I just want to know if reports are failing for anyone else and if we need to be making some kind of plea to the admins/moderation teams/developers to try to fix it.
Because not being able to report this obvious spam is a problem.
Boost refuses to mark their messages as read for some reason
It does that with any unread message for some reason. Going to your replies and then tapping the double check mark in the top right marks everything in the inbox as read, as well.
I should probably let the devs know about this bug… When I don’t forget.
I’m assuming the issue is the new nicole has already been banned by the time you are responding to it, thus there’s nothing new to report.
Far as solutions, only ones I can imagine are I guess to allow instance owners to auto block and ban messages featuring certain words/sentences, and maybe auto delete messages from already banned users, in terms of technical solutions to getting ahead of the whack a mole game.
So you rewrite the script, and if the message gets banned it replaces the banned words with similar words from a thesaurus.
I assume the messages would get jumbled after a while.
“Hi, me name be Nicole. The fediverse baby chicken.”
Well yeah it’s still at best a cat and mouse game. Really I suppose the biggest red flag a system could detect is most likely sending out multiple DMs on an account with little other activity.
can’t really ballpark but I’m guessing the bot must hit several hundred if not thousands of us in rapid succession on creation (otherwise it would be a lot slower and only a few of us would have seen it).
That being said, auto detecting the links, contact information etc… would be pretty direct as well, as that would require every banned iteration to also remake whatever methods it tries to phish etc… as well.
Some admins have been developing and sharing solutions, from regex filters to going down into the databases itself.
For my small instance and personally, I haven’t seen any new messages in a while now, after adding this regex to the slur filter mechanism:
(.*(\n)*Hi.*[nN]icole.*[fF]ediverse [cC]hick(.|\n)*)
Sadly my brain is blanking which instance admins I should credit for that, but it was shared around somewhere.
Ha, I literally just got a new message with spaces between every letter to try to get around filters like this.
Makes sense, it’s most likely an ongoing arms race - I’m probably coasting under the radar a bit from running an instance with effectively only 3 people to target at the moment.
I love how useful Regex is. I hate writing Regex. Special shout out to anyone who can do it well.
Regex is great, but PCRE deserves a special place in hell. You don’t know unreadable until you’ve encountered regex that uses recursive matching, backrefererences, and subroutine calls.
💀
Thanks for the response and good suggestions.
When I looked at the profile, it didn’t say it was already banned, and I looked moments after I got it, but maybe it’s banned in the seconds between that check and when I submitted the report.
Sorry every one, Nicole made an account on my instance yesterday, since banned
May want to cross-post this to !lemmy_support@lemmy.ml.
But lemmy.ml is evil!