Regarding the recent wave of spam - what are we doing about it?

Question

There's been a lot of spam and nonsense posted lately, everything from love connections to kids struggling through puberty demanding to be made moderators.

What is being done about this, and what can I do to help?

Answer 1

For those that don't know me, I work for Stack Exchange.

We've been on top of this since it started. We noticed some abnormal spam activity going on here which your moderators quickly confirmed.

We've taken the following steps to keep it out:

• Registration is now (temporarily) required to post a question
• Spam originating from Ask Ubuntu is dealt a much harsher penalty score in our network-wide spam fighting system. This means spamming AU is going to get you blocked very fast

The problem has been that some of the people / bots doing this have been cycling through Tor nodes, some so distant from where we suspect the abuse is originating from that we're amazed the sites were even functional for them.

The system works on a level that's lower than individual I.P. addresses, so it's not terribly difficult to keep these people out. In fact, there's a very good chance that the spammer that wrote whatever you just found has probably already been, or will very shortly be blocked. However, the increased volume leaves more artifacts of why they were blocked around to be cleaned up.

The good news is, we're still rejecting significantly more than we're flagging, and more granular tuning should make this even more effective. Here's what it looks like now:

I've got some tweaks and changes in the hopper that I want to experiment with briefly on Monday, and quite possibly push if they look promising. I hope to bring this back to where it was around January 5 - 7. Part of the reason for this is one of these fools keeps cycling through Tor nodes like some people cycle through Pringles. This is easily and quickly dealt with (I anticipated this when I designed the system), but it does leave some more artifacts of the carnage lying around, which is what you've noticed.

I'm also considering doing something to delete any zero or worse scored post contributed by something the system (through the signals you send it) determined to be evil - this will help minimize the amount of crap you have to see in order to keep it out effectively through flagging - but it's also kind of dangerous. I'm just trying to be as transparent as I can when it comes to what I'm thinking about doing next.

Now, more importantly, what can you do to stop it?

Exactly what you've been doing: flag the post as spam and move on. Don't interact with it via comments. If it's a bot you're just wasting seconds of your life; if it's a moron then you're wasting seconds of your life and encouraging them even more. Burning this stuff should be devoid of any sort of ceremony, just like picking up litter.

There are rare cases where you might find the need to leave a comment as a beacon to other users that the post is spam, such as someone impersonating a high reputation user or moderator to try to confuse folks. However, engage your fellow community members, not the person that wrote the junk.

It should go without saying that you should definitely not flag something as spam that is not a blatant advertisement or purely nonsensical trolling attempt of some sort, and not flag something as offensive unless a reasonable, non-technical person would find it offensive as well. These flags carry serious weight, so use them wisely.

I'm going to be continuing to monitor how effective the system is being, and making additional changes to the configuration for AU. You can take some comfort in the fact that the work you're doing by flagging is helping to stop it, and that we're closely monitoring this and doing things on our end to help.

Thank you, to all that have worked tirelessly to keep the site clean. You've been doing an awesome job, and I hope to see the end of this soon.