Though Lemmy and Mastodon are public sites, and their structures are open-source I guess? (I’m not a programmer/coder), can they really dodge the ability of AI s to collect/track any data everytime they search everywhere on Internet?
Though Lemmy and Mastodon are public sites, and their structures are open-source I guess? (I’m not a programmer/coder), can they really dodge the ability of AI s to collect/track any data everytime they search everywhere on Internet?
Voting is done through ActivityPub because that’s the only reliable way to do it. If you don’t even require account names, sending a million downvotes or upvotes to a post becomes trivial. ActivityPub votes are signed with account keys so the amount of spam votes is restricted somewhat.
Lemmy 0.19 added a nifty feature to the web UI that allows server admins to see who voted what on comments. Previously, it was possible to extract that data from the database, but now any admin can just click the menu button on a comment and click “votes” for an overview.
This makes a lot of sense for manually verifying things like voting rings or butthurt people who will go through someone’s profile and downvote every comment.
As for reading: Lemmy maintains a read state for posts so its “hide read” feature can do its job. It doesn’t store this information about comments, and for notifications you’ll have to interact with them or manually dismiss them for the read state to get updated in the database.