• 0 Posts
  • 30 Comments
Joined 1 year ago
cake
Cake day: July 11th, 2023

help-circle

  • PlantJam@lemmy.worldtoTechnology@lemmy.world*Permanently Deleted*
    link
    fedilink
    English
    arrow-up
    4
    arrow-down
    4
    ·
    6 days ago

    Isn’t paraphrasing/summarizing the top result a pretty good use case for LLMs? If I search “what temperature should I bake cupcakes at?” I really just want a simple answer, not dozens of links to life story style recipe blogs.

    DDG didn’t provide a summary, but Google did (and it was very long). I assumed the answer was 350F, but the summary suggested 325-375. Lower for flatter cupcakes, higher for more domed. Interesting.

    This type of summary wouldn’t be nearly as helpful for a technical programming question, but I doubt that describes the bulk of search queries.














  • I recommend some keyword filters. I use them to reduce certain topics from my feed, specifically about certain wars or politicians or celebrities. I did the same thing on reddit to filter out some of the “awareness” campaigns where everyone posted about the FCC chair to every single sub.

    I also just browse local communities instead of the entire fediverse. This defeats the purpose of the fediverse, but it drastically reduces the number of duplicate and NSFW communities. I then mute the communities I don’t want to see. I’ll also mute the hyper specific communities that usually have complete overlap with the more general community, i.e. dogs and dogpictures.

    It takes some work to get your feed to your liking, but it’s worth it in the end. There’s still far less content available than on reddit, but the amount of quality content feels similar.