Sybase iAnywhere SQL Anywhere Mobile and Embedded Database

I’d rather play golf


Thoughts on data management, autonomic computing, and self-managing database systems.

header image V6

Fighting spam with Mollom

September 22nd, 2008 · 1 Comment

According to Dries Buytaert, the creator of the web-based CMS framework Drupal, and his colleague Benjamin Schrauwen, some 80 percent of all blog comments worldwide are spam. As unfortunate as this is, the number isn’t a surprise to me since I have to manually deal with spam comments everyday that break through the spam filter this site currently uses, which is Bad Behavior.

Buytaert and Schrauwen have now released a commercial SaaS package called Mollom that offers significantly better spam filtering than typical packages, using a combination of proprietary AI techniques and resorting to CAPTCHAs when the content is in-doubt. Traffic going to and from Mollom servers is encrypted using a public key cryptography system. That level of security is insufficient for some high-security applications, but is perfectly adequate for the vast majority of web-based applications. I’ve been using Mollom (as a beta customer) for several months for one of the websites that I manage and the effectiveness of Mollom’s techniques are impressive. Mollom can be used to filter not only blog comments but any content submitted to a website.

While Mollom is not generally free, it is free for sites who generate fewer than 100 true posts or comments per day. You can check out a FAQ about Mollom here.

Tags: Uncategorized

1 response so far ↓

  • 1 Fishman // Jun 24, 2009 at 11:42 am

    There should be some sort of database where all of the spam comments and broken CAPTCHA entries live.

Leave a Comment

Note that all comments are currently being moderated until I have a better handle on spam, so your comment may not appear for a couple of hours

Sybase Privacy policy