Mutual aid spam is becoming a problem on the Fediverse.
And to be sure, I'm not against mutual aid. What I am against is spam.
This person has not verified who she is -- or even if the profile picture is hers. Additional research on her name states she is a scammer with a record of grifting. I am therefore skeptical that any donations will help anyone in need.
Folks, please be cautious with mutual aid requests. Yes, people sometimes need help. But people also lie.
That's not mutual aid, that's scam spam. Report it.
One problem with reporting private messages on Lermy is, as an admin i don't see who sent the message. I only see who reported it. And i don't have any actlon available, other than marking the report as handled.
with reported posts, i can ban the poster. With reported messages i'd have to ask the reporter who it was, trust their answer, search for the account manually and then i could ban. Not really efficient or fast if there ever was a spam wave.
of course sparmers could then just register a new account on a open instance and i might need to defederates which would lead to a fractured landscape of spammy open instances and likely inactive private instances.
there's also not even rudimantary spam filtering in lemmy.
The main saving grace is that Lemmy is too small to attract a ton of spam yet.
maybe some of the above is just due my pick of clients (jerboa and the web interface), and there's better tools? If so, i'd love to hear. But as things stand right now, there's a lot to be desired
I wish I had approximately double the hours in a given day, and also vastly more coding skill to help in meaningful ways.
It seems sort of odd that comments or messages reported for spam don’t offer any tools. Even a simple url pattern match that gives mods/admins the ability to click a checkbox to remember the link and take some predefined action in the future would be a rudimentary but effective option.
I mean, heck, it’s the fediverse. In my fantasy implementation of an anti-spam approach, it would be possible to federate these lists of untrusted links and assign consensus-based confidence scores for links generated from moderator actions across instances. (With options for instance admins to tailor their own trust scores of other instances, so that each instance can choose for themselves who they trust, just in case a couple rogue instance admins try to poison the spam filter.)
Same concept can be applied to banned accounts, although in that circumstance, I’d suggest they find a way to mask the email address when sharing it. Not that folks won’t just spin up a new email. But, you know. Something is better than nothing.
Hopefully that makes sense. I’m losing my mind with sleep deprivation.