The thing is, I've actually been doing that for years now. I haven't been opening the videos, nor have I downvoted them, just clicked the side icon and "don't recommend the channel". I could technically click "not interested" but I wouldn't want Youtube to stop giving me music I do want to hear.

So it wasn't just me ! I keep telling Youtube to stop pushing that stuff but it keeps somehow popping up on recommendations from completely unrelated videos, I swear...

I'll keep the current list up for a bit longer but I'll remove it afterwards. Truth is, having a 3000+ user list simply existing on my account does make certain bits of the Bluesky interface considerably slower, and I'd rather keep that isolated somewhere else as a result.

I'll link to mine here as well. bsky.app/profile/kana... It started as a simple process to weed out a few trolls but turned into a massive undertaking to add all of the accounts spamming the same stuff over and over, all verified manually. I don't know for sure how many are in there at this point.

Do you have some kind of tool or method to make the process faster ? I realize on my own list that the longer it gets, the longer it takes for the moderation buttons to even appear in the first place. Even with some automation, it becomes a hassle from my experience.

Sometimes they really do spend time writing posts/replies that aren't just copy-pasted, but then their username is just a slur with some numbers (not sure why Bluesky even allows it) and bio is something like "here to own the libs". Some real whiplash when witnessing their behavior.

Je dirais que la modération fait ce qu'elle peut mais est clairement débordée (bsky.app/profile/safe...). Je fais ce que je peux de mon côté pour maintenir une liste de modération mais c'est pas génial non plus ; plus la liste est longue, plus l'interface de blocage met de temps à apparaître.

Here's my own, mostly focused on obvious trolls going for that one spam phrase and similar ones. I check each account by hand and then use an automated script to add them to the list in bulk, making the process a bit less painful. I wish I knew how many are in there by now. bsky.app/profile/did:...

@bsky.app Currently, I think that user moderation tools are ill-prepared for targeted attacks of this scale. Too many clicks, too much waiting for buttons to appear, some bugs. I built an automated tool to help with the process, but it regularly runs into snags like failing to find users that exist.

I have been at work building up a massive block list for as many of them as I could find, but they just keep coming. I try to keep it updated but catching them while not misfiring is an ordeal. Quite a few people use the same words as them, in order to mock or disprove them. bsky.app/profile/did:...

I have been trying this, the problem is that the waiting time is very inconsistent... sometimes almost instantaneous, sometimes more than 2 seconds. I'm looking into opening a new tab into the moderation list and add the account through that interface to see if it's more consistent.

I've been looking into this myself by learning how to use the Automa browser extension, but a snag I've been running into is that the button to add an account to a block list can take something like 2 seconds to appear, while it checks whether the account already is in the list or not.

Oh, I see they've been doing the same I've been doing. I've been feeling my own block list using the same manual methods, but I would really love for there to be some kind of mass blocking tool as the process of opening each individual profile is tedious.

Sadly I don't, but I sure hope there is some kind of solution because trying to block all of them manually is seemingly an impossible task at the moment.

Forget about "the other day", search right now for "two genders" and look at the latest posts, there is an influx of new accounts over the last hour spamming messages like that and they just keep coming.

Personally, I've been getting some mileage out of "libt*rd" (a or u), or "left*rd". That being said, if you're not a fan of moderation lists, maybe you'd prefer labelers ?