/kbin meta

3 readers

12 users here now

Magazine dedicated to discussions about the kbin itself. Provide feedback, ask questions, suggest improvements, and engage in conversations related to the platform organization, policies, features, and community dynamics. ---- * Roadmap 2023 * m/kbinDevlog * m/kbinDesign

founded 1 year ago

Spam from unmoderated communities/magazines (feddit.ch)

submitted 1 year ago by theKalash@feddit.ch to c/kbinMeta@kbin.social

19 comments fedilink hide all child comments

Both !infosec@kbin.social and !science@kbin.social are being spammed constantly with opioid ads and other non-sense.

It's quite annoying if you want to use the all-feed. I've blocked at least 6 spam accounts just today.

Just wanted to bring that to your attention. Thanks.

you are viewing a single comment's thread
view the rest of the comments

[–] ernest@kbin.social 10 points 1 year ago (1 children)

Of course, I'm taking action, and on kbin.social, those posts have been removed, which you can verify. However, for some reason, it doesn't always federate with Lemmy. It used to work before, but practically after every Lemmy update, there's something to fix in the code, or it could be related to API limits because some requests receive a "fail" status in the queues. The same applies to spam from Lemmy instances - I had to manually clean up hundreds of posts from other instances because they didn't always send AP requests, which I verified in the logs.

Do you seriously believe that it's a matter of individual accounts, and banning a user solves the problem?

[–] LinkOpensChest_wav@beehaw.org 1 points 1 year ago (2 children)

Am I taking crazy pills? Ban the user, or they'll keep posting spam. Vet your applications like a valid instance instead of letting any spambot create an account.

You are not taking action if all you're doing is removing posts. You are handling this in an entirely inadequate and incompetent way, and you've made kbin a blight on the fediverse. Get your shit together.

Do you seriously believe that it’s a matter of individual accounts, and banning a user solves the problem?

It's certainly better than allowing these accounts to continue to post harmful spam. Then when people sign up, there should be some red flags in their application that you can watch for. It's not like these spammers are highly intelligent. They're pretty blatantly spam accounts, especially when they have literally just a number's difference in their username.

[–] ernest@kbin.social 9 points 1 year ago

By the end of September, kbin.social will receive an update that will address some of the issues. For now, I'm banning spam accounts, and registration is disabled. Additionally, external links are marked as "nofollow." I just want to emphasize that this is not solely a kbin issue - I had to manually remove spam posts from Lemmy instances as well because moderation didn't federate, which doesn't happen with, for example, Mastodon.

For now, the admin of your instance can choose to defederate or remove communities, as the admin of lemmy.world did. Once I've dealt with the update, I will reach out to other admins and discuss the possibility of restoring valuable posts.

[–] ernest@kbin.social 7 points 1 year ago

The problem is that at this scale and with this infrastructure, hotfixes are no longer as straightforward as they used to be - when I could solve these kinds of issues in a matter of hours.