My six months with Mollom

As many Drupallers know, Mollom has been out for a while now, six months in fact, and I just wanted to share my enthusiasm about it.

I had the pleasure of speaking a bit with Dries about it in the shuttle bus back from Szeged, and it seems that Mollom’s future is truly exciting.

Before Mollom, I’ve been using spam.module and Akismet with varying degrees of success, and when Mollom first came out, I wondered why we needed another Akismet, but decided to try it out to support Dries.

The answer to that question is that Mollom is in fact not just another Akismet. Spam blocking is just one of the things it does. That does not mean that Mollom is not good at blocking spam. My experience is that Mollom is more effective than Akismet and spam.module combined.

But the thing that Mollom does is intelligent text recognition with a much broader scope. Mollom can decipher what language content is written in, and even more importantly, the “quality” of the content. Quality is such an elusive thing, and it is of course not all aspects of quality that Mollom is able to figure out, but Mollom has learned many of them, and I think we’re going to see more services in the future taking advantage of this.

Because spam is not the only challenge to internet users today, it is merely the first and most glaring. The ability to find the proverbial needle in the haystack, the few well-informed and interesting remarks in the torrent of ill-informed, poorly written tripe that is the standard Digg comment page.

Or perhaps the ability to filter out content that is not necessarily spam, but just unwanted for any kind of reason. Recognising link-whoring, trolling, racism and similarly unpleasant content.

As I understand it, all that and more is in the power of Mollom, and it’s going to be highly interesting to see where Dries, Ben and Wim are going to take it in the future.