Fighting bad bots - Anubis activated

[nephele]

@Pap

Bots get smarter these days, and they can bypass approval of the first post by copying an older legit post somehow, and paste that one as their first “post”.

Sure, maybe? But that would mean they have to fight against my memory

There is always options to bypass this, the cheapest option used to be “pay people in third world countries”, now the cheapest option is “use AI”, just that AI is much easier to detect for us…

@Pap

Spammers are always one step ahead,

I disagree… There have been attempts to spam this forum for the entire team I’ve been moderating it, but they only manage to get anything visible only ocasionally, this this morning was an exception, not the rule.

[Pap]

@PulkoMandy

This means new users can’t join the forum in any meaningful way. Which seems more annoying than having to deal with spam messages from time to time.

It certainly delays new users from joining for good, but only for a little while. You post something and it may or may not be visible the next day, depending on moderators’ time. But I think whoever wants to join Haiku’s discourse won’t be stopped by that. Most of them already did the first step to try Haiku, a much more important step. And apparently they liked it, otherwise why would they want to join Discourse. I doubt such a moderation measure would make them give up.

FreeBSD has this model in their forums. Now, admittedly, FreeBSD forums are not exactly famous for newcomers’ welcoming. But that’s not because they have this “first 10 post hidden until approval” rule. And the rule seems to work.
At any rate, I read today’s spam attack is not the norm here, so maybe such a strict measure is not needed. It’s up to the moderators really, I just wanted to mention another way to monitor spammers, just in case.

[nephele]

@Pap

And apparently they liked it, otherwise why would they want to join Discourse. I doubt such a moderation measure would make them give up.

there are quite a number of users here who do not run Haiku normally, and i know of severall instances where people signed up to write a single sentence reply to something. That is perfectly fine on it’s own. As for moderating messages that are to be aproved we should probably start with 1st message only. Currently it is 1st message only but only if user wrote it in a time span shorter than X, so it’s a bit of a heuristic.

If this ever prooves insufficient one can decide on more harsher things like having the first N posts be hidden until aproved, but i’d rather not do that when it isn’t absolutely neccesary.

[nipos]

I’m generally not a fan of those centralized/proprietary spam detection services,but what about using Akismet to detect if the first post is spam?
As far as I know,it’s already in use on Trac and for every post there(?)
Such algorithmic automatic services can make mistakes,both false-positives and false-negatives,but if it only decides if the first post by a user is shown or to be manually approved,I think it’s okay and better than hiding every first post of a new user.

[smashIt]

@PulkoMandy

At worst it will be manual approval of the first post by every new user by the moderators. Not fun to do but it would surely slow down these attacks

In my experience the only reliable option.

[win8linux]

@michel

Judging by this morning’s thoroughly be-spammed inbox, there are scams the mighty Anubis is powerless against.

IIRC Anubis isn’t used for the forum?