9

I'm a moderator on a small site. I can't afford to build my own filtering system for automated spam. But surely this is a solved problem; some sites do have built-in tools for identifying spammers and blocking their IPs and content. How can I do that on my site? Is there, for example, some centralised tool for managing such lists across sites?

For a small site, managing automated spam can be difficult without the resources of larger sites, so something like this would be invaluable.

3
  • 4
    This question appears to be off-topic because it is a request for an external resource.
    – ChrisF
    Jul 29 '14 at 21:42
  • 2
    @ChrisF How do you get from A to B? Ah, you've raised the issue on meta (you should have posted a link to that). Jul 29 '14 at 22:08
  • @Gilles - ah yes. Forgot to do that. Sorry.
    – ChrisF
    Jul 29 '14 at 22:09
7

There are many technical tools that can be integrated with existing software.

There are non intrusive methods (methods that don't require extra user interaction):

  • Akismet has methods to integrate with content management systems, forums, e-commerce sites. It also has libraries for several languages if you've built your own community application. It advertises that it catches 7.5 million spam comments around the web an hour. It is free for personal sites.
  • Stopforumspam is an API accessible database that contains known spammer information, which you compare against. It has plugins for almost every forum software imaginable, plus wordpress and contains a Java library for you to develop against. This option is also free.
  • Blacklists. I haven't seen an all encompassing one for forums, so this would need to be searched on a case by case basis. Something to look for though is that the plugin looks at multiple blacklists, not just one. As an added feature, if you can change the weight of the blacklists and block spam only when certain thresholds are reached, you are less likely to block legitimate traffic. One plugin I've used for a PHPBB board is Advanced Block Mod. It has prevented a lot of spam from appearing, according to the logs it keeps

BL Counts

There are also intrusive methods (extra steps someone has to jump through to add content to your community)

  • Many forums have built in methods of spam prevention. These include answering questions to proceed. Examples I've seen include simple math questions ("What is one plus one?"), simple questions ("What is the name of this community").
  • Recaptchas. Everyone knows what they look like. These can be added at registration time, comment posting time or a combination of both.

Recaptcha

Have you noticed that these have gotten easier recently? That is because Google updated it based on a decade of detecting humans vs bots around the world.

As part of this, we’ve recently released an update that creates different classes of CAPTCHAs for different kinds of users. This multi-faceted approach allows us to determine whether a potential user is actually a human or not, and serve our legitimate users CAPTCHAs that most of them will find easy to solve. Bots, on the other hand, will see CAPTCHAs that are considerably more difficult and designed to stop them from getting through.


A combination of intrusive and non intrusive methods is most effective. However, personal experience has shown that every intrusive method added reduces the likelihood that people will contribute. Especially if they have to jump through a hoop every time they wish to post. Most users are understanding of these methods at registration.

The non-intrusive methods run in the background and normal users may never know they even exist. This is ideal, because it doesn't interrupt their flow of ideas or affect what they need to do to submit content to you.

2

One of methods for identifying automated spammers: create several 'textareas' for content entry, mixing up their names and holding the right one in some key. Hide the remaining ones through CSS. If any of the dummies contains any content, it's a bot.

1

There are multiple companies and sites offering various capabilities. For forums, I would employ something like Akismet to block spam posts, but you can also often find out whether an IP is regarded as spam by entering it into Google and seeing if results appear, or by using a site like stopforumspam.com

0

On a site I'm on, there is a place where you can get free bitcoins called a "faucet". What he did was make a dummy button, and put it in before the real button, and if the dummy button got clicked from a bot trying to automate the faucet, then it would automatically ban them.

I think the same thing can be used here, by putting that in, and if a bot tries to post something, by clicking on the first link, it would automatically ban them.

0

I basically solely manage an old but active forum and it gets hit with spammers on a daily basis. Since the owner is MIA and the subscription for the spam filtering service (IPB...) ran out, I can't use reCAPTCHA or automated banning methods.

To catch spammers this way, I need to employ manual methods. One such method is to moderate a new user's first few posts to ensure it's non-spammy. I also look for email addresses that are clear gibberish (such as juwe78tgri3@somemail.com) which often are paired with similarly gibberish usernames.

If you have automated tools, use them. Most software has some sort of spam-capturing software either built-in or as a downloadable plugin that's installed. Otherwise, instinct is your key ally.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.