We also need shareable blacklists if they federate as well then that would be ideal. Surly someone can write a community thats a bot that gives democracy to blocklists.
I wish there was a seperate block as spam that would auto report. Some things I block due to lack of interest but others spam and our current report takes more clicks than I like. Then I would also like to be able to subscribe to peoples spam block list. If I like the cut of someones hib I might just subscribe to block any spam they block.
yeah I want as much power pushed down to the individual as possible. I was even thinking you just sorta have recommended mods. like by default you are subscribed to their block lists for the community but if you subscribe you have an option to see unmoderated or such.
Some instances may not want to use democracy to decide on what to block. But some web of trust would be nice perhaps. Like I trust an instance and if that instance blocks another instance, maybe I’d block it too.
I don’t think they are. The closest thing is hashtags but tags aren’t really a thing in ActivityPub as far as I’m aware. This kind of system will also always depend on people correctly tagging their stuff of course.
Seems like a user pattern that should be fairly detectable. Might not be a bad idea to scrape logs to look for spammy/copypaste comments, particularly on very new accounts, and either flag them for manual review or just ban them outright.
We also need shareable blacklists if they federate as well then that would be ideal. Surly someone can write a community thats a bot that gives democracy to blocklists.
I wish there was a seperate block as spam that would auto report. Some things I block due to lack of interest but others spam and our current report takes more clicks than I like. Then I would also like to be able to subscribe to peoples spam block list. If I like the cut of someones hib I might just subscribe to block any spam they block.
Exactly could have different lists made by different communities democraricly.
yeah I want as much power pushed down to the individual as possible. I was even thinking you just sorta have recommended mods. like by default you are subscribed to their block lists for the community but if you subscribe you have an option to see unmoderated or such.
Some instances may not want to use democracy to decide on what to block. But some web of trust would be nice perhaps. Like I trust an instance and if that instance blocks another instance, maybe I’d block it too.
I thinking more on a per user basis content tagging should defiantly have blacklists as well. But tags arent a thing yet?
I don’t think they are. The closest thing is hashtags but tags aren’t really a thing in ActivityPub as far as I’m aware. This kind of system will also always depend on people correctly tagging their stuff of course.
Somewhat already exists, called Fediseer
There are a few other instances that subscribe to lemmy.world’s censures list on there
Is that a content blacklist or a federation blacklist cos i left .world cos they kept degenerating with people.
Not necessarily a bad idea, but most the spam I’ve seen is from new accounts on larger instances, so I’m not sure it’ll help with this.
Seems like a user pattern that should be fairly detectable. Might not be a bad idea to scrape logs to look for spammy/copypaste comments, particularly on very new accounts, and either flag them for manual review or just ban them outright.
yeah usually less than an hour to a few hours old. occasionally they make it to days.