As atproto matures, adversarial actors will reveal all the places where trust is implicit rather than managed - maybe time to build reputation mechanisms for PDSs like we do for email servers? It'll likely need to be default-block/prove-healthy, or default throttle until proven typical?
Scoiattolo
They've got a huge network that I'm in the process of graphing. They're now setting up their own PDSes. It's sophisticated.
Mar 17, 2026 at 5:44 AM
Trust Infrastructure on ATproto
I agree with Nick that you have noted a big hard problem, with Zooko’s triangle type trade-offs. (For a semi-related example that is “easier” but still crazy difficult, see our “PeerFlow: Secure Load Balancing in Tor” https://doi.org/10.1515/popets-2017-0017 ) As noted or implied already in this thread: in the end, if someone can cheaply spin up indefinite unlinked identities and build good reputations for them, and if trust is based on such reputations, it is hard (impossible?) to avoid them s...
Scott Lanoue - CoMail
Ricardo J. Méndez (@ricardo.bsky.social)
If WSocial sold people on the idea that everyone they would interact with is ID-verified, then indeed, every single open ATProto community would act as a back door of sorts. The only way for them to deliver would be to de-federate entirely. And it's always a bad idea to bet against openness.
ATProto ideas
If AT Protocol becomes “a meaningful layer of the web” — even if it’s only tens of millions of users — then the winning ideas probably won’t only be futuristic...
Adversarial ATProto PDS Migration | Blog
An ATProto account lives on a Personal Data Server (PDS), and that service needs to be hosted somewhere by someone. There are a lot of options here, like a Raspberry Pi in your closet, a rented VPS in a datacentre, or a batteries-included service like bsky.social.
Attack Resistant Trust Metric Metadata HOWTO
3 Jul 2002: Advogato is now running an implementation of the generic metadata engine described below. The code is available at the mod_virgule CVS, hosted on casper.
Adversarial PLC directory migration - microcosm
Contingency planning: what's our credible failover for the centralized underpinning of ATProto identities?
Adversarial PLC directory migration - microcosm
Contingency planning: what's our credible failover for the centralized underpinning of ATProto identities?
Jesse Ruderman » The Advogato trust metric is not attack-resistant
The project for the graduate algorithms class I took in Fall was to compare two algorithms for a problem of our choice. I chose to compare two trust metrics, the Advogato trust metric and Google's PageRank. Trust metrics attempt to solve the following problem: "Given a directed graph where an edge from A to B means 'A trusts B', whom should I trust?".
As atproto matures, adversarial actors will reveal all the places where trust is implicit rather than managed - maybe time to build reputation mechanisms for PDSs like we do for email servers? It'll likely need to be default-block/prove-healthy, or default throttle until proven typical?
Scoiattolo
They've got a huge network that I'm in the process of graphing. They're now setting up their own PDSes. It's sophisticated.
atproto offers an fascinating example of what it means to restructure this relationship. End-users get to choose who and how much trust they delegate to software producers, the delegation of trust becomes tangible and revokable.
𝕮
Networked computing, in a fundamental way, changed the producer-consumer relationship we have with corporations. In the material world that relationship ends with the transfer of material and money; I buy a pencil and some paper, and the producer is out of the picture while I use them to write.
If WSocial sold people on the idea that everyone they would interact with is ID-verified, then indeed, every single open ATProto community would act as a back door of sorts. The only way for them to deliver would be to de-federate entirely. And it's always a bad idea to bet against openness.
Wimster9030 🇪🇺 🇧🇪
I know, Daniel, but that's not the way WSocial "sold" it to us. We could only access trough ID-verification so we know for 100% they are EU and a "real person". That was the whole point. So we went trough that process, but now it seems that the backdoor are the other ATProt platforms to get in too.
Some of us in the ActivityPub world feel jealous of AtProto’s success (like me). But more decentralization is a win for everyone. With Bridgy from @anew.social connecting protocols, this isn’t competition, it’s momentum. We’re stronger together, so let’s build the social web humanity deserves 🚀
