The Lies Social Networks Keep Telling Themselves

British developer Tom Hume recently went to hear a talk by the anthropologist Robin Dunbar — who famously posited that most humans can only handle around 150 social relationships — discuss his views on our ultramodern ways of staying in touch.

The session, titled “How Many Friends Does One Person Need?”, took a look through our lives and established that Dunbar’s number appears in all sorts of ways and in all sorts of places: in the military, in families and in business, for example. But aside from encouraging some fairly standard questions (does anyone really get value out of having 5,000 friends on Facebook?), it also prompted Hume to really examine what social services online are getting wrong.

As a result, he has outlined what he calls four “lies of social software:”  the set of assumptions almost every social service online makes, despite the fact that there is ample evidence that they’re wrong.

And they’re pretty bang on:

  • Your friends are equally important
  • Your friends are arranged into discrete groups
  • You can manage hundreds of friends
  • Friendship is reciprocal and equal

Almost every service offers you a way to make a connection with as many people as you want, and tools to help you categorize that connection into one of a few buckets. Many of us have started to adopt this way of managing our online friends, to try to eke some efficiency out of the system, but let’s be honest: Very few of us manage our lives in this way. We have siblings who are friends, and siblings who are not; we have co-workers we’d share intimate secrets with, and those we just can’t stand. We have friends who are closer to us than we want, and acquaintances who are further away than we’d like. In short, people are messy — and very few pieces of social software are able to reflect the complexity of real relationships.

When they do, they rarely get the credit they deserve. For example, I think one of Twitter’s great benefits was that it made the relationship between two users asymmetrical. I can follow anybody I like, but there is no real reason — besides a sort of social etiquette or the need for backchannel communication — for them to follow me back. That gives more power to celebrities and broadcasters, which can bring more new users in, and it also reduces the influence of spammers and makes the system more scaleable for different sorts of users. Your network can be pruned to be large, small, broadcast, narrowcast — precisely can choose who to follow without finding yourself overwhelmed.

Hume wonders whether, ultimately, we aren’t the ones holding things back:

Managing lists of friends is unpleasantly icky. I bet Google (s goog)  or Facebook could take away much of the pain of creating these lists by analysing my flow of communications. I bet they could notice and prompt me to confirm changes (“you’re emailing Freda a lot at the moment — working late or is she a friend outside work nowadays?”). Perhaps the challenge is less technical and more how to present this to a privacy-concerned public;

Context-aware sociability is definitely possible. For all that people deride Google’s social efforts, for example, the company has shown with Gmail’s priority inbox feature that it can use our patterns of communication to determine what is actually important to us.

But I don’t think fear over our privacy is why we don’t let these companies in. In fact, we’ve been complicit in helping them erase privacy in many senses. Instead, I wonder whether it’s just that the laws of social networking are simply based on what others have done before.

In many ways Facebook is not a great deal more advanced than it was when SixDegrees and LiveJournal helped set the standard: and it still, by and large, subscribes to these same mistakes about how human relationships work.

Is it something that will ever be fixed?