The image is a reddit post with the following text (automatically transcribed):
I remember I got into an argument on reddit awhile ago with a person over Italian food. It got to the point they were following me into other subs to harass me. I clicked on their profile to block them and their most recent post was them drinking their own piss on r/piss. At that moment I realized I had spent so much pointless time arguing about the taste of food with someone who drinks their own piss as a hobby. This site is a shit hole.
The total score is appearently stored on your user somewhere in the database but not displayed… Someone said. They lied?
I would assume it’s a hard thing to implement reliably in a system such a lemmy. You could just spin up your own instance and give yourself a ton of points if you really cared for the numbers.
That’s true… I wonder how it handles downvotes and upvotes on comments or posts, they are also replicated right? What if some instance goes into the db and changes those numbers?
Then it’s only changed on their instance. They’d need to create fake users and send votes for them. At that point others should be able to detect it as botting, at least if the impact is big enough.
I use Summit for my reader and it keeps a total post/comment score. Seems extremely inaccurate though
You can always go through someone’s comments and add up the scores
Sounds like a fun job… :)
You can do local counts as the instance owner can see that but afaik the who voted doesn’t federate. Might be wrong I need to go back through the docs.
I read that you can see who voted on kbin, so it should be federating for that to work
Kbin is different to lemmy but activitypub only shares certain data. Looking at the tech info the primary server only shares the count to the federation but would know who directly liked it. https://www.w3.org/TR/activitypub/#liked
Again, without testing on my own server it is hard for me to say with 100% confidence.