cross-posted from: https://hachyderm.io/users/maegul/statuses/111820598712013429

Is decentralised federated social media over engineered?

Can’t get this brain fart out of my head.

What would the simplest, FOSS, alternative look like and would it be worth it?

Quick thoughts:

* FOSS platforms intended to be big single servers, but dedicated to …
* Shared/Single Sign On
* Easy cross posting
* Enabling and building universal Multi-platform clients.
* Unlike email, supporting small servers

No duplication/federation/protocol required, just software.

#fediverse
@fediverse

  • @ryathal
    link
    English
    -15 months ago

    The fediverse isn’t over engineered, it’s just not quite focused on the right aspects. A federated social network needs to be more like a block chain, where the content is centralized, and the instances (miners) are decentralized. The content is the important part, and with everything being tied to an instance, it makes the content harder to access. You have instances defederating, going down, closing, and version conflicts, all that makes it harder for a network to gain traction.

    • @[email protected]
      link
      fedilink
      English
      5
      edit-2
      5 months ago

      You are describing a different thing than what the idea of the fediverse is. Content is collected at an instance and these instances federate. That’s why its called Fediverse: people basically form groups, these group federate. It’s a social thing, there is trust involved. With blockchain, the idea is that you don’t need to trust a central entity.

      I think you talk about something like nostr.

      • @ryathal
        link
        English
        15 months ago

        It doesn’t need to have the full trustless or buring energy for fun, but it does need to be resilient against instances going down, which currently isn’t the case.

        • @[email protected]
          link
          fedilink
          English
          25 months ago

          What specific edgecase do you have in mind. The fediverse is coping data and is quite resiliant against data loss.

          • @ryathal
            link
            English
            25 months ago

            If an instance shuts down everything from it is just gone. ML already ran into dns issues once, if it goes, 20% of lemmy is just gone.

    • @[email protected]
      link
      fedilink
      English
      15 months ago

      But who will moderate the content? Who is to say to what is legal and where? In USA and in EU different pieces of information can be shown. CSAM needs to be removed. Main lemmy devs removed only active mod on !anime[email protected] due differences in censorship.

      • @ryathal
        link
        English
        15 months ago

        The same people that do now?

        • @[email protected]
          link
          fedilink
          English
          15 months ago

          Blockchain is immutable so you can’t remove the content.

          The same people can’t moderate the content if content is centralised, there would need be an overlord that sets the rules.

          • @ryathal
            link
            English
            15 months ago

            Block chain was more on analogy than implementation. The key is that data isn’t bound to an instance, and ideally most people never need to know about instances.

    • @[email protected]OP
      link
      fedilink
      English
      05 months ago

      Sounds like you’re describing BlueSky there. Have you looked into it?

      Unless you’re talking about something more nostr/web3.0?

      • @[email protected]
        link
        fedilink
        English
        25 months ago

        bluesky depends on one single entity. they promised a lot about their protocol, but they have yet to show that other instances other than the official one can operate in a fully independent manner.

        • @[email protected]OP
          link
          fedilink
          English
          15 months ago

          I was under the impression that it’s clear that additional relays can work within their system? Have they not setup anything in the protocol for how that’d work?