Mollom: spam killer minus the annoyance

I’m late to the Mollom announcing party, because I was on a vacation. Nevertheless, I hope I can still interest some of you with a slightly different angle.

The major issue with spam prevention is that it often (currently virtually always) involves extra steps for normal users. And more steps means less participation. Less participation means less traffic. And less traffic means less popularity, revenue and whatnot. So clearly there is much to be gained to prevent spam without annoying normal users.

And this is exactly what Mollom tries to do: minimize spam and minimize annoyance. Additional benefits are that you no longer have to moderate content (nor users, because Mollom applies its magic on the user registration form as well), and – this one is pretty amazing IMO – ”improve the overall content quality”. Konstantin Käfer explained this pretty well:

While I get virtually no automated comment spam anymore, I now get actual humans who post spam comments. These comments are usually somewhat related to the blog post. That means I’m not getting generic “This content is great!” comments, but for instance, “This bundle is very useful. Thanks!” on my blog post about the TextMate Drupal bundle. Sounds like a valid comment, however, the supplied comment author homepage contains clearly a spam URL.

I can confirm this observation: when I get spam comments, they’re actually relevant to my content, but it’s the author’s URL that makes it spam.

I’m also one of the private beta testers and I’d like to thank Dries for choosing me to participate, because Mollom is really by far the least intrusive spam killer, and for that reason a growth accelerator. I will be recommending it to all my clients.