Not sure about the technicalities of it. Are the protocols similar enough for this?

Edit: not sure why this was double posted. I deleted the other one.

  • deejay4am@lemmy.world
    link
    fedilink
    arrow-up
    2
    ·
    1 year ago

    Nope, the instance is the server; same thing. Lemmy vs kbin are actually two different services entirely; but since they both run on ActivityPub (the underlying software that allows different sites to “federate” together), they can send and receive content from one another.

    Twitter-like replacement Mastadon is also part of the network of federated sites (and has a similar “choose-your-instance” federation) and the weirdest and most fun part is that you can see posts from any of them on the others! You can post to a Lemmy community from Mastadon, for example.

    Back to the login thing, your “home” instance is the one you signed up on (mine is lemmy.world for instance). By default, when browsing the Lemmy service, you will see local posts; that means anything posted to a Community that was created on that server. (Communities are like message boards, or subreddits).

    You can switch views to Subscribed; this will show you posts just from the communities (local OR remote - that is, communities from other Lemmy or kbin sites that have federated with your home instance to which you have subscribed).

    And finally, you can also view All; this shows you all local communities posts and also all posts from any remote community that someone from your home instance has ever subscribed to.

    You can kind of imagine it like instead of being one big Lemmy in a data center somewhere with offices and shareholders, there are 1000 little Lemmys; they all do their own thing as a monolithic version would, but then they all share all of that with each other. Users only interact with the service through their home instance, so that instance bears the load of their activity. Other instances might occasionally sync with each other, but that traffic is much less than users bouncing requests around all over the place constantly.

    This does come with some issues: for example, there is more than one “technology” community. Currently, searching for remote communities is not really great (I heard there are some external tools/sites to do it for now while they work on building more discoverability in). I guess the idea is to see which one eventually ends up on top; but this fragmenting also has the advantage that if the big popular one decides to close their API, de-federate and have an IPO, you’ll have other options. The hope is that’ll be incentive enough to make such actions undesirable, and the people who run the instances won’t be incentivized by profiteering rugpulls.