(this is a bit of a rant, i’m sorry)

what in particular do you mean by lack of discoverability?

like, i want to see posts from communities that i already subscribed to, but because there’s more than 1000 communities on the fediverse and i’m only subscribed to a small countable subset of them, i inevitably lose out on a lot of content. (The “all” feed sucks unfortunately). So how to solve this?

The lack of discoverability is non-starter for many.

The Fediverse significantly lacks behind on the Content Discoverability technology.

I guess this is because there was a loud public outcry in the last 20 years that whoever makes your feed (this is called an “recommendation algorithm” or abbreviated “the algorithm”) has a lot of political power to decide what you see and what you don’t see, and that’s frowned upon. Because everybody that has power over what you see and what you don’t see is bad. That is why nobody wanted to provide an recommendation algorithm for the fediverse, because they would expose themselves to wild accusations. There should be an open-source recommendation algorithm, though; I’m sure of it.

  • PumpkinDrama@reddthat.comEnglish
    21·
    14 hours ago

    You can’t have a recommendation algorithm on open-source software because it requires a lot of compute to calculate personalized recommendations for each user, which simply isn’t feasible for most instances. Instead, there should be an API endpoint that returns post metadata for the last week, allowing users to implement their own ranking algorithm via a userscript running on their own hardware.

    I also believe there should be a more personalized “All” feed per instance. Each instance could surface different content tuned to the admins or to a subset of long-term users—something stable that doesn’t change often but varies from server to server.

  • squirrel@piefed.kobel.fyiEnglish
    4·
    1 day ago

    Mastodon has an algorithm in their official app. The Explore feed shows you posts that are popular among people you follow and interact with.

    Loops has a For You Page algorithm.

    There are algorithms on the fediverse. They are rare, but they do exist.

    • Coelacanth@feddit.nuEnglish
      1·
      1 day ago

      Is the Mastodon algorithm new? I could have sworn they used to pride themselves on the chronological-only feed.

      • squirrel@piefed.kobel.fyiEnglish
        3·
        22 hours ago

        It’s been there for a couple of years. Found a GitHub comment by Gargron talking about what it is in 2022

        The explore tab is in essence a moderated, quality-filtered federated timeline. Its purpose is to help you discover other people and expand your visibility, but without being a vector for spam and abuse.

  • FoundFootFootage78@lemmy.mlEnglish
    4·
    10 hours ago

    The problem with recommendation algorithms isn’t just the power, it’s the fact that it deprives us of a shared reality. It’s one thing if we filter ourselves into a bubble but it’s another if the site itself does it.

    • unwarlikeExtortion@lemmy.mlEnglish
      1·
      2 hours ago

      An algorithm is a tool. And if xour tool is a hammer and nlthing else, then yes, everyone will be one-track minded in an echochamber.

      Bit the solution isn’t to go throw hands up in defeat and get rid of the hammer because it proved dangerous, but to make other tools:

      Make the algorithm into a toolbox.

      Because, from what I see just from browsing Lemmy, Lemmy has a few elementary sorts (Top, Active, Most Comments) and two filters: a hard Local/All toggle and the Top sort has filters for Day/Month/Year/All time. I know there’s a few I missed, but my point stands.

      The algorithm should have toggles to shape how it works.

      Instead of a Local/All toggle, add a map of connected communities and the ability to “jump” between their feeds - without needing to switch to an acvount on that instance.

      Add a “randomize” option: not to the feed, but to the algorithm’s parameters. Don’t just shuffle the “selected” posts like a playlist. Fuzzy the criteria for selection themselves up to get a different batch altogether!

      Learn from the “big” players: use their core tactics and ideas, but with one huge caveat: make the parameters transparent and accessible. Make “presets” (which we already have).

      In essence, don’t make a recommendation algorithm, make recommendstion algorithms. Ones easily accessible to users and easy to understand.

      Or, in other words, instead of a crutch, make a toolbox.

      Actually, the algorithm is less a hammer than a crutch. It’s not something you use to build, it’s something you use to walk.

      • FoundFootFootage78@lemmy.mlEnglish
        1·
        10 hours ago

        That’s not filtering ourselves, that’s letting ourselves be filtered. If an algorithm does the mental work of filtering us into bubbles, that makes it harder to escape.

  • Ephera@lemmy.mlEnglish
    2·
    1 day ago

    There should be an open-source recommendation algorithm, though; I’m sure of it.

    Problem is that the kind of algorithm you envision is technologically a black-box, not just by choice. It’s a machine learning model. At best, you could make the training data and instructions public, but it would still be hard to reason why it makes certain decisions. Corporations traditionally try to eliminate biases by throwing as much data at it as possible, but that makes it even harder to reason about it.

    I guess, maybe you could try to split the tasks. So, set up a list of e.g. 50 topics, such as sports, IT, politics etc… Then use a small language model to decide into which categories each post fits. And then you could let the user decide the weights for the topics + weights for recency and vote count.
    Or I guess, automatically decide the weights based on what the user upvotes and then make the weights transparent to each user.

    But yeah, I don’t think there’s prior art in this respect, so would probably need lots of experimenting still.

    • gandalf_der_12te@discuss.tchncs.deOPEnglish
      1·
      1 day ago

      hmm, i think you’re overthinking this. what if the recommendation algorithm simply gives you stuff from communities and you’ve subscribed to and “similar” communities (these would have to be linked from the original communities / link to the original communities)?

      that should be reasonably easy and not involve any neural networks. i think basically it constructs a “feed” (post list) which is basically a remix of other lists (which are the individual communities that stuff is taken from), maybe weighted with a certain scalar factor.

  • Die4Ever@retrolemmy.comEnglish
    4·
    1 day ago

    i want to see posts from communities that i already subscribed to, but because there’s more than 1000 communities on the fediverse and i’m only subscribed to a small countable subset of them, i inevitably lose out on a lot of content. (The “all” feed sucks unfortunately). So how to solve this?

    You don’t use the Subscribed feed? I like Subscribed+Scaled

    The Fediverse significantly lacks behind on the Content Discoverability technology.

    https://quiblr.com/

    https://quiblr.com/understanding_your_private_personalized_feed