A common concern among clients who want to engage their customers via communities is the need for moderation. There is a balance that need to be struck between freedom of expression and the need to tame excessively inflammatory posts. Clive Thompson of Wired recently described several interesting techniques to rain in trolls without having to resort too heavily on expensive moderation. Instead, here are some tricks to let your community do the policing:
1. Crowdsourcing. Let a randomly picked group of readers describe posts for a limited period of time. These descriptions are then converted to a rating. Users can then filter out low rated posts, making them less apparent.
2. Selective Invisibility. Again, based on reader ratings. If a comment gets too many negative ratings it gets removed. The trick is that it becomes invisible to everyone except to the one who posted it. This way the troll thinks everybody is ignoring him and hopefully goes away.
3. Disemvoweling. Using this technique, a post that crosses the line gets all its vowels removed making the message still understandable (and thereby not really censored) yet still clearly flagged as inappropriate.
Of course, the best way may be to simply ignore the trolls. NPR, for instance, has a rule not to “feed” the trolls. They remove both the original inflammatory post as well as replies to curb the practice. In addition, requiring registration or some form of identifier in order to post will help create a barrier to entry for the random hatemonger.