One of the best things about reddit was looking for answers or other users with the same problem as you, and since Google didn’t really help with that anymore and instead insisted on giving you business results, the best practice was to put your search terms in followed by ‘reddit’ and you’d find your answer.

  • marsara9@lemmy.world
    link
    fedilink
    arrow-up
    22
    ·
    edit-2
    1 year ago

    I’m working on a specialized search engine just for the fediverse. https://github.com/marsara9/lemmy-search

    If anyone wants to help out, feel free to reach out, but I hope to have something ready to release soon.

    The idea with my version is that it’ll search as much of Lemmy / the fediverse as it can and you can select the preferred instance that you want to open any link with.

    • qisope@lemmy.world
      link
      fedilink
      arrow-up
      6
      ·
      edit-2
      1 year ago

      If you are looking to return relevant, well ranked results based on freeform queries you’d be better indexing into something like elasticsearch. Otherwise you’ll be reinventing solutions to well understood problems, like stemming as a very basic example.

      • marsara9@lemmy.world
        link
        fedilink
        arrow-up
        3
        ·
        1 year ago

        For the initial release the search is still fairly basic, but A LOT better than the built in search here.

        Right now I just look for IF the individual words match ANY of the words in the post title or body and then rank based on the number of upvotes that the post has.

        Future versions may look at using elastic search, etc… But for MVP it just looks for the number of hits + the score of the post as I assume the higher the score the more trustworthy the post, and obviously the more matches that to your query the more relevant the post is.

    • Excel@lemmy.megumin.org
      link
      fedilink
      English
      arrow-up
      2
      ·
      1 year ago

      How is this different from just searching for posts on the original “seed instance”? Presumably you’re crawling through everything on all of the instances that it’s aware of, as opposed to the Lemmy built-in search which would only search communities that have a subscriber?