A group is its own worst enemy

Lots of people and companies want to know how to put together a great online community. Or scale an existing one way up. Not as many people consider what happens when a community goes bad. This was the topic of a great presentation by Chris Tolles at SXSW called When Communities Attack.

Here are a few points I found interesting.

The tone in discussion forums gets more friendly if posts are geotagged. The rationale, while not proven, is that a degree of anonymity is lost this way and that no one wants to associate a shameful post with their vicinity (or perhaps even suffer being located).

Lots of people have “conversations” with themselves in online communities using alternate screen names to establish credibility.

The word for the non-machine-readable letter grid that is often required for users to input to validate themselves is called a “captcha”. Didn’t know that.

Registration often works against decorum by keeping out good posters who prefer anonymity and by encouraging flamers since registration implies that this community has something good going on inside of it.

I’ve been noodling on this question from a different panel, called Bridging the Online Cultural Divide, since Austin:

Do social networks conjoin communities — i.e., technology facilitates connection where it could not be accomplished before — or does it merely create closed communities by allowing like-minded folks to cohere and separate from the ‘others’? Put another way, is a social network inherently based on segregation or inclusion?

Post title from this great paper by Clay Shirky.

That’s it for recaps of SXSW. Back to the other stuff …