this post was submitted on 22 Mar 2025
279 points (99.6% liked)
Fediverse
32164 readers
798 users here now
A community to talk about the Fediverse and all it's related services using ActivityPub (Mastodon, Lemmy, KBin, etc).
If you wanted to get help with moderating your own community then head over to [email protected]!
Rules
- Posts must be on topic.
- Be respectful of others.
- Cite the sources used for graphs and other statistics.
- Follow the general Lemmy.world rules.
Learn more at these websites: Join The Fediverse Wiki, Fediverse.info, Wikipedia Page, The Federation Info (Stats), FediDB (Stats), Sub Rehab (Reddit Migration)
founded 2 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
I know that, in my mind the algorythm or maybe better to said the algorythm's parameters run on client side, no need to create a Fediverse instance on client side. We can already to something a bit like that with word filters.
I'm not sure you understand how a suggestion algorithm works, it's not something that can be run locally without having to then load and process thousands of messages. There is no aspect of that that could be run locally while simultaneously not having to load all the messages you're attempting to sort and make recommendations on.
It is not at all equivalent to a word filter, a word filter is not attempting to find something, it's a filter that applies to things that are already being loaded regardless. When a post comes in the word filter looks to see whether or not it has a word that needs to be filtered. An algorithm needs to find new content out of all the content available and then attempt to determine whether or not it's something that should recommend meaning the client would have to load thousands upon thousands of random posts from the Federalists in order to make that algorithm work