Element has chosen to pursue future development of Synapse, Dendrite and associated server-side projects under the terms of AGPLv3.

Synapse and Dendrite relicensed to AGPLv3

  • @[email protected]English
    6•1 year ago

    I really, really hope this leads to development of data portability/server migration options. When I set my homeserver up, I chose Synapse as I didn’t know about the other servers. Now that I do, and would like to switch away because of Synapse’s performance problems and the new CLA stuff, I realize I and all my users are fully locked in, and would have to start from scratch (lose all chats, profiles, etc) to migrate.

    • @[email protected]English
      1•1 year ago

      Just give XMPP a shot… It’s what Matrix has been promising to be without ever actually delivering, was there a decade prior and still will be after the next.

      Matrix’s purpose was to be a VC unicorn for is founders, it’s not tackling a new problem, it’s not bringing a novel or interesting solution, and everything that it does differently on a technical level seems to go against its goals and those of its users. Time to drop them was someandsome ago, but it’s never too late.

      • @[email protected]English
        4•1 year ago

        Is there a good way to handle e2e encryption on xmpp? I feel like that is one of the selling points of matrix. Iirc there was an OTR plugin for gaim/pidgin, but I have no idea what’s normal these days

        • @[email protected]English
          1•1 year ago

          XMPP’s E2EE is comparable to everything else out there (Signal, Matrix, WhatsApp, …) in that it uses Signal’s double-ratchet algorithm (guaranteeing perfect forward secrecy and al.). All maintained clients support it: https://omemo.top/

      • @[email protected]English
        1•1 year ago

        My hope for XMPP is that new and better clients emerge than the ones that already exist. Matrix has this same problem as well.

        • @[email protected]English
          1•1 year ago

          Of course I wouldn’t complain if more and more clients popped up, but IMO what matters is that the ones we currently have remain well maintained over time, in the hands of people who know what they are doing, and that there is one such client per platform, which is effectively the case today.

          If you have ideas on how to improve current clients, I’m sure people would take your remarks and contributions kindly (for instance, there’s a zoo of pretty clients that forked from Conversations on Android, Dino tries to be a minimalistic client for the simple needs and is getting an official Windows build soon, etc)