I want to make signing up to my instance as unattractive to bots and vote manipulation as possible. If I could turn off all voting on Lemmy I absolutely would. I think activitypub votes are spam and just because reddit used them doesn't mean that it should be required. I used forums and other websites for a long time before Reddit even existed without voting and my time with Reddit and now Lemmy has left me unconvinced that it is worth while. I think engagement with comments is the only thing worth anything on this platform and some day I'd like to see a sorting algorithm that only took traffic and comments into account for scoring.
Anyway, I try my best to review the users who sign up and the comments that are made to my instance. I typically deny new applicants with disposable email addresses as well.
But there are some accounts that were created that have no activity that I can see easily from lemmy-ui and I'm not really at the point where I am reviewing SQL database. I want to confirm disabling downvotes at least prevented that kind of bot activity and I was wondering if there is anything else I can be doing? I can see how upvoting fascist and western centric points of view can also be problematic by boosting a state actor's propaganda in a mirror of downvotes silencing minority views.
Disabling downvotes is a feature of Lemmy, I have it disabled on vegantheoryclub.org from the admin preferences. I don't know though if that just disables them on Lemmy-ui and compliant apps or if it actually disables them completely. What I am concerned of is someone creating an account, not using a reputable app, and sending API calls directly to my instance to federate to others.
You can have a look at reddthat.com or Lemmy.blahaj.zone communities from other instances and see if the downvotes count are consistent
I would prefer a way to detect the manipulators from my SQL database then doing detective work to be honest. Someone else mentioned that they can actually do this so I will have to research this.