About     Social Interaction Design     Writings     SxD Blog

Social Media Research: Observations

Summary: Six degrees of separation don't make a social system. In social software applications and social media, the first degree of relation involves trust not belonging to the second degree. But use of social networking sites benefit from two degrees because that's where they create new relationships. Social media have moved from social networking as a means of forming relationships to their use in constraining relationships and facilitating discovery.

Friendship in the first degree

In trust-based systems, which includes human communication and interpersonal relations, there's a big difference between first and second degree. Friends, those assigned a relation of first degree in today's social software systems, are people with whom we have a real and existing relationship. That's why in friendship-based social software system like Friendster, our friends have to accept (reciprocate) an invitation to join. The acceptance qualifies them, and creates a first degree relation.

Friendster's success is a fascinating example of that unique way in which communications technologies are able to capture and leverage existing social dynamics. Friendster provides a platform, a place, and tools with which its members can do something they already do in real life, only differently. Like all communications technologies, Friendster is means of production: production of social interactions.

What's captured our interest in social software systems like Friendster is how they leverage the value of social relations.

The irony in them is that while they grow out of existing relationships, they create value by producing new relationships.

They do this by extending what we call trust.

Friends are people we trust personally. Why do we trust them? Because we have a shared history of successfully negotiating various situations involving risk, exposure, vulnerability, etc. We can get trust out of sharing a good time, but it's rough times, not good times, that really test and produce trust. Trust is one of those attributes of interpersonal relations produced by overcoming the potential embarrassment and betrayal inherent in any kind of intimate or close relationship. That's the psychologist's take, and it works for us.

Friendster offers its members the chance to use that trust to make new friends and contacts (read: get dates). Social software systems like Friendster give us the impression that second degree relations, or friends of friends, are close enough to merit or benefit from the trust we have in first degree friends. This would seem like a pretty innocuous assumption. But it challenges two key facts of human communication.

First, trust is dyadic, that is, based on interaction partners (pairs). I trust you, you trust me, and our trust is a foundation and currency of our interaction. The ingredients that produce trust are tested and resolved through these paired interactions. And we can only do this one person at a time. So in all trust-based relationships, there's a fundamental degree of one, practiced one at a time. (Think about your friendships. They're with individuals, not groups).

Second, friends of friends are acquaintances, else they would be our friends in the first degree. Our relationships with them are not personal enough to count as personal relationships. So the best we can do is project onto a person known to us in the second degree, what we know from our first-degree broker, or friend in common. We don't have trust, though we may presume it. The moment we do have trust with that person, really, they're a first degree relation. Second degree relations are good enough to give us a filter, to help us qualify a pool of unknown strangers, and so on. And to that extent the second degree plays a truly valuable role in many kinds of social networking.

The distance between nodes in a network is equal from a network perspective. Indeed, first, second, third, fourth (etc.) degrees have the same numerical incremental difference. But viewed in terms of the economy of interpersonal relations, the difference between first and second is not quantitative. It's qualitative.

Here's the irony of social software systems based on the trust that exists in the first degree: the first degree relation is the closest and strongest, but the second degree relation is the most useful.

We'll see over time what social software systems are good for. It might be that career networking is well served by social networks because they are good at brokering introductions to acquaintances. It might be that they're good for resurrecting old networks (such as alum and classmates networks). If they're to succeed, though, they will need to evolve past their present value as novelties. They will need to offer us the benefits of conducting these kinds of searches and relations online. And that's been the challenge for any social technology.