult is something that will be immutably so regardless of how centralized or decentralized you are: moderation. The author of this piece wrote about this common misconception abou…

$30 a month is a pretty massive costs for such a program. Why does it require such a beefy VPS, is it just the initial bootstrapping that requires such resources?

Re: There are no instances in ATProto

#156

Post by neko-moe » Fri, Jun 19, 2026, 9:31 PM UTC

danabramov wrote: post #145

Groxx wrote: post #130

Earlier quoted context omitted.

it's somewhat similar, yeah. minus the part where bluesky itself is by far the majority host of people's data. that puts it more in the realm of "Office 365 Online + OneDrive storage" than "Word" - a lot of people will lose a lot of data, though something resembling it can be started up again. and people with backups (their own PDS) will just move to OpenOffice for a bit. Blacksky finishing their full forking does fi…

Yeah that’s a fair clarification. At the very least I think Bluesky hosting should start doing something for automatic backups.

They're funding Fig to create and run a "full network backup" solution. What Bluesky really needs to do is figure out a way to get users to own their own backup recover key, whether personally or through some third-party service.

Re: There are no instances in ATProto

#157

Post by danabramov » Fri, Jun 19, 2026, 9:31 PM UTC

Fraterkes wrote: post #154

danabramov wrote: post #145

Earlier quoted context omitted.

Yeah that’s a fair clarification. At the very least I think Bluesky hosting should start doing something for automatic backups.

I'm sorry, just to clarify: in your scenario where the app/company is suddenly vaporized from the face of the earth, if that happened to Bluesky right now it would effectively mean that >90% of content currently published using Atproto would be lost?

No and yes and no / it depends.

Realistically, I can't say "yes" because I'm sure there's plenty of copies of entire network by now. They would be out of date but would have all old records. So that could be maybe 70% that's already backed up. I guess they likely won't include images/blobs. There's an ongoing project to build an always-available full archive with this specific purpose (https://atproto.com/blog/introducing-hubble-a-public-mirror-...) so it is also an active area of work.

If we imagine that nobody has a full copy today or is unwilling to share it, the answer would technically be yes.

I'd still say that, for an app going down, the answer is "no" because "Bluesky app" and "Bluesky hosting" are like two separate services. The point I was making was that specifically "apps going down doesn't destroy data". (The distinction between "Bluesky app" and "Bluesky hosting" isn't completely contrived because I'd expect the cost of running the app to be many orders of magnitude higher than the cost of running hosting.)

But if you pick a hosting company, and users don't have backups, and nobody does mirroring, then yes, hosting disappearing would destroy data. As with literally any hosting.

Re: There are no instances in ATProto

#158

Post by kmeisthax » Fri, Jun 19, 2026, 9:33 PM UTC

INTPenis wrote: post #34

ATproto sacrifices true decentralization for consistency, Mastodon and AP does the opposite, sacrifices true consistency for more accessible decentralization. At least that's how I understand it, because running an AP node is much more accessible to regular selfhosters than running one of those content relays in AT. So all you'll ever "decentralize" in AT is your own data, it's more about owning your data rather than…

Mastodon also has content relays!

But really, I actually would argue the opposite: ATproto is (or at least, wants to be) more decentralized.

In the ActivityPub world, identity, application, and hosting are intrinsically linked. If I want to use Lemmy, I can either register a second, permanently separate ActivityPub account on that Lemmy instance, or ONLY use Lemmy to the extent my Mastodon instance knows how to send messages that Lemmy understands. EVERY new ActivityPub app is a new set of interoperability concerns, because each app owns its own identity and hosting. There's no way for my self-hosted Mastodon instance to provide an identity to a Lemmy server and then for that Lemmy server to tell that instance to host content on my behalf.

That's the bare minimum you'd need to match ATProto, at least in the version being told to me by the ATProto people. No idea if any of it applies to actually-existing ATProto, in the same way that actually-existing ActivityPub can't interop the way ActivityPub supporters claim it can.

Re: There are no instances in ATProto

#159

Post by ryukafalz » Fri, Jun 19, 2026, 9:36 PM UTC

chokolad wrote: post #133

NoGravitas wrote: post #89

Earlier quoted context omitted.

The scale-down floor for running an ATProto relay or appview is much, much higher than the floor for running a fediverse instance.

You can scale down as much as you want. You don't need to run full relay if you want to follow only a dozen of accounts. I bet you can run something like that on a raspberry pi or something similar. You will not get the search over all of the network, but that's something you don't get with your personal mastodon instances either.

Wouldn't you then not be able to see replies from anyone besides the dozen accounts your relay follows too? If I run a personal Mastodon instance and someone replies to one of my posts, their instance will send it directly to mine and I'll see it. My understanding of the ATProto architecture is that it doesn't support directed messaging like that.

Re: There are no instances in ATProto

#160

Post by supern0va » Fri, Jun 19, 2026, 9:40 PM UTC

dbingham wrote: post #153

notthemessiah wrote: post #96

Earlier quoted context omitted.

> Relays are quite expensive to run While relays are among the more intensive parts of AT Protocol infrastructure, their cost of operation is still something most people can afford: approximately $30/mo now. What is truly expensive and difficult is something that will be immutably so regardless of how centralized or decentralized you are: moderation. The author of this piece wrote about this common misconception abou…

"Most people" cannot afford $30 /month. And of those who can, most of those cannot afford the time and effort required to run a relay. And I honestly think this is one of the fundamental problems with the push back towards protocols and decentralization. We're overestimating the bandwidth and capabilities of the average user and we haven't fixed the problems that pushed everyone towards centralization in the first pl…

>"Most people" cannot afford $30 /month. And of those who can, most of those cannot afford the time and effort required to run a relay.

I don't think that was a recommendation for people to do, but to demonstrate that it doesn't require much in the way of resources. That VM could probably host hundreds of people, since relays scale and de-dupe shared content.

If someone said that email or a personal website was inexpensive and could be hosted on a cheap VM, I don't think you'd make a comment like the above dismissing it as impractical: the idea is that if it's that cheap, then there are people that can host cheap/free (maybe ad supported) relays, similar to how cheap/free webmail exists today.