People describe ATProto as federated, and that has always bugged me.
-
@lrhodes @jdp23 @rwg @jenniferplusplus
on a side note, i can definitely recommend this new blog that came with last weeks update to atproto
ATProto for distributed systems engineers - AT Protocol
AT Protocol is the tech developed at Bluesky for open social networking. In this article we're going to explore AT Proto from the perspective of distributed backend engineering.
AT Protocol (atproto.com)
-
Thinking about this again after Mastodon's announcement of Fediverse Discovery Providers today. https://www.fediscovery.org/
These seem to play a similar role to AT-style Relays. If they want to provide global search (which is at this point table stakes for a big world Twitter alternative), it's not clear what other scalable options there are. If the implementation is as efficient as AT's the costs of a running a whole-network service provider are presumably likely to be roughly comparable to the costs of a similarly-sized Relay, so very expensive for a whole-network provider.
Of course FDPs are optional. But once they're available, instances using these services will be a lot more more attractive to people who want a Twitter alternative that instances that don't provide global search. So they're not really optional for that use case (just as AT Relays aren't really optional for something Bluesky-scale). Networked communities that only care about search and hashtags of a subset of the network probably don't need them, or if they do the scope of the discovery provider will be smaller so it hopefully won't be prohibitive (like AT Relays scoped to a subset of their network).
And as the announcement mentions, they'll support other kinds of providers. Providers that can label posts, accounts, and instances with information useful to moderators and moderation tools? Providers that give instances and individuals access to customized feeds? The possibilities are endless!
So on the one hand, you can build the AT architecture on top of today's AP (ActivityPods are PDS-like). And as Robin Berjon has pointed out, you could do AP on top of AT. At least today they clearly optimize for different things: AT for scalability and all-public networks, AP for flexibility (which comes with a complexity cost) and scoped visibility ...
My intuition is that today's AP is probably better for connecting disparate systems (Wordpress, Threads, Flipboard, Mastodon, Lemmy) that aren't necessarily built with AP internally. Today's AT is better for building large all-public somewhat decentralized (although not in the same way as today's instance-oriented fediverse) social networks that an then be connected with AP (or whatever).
@[email protected] @[email protected] @[email protected] -
@jdp23 @lrhodes @laurenshof
I expect Mastodon's search things will be much smaller and less revolutionary than you're imagining.Also, fwiw, I think ATP is better both at interop between disparate systems and all-public publishing/sharing. And I think it's more flexible in practice. AP's most significant benefit is scoped visibility, as you call it. And it's not even good at that, it just doesn't prohibit trying.
-
Hard to know on how much impact this will have. If it's only Mastodon I agree with you, but also it depends to at least some extent on what Threads does. If they're serious about the fediverse -- or just want to pretend they are for PR/regulatory reasons -- I could certainly imagine them having a discovery provider (just as they've already talked about providing moderation tools). Or I could see Flipboard doing it, they seem to be betting on the fediverse pretty big (and I've been impressed with Mike McCue's strategic thinking). Time will tell.
And maybe AT is better at interop, I was just noticing that's where AP has traction now. And agreed that AP isn't particularly good at scoped visibility ... hmmm ...
@[email protected] @[email protected] @[email protected] -
@jdp23 @jenniferplusplus @laurenshof Seems to me there are some pretty fundamental differences between what they're describing and what AT relays do. You could (with limitations) build the AP equivalent of an AT relay, but only by violating a lot of the norms and expectations of federated networks, and the fedidiscovery project have stressed that they don't intend to do that. Maybe it could add features like moderation labeling, but not without huge redundancies in the system. But who knows.
-
Agreed that the fediversity discovery proposal is stressing that they're not violating fediverse norms and expectations -- and yes, there's certainly a fundamental difference between AT's all-public approach and AP's scoped visibility and the fediverse's opt-in to search (well a lot of the fediverse, Lemmy as far as I know doesn't even give the ability to opt-OUT of search, not sure about Misskey, etc). That's a different difference than how much things are mediated though.
In terms of other uses for this architecture I was just at the fediforum session on this, and people at the session talked about using it for moderation, spam-fighting, trending topics, and link previews ... so, it's clearly designed for broad use and that's how people are thinking about it, but of course we'll see how it plays out.
@[email protected] @[email protected] @[email protected] -
@jdp23 @jenniferplusplus @laurenshof I'm sure we COULD incorporate some or all of that into search. It would just require more overhead than in an ATProto, because Fediverse servers pass messages directly, whereas nearly all traffic on an AT network flows through the relay before it reaches the PDS. So for a Bluesky-style labeling scheme, you've have to get the label and a reference URI from the search server in addition to getting the post from its originating server, then match the two.
-
@jdp23 @jenniferplusplus @laurenshof That, in turn, creates a situation where accounts can avoid labeling by opting out of search, effectively using privacy and safety features to bypass the filters those labels facilitate. So sure, the technical capacity is there, but if people are envisioning this as a way to adopt some of the Bluesky features they like, they really need to think hard about the form the differences are likely to take.
-
It's not incorporating all that into search, it's separate providers for the different functionality (just like AT has separate labelers and feed generators). From what I understand, the current plan is for the instance to enable sending data to one or more service providers, which is different from the centrality of the Bluesky Relay ... but then again depending on how it's implemented (and the licensing restriction) I could also imagine other service providers taking the output of the Mastodon gGmbH search provider as a starting point.
@[email protected] @[email protected] @[email protected] -
I'm not sure people are thinking of it specifically as a way to adopt Bluesky features, I think it's more that they're thinking of it as addressing (real) problems and coming up with approaches as Bluesky.
Agreed that using this approach for moderation and spam conflicts with opt-in. It'll be interesting to hear what their thougths are on that. (This is a challenge for consent-based privacy laws too!).
Still, even before this, I've seen proposals for using Akismet where nobody's pushed back on consent. And instances that use CSAM scanners already ignore consent (to be clear I'm not saying this is a bad thing, just observing. that it's the case). So it's not clear to me how either norms or the technology will evolve on this front.
@[email protected] @[email protected] @[email protected]