- cross-posted to:
- privacy@lemmy.ml
- cross-posted to:
- privacy@lemmy.ml
Interesting counterpoint to the stuff we sometimes talk about here. It’s more for public chat rooms though. MLS (RFC 9240) still interests me and I’ve been wanting to try coding it.
Different models for different cases, right? It’s only a counterpoint if one believes that the message is “federate everything.”
There is no privacy guarantee in federated systems. There’s no privacy in public forums, full stop; nor should there be any expectation of any. You can barely achieve any privacy trust in 1:1 communications, and much less any form of many-to-many.
Anonymity is another matter - you can have anonymity, if you’re careful, but privacy? No. Federation is not about providing privacy, or even anonymity - there’s no section either in the AP specification.
For private communications, you use different technologies; P2P is a good option, but you can get privacy and anonymity in federated systems like Nostr, which - while capable of being a public forum - has all the parts needed to achieve anonymity and privacy.
Tox has a good model which they can’t seem to get to work. 0xchat is interesting.
It gets said repeatedly: you tailer your solution to your threat model; there’s one size fits all.