Replies (1)

• @teajaygrey@snac.bsd.cafe 2026-02-21 17:17

  IMHO, SILC (Secure Internet Live Conferencing, previously: https://silcnet.org code archived here: https://github.com/silc/silc.github.io) solved a lot of IRC's shortcomings, decades ago and also, not coincidentally: implemented end to end encryption. Alas, it probably had an apogee of around 300 daily active users circa 2005 (with a peak of several thousand when it was used during a Chaos Communications Congress) and despite having multiple clients (Colloquy was particularly neat for its era) never seemed to make much of a dent outside of hacker realms. In particular: messages were encrypted from server operators, which is a threat model too few seem to acknowledge is real. Perhaps, now that whistleblowers have outed WhatsApp/Meta as sharing users' presumed private messages en masse to internal teams in litigation? That will now be acknowledged as an actual threat that Pekka foresaw and attempted to mitigate decades ago? There's also PSYC2 (https://about.psyc.eu/) which I think was at least attempting to meaningfully advance the state of the art of end to end encrypted real time chat, and perhaps of particular interest, seems to be attempting to address some of the perceived deficiencies of federated models: https://about.psyc.eu/Federation No. Neither of those are as popular as things such as XMPP or Matrix, but both were designed for secure comms, by folks who knew what they were doing. For example, PFS (Perfect Forward Secrecy) is antipode to "message persistence" and many (e.g. Matrix, Session, etc.) seem to have missed the memo, thus putting their users at risk. Whereas SILC, defaults to Perfect Forward Secrecy, sacrificing that "convenience" for security. I'm also really curious how "moderation" would be manifest in Spritely's efforts, since, more often than not, particularly in 2026: moderation is synonymous with censorship. CC: @suika@pl.ibuki.club @3@social.xenofem.me

  Open ##2551417