I apologize if this has been asked before, but I’m wondering if it would be feasible to implement a new approach to defederation that offers the option of choosing between complete or partial defederation from another instance.

Currently, defederation blocks both the locally made posts on the defederated instance and its entire userbase. This can be excessive, and in many cases it may be better to block only the posts made on the other instance while still allowing its users to interact with the instance that defederated — user behavior may differ between their home instance and other instances. This partial defederation (or limited federation) would facilitate normal interaction without negatively affecting the content of a feed.

Problematic users could be managed on a case-by-case basis using bans, similar to how it is done for federated instances. Automated tools could simplify this process in the future. Complete defederation would still be necessary in extreme cases where no positive user interactions are expected, such as with instances that promote Nazism.

Instances are being forced to choose between a sledgehammer and nothing at all, and I think a compromise is warranted. I’m curious to read others’ thoughts on how to solve this existing challenge.

EDIT: I added a rough sketch that outlines the proposal. On the left side is the system as it works now and on the right side are two possible scenarios for limited federation (1 direction or bidirectional)

  • theory
    link
    fedilink
    English
    arrow-up
    10
    arrow-down
    2
    ·
    1 year ago

    This needs to happen. This is getting ridiculous

    • OrangeSlice@lemmy.mlM
      link
      fedilink
      English
      arrow-up
      9
      ·
      1 year ago

      I have a feeling that given a couple of years, things will settle out a bit and be more like Mastodon.

      Could you imagine if your ISP/Gmail was so particular about what servers you could send email to?

      There will always be valid reasons to defederate, although I think the bar for that is going to end up pretty high and well-defined in the future, but it’s sort of an organic process to get there.

      • zero_iq@lemm.ee
        link
        fedilink
        English
        arrow-up
        4
        ·
        1 year ago

        When you look at Lemmy as a whole though, the growth is significant, but the total is not that huge.

        The number of Lemmy users has increased from ~50K at the start of the month to ~135K today, so a bit under 3x. (For comparison, that’s approx. 0.002% of reddit’s active daily user accounts, or 0.00008% of reddit’s active monthly user accounts.)

        That we are seeing technical, trust, financial, and social/management scaling problems leading to defederation, servers being overloaded, etc. at this relatively tiny level of engagement is a bit worrying, but also kind of encouraging in a way. Better to encounter these things and address them early on, while the system is up and running.

        The good news is that there seems to be no shortage of people willing to help out. Lemmy is working for now, but these rumblings of future scaling problems need to be tackled. We have a growing user base, and there seems to be no shortage of motivation for creating a viable reddit alternative.

        • psyspoop@kbin.social
          link
          fedilink
          arrow-up
          2
          ·
          1 year ago

          Just a correction, 135,000 is 0.2% of 52,000,000, not .002%. If 135,000 users was .002% of Reddit’s daily active users, that would mean Reddit would have over 6 billion daily active users.