I’m focused here on the "social" not the network, and the average user. Twitter may very well be the foundation of the real-time web and possibly an enabling component to a future semantic web. However, it has become clear to me that it is not a very good social network.
The realization that this was the case came over time while increasing my use of Twitter, putting messages out there and watching what came back. I noticed that the people responding were not the people I was following. I expect this is counterintuitive for many new users and likely leads to confusion. The people you follow are often not the people who follow you. Your view of Twitter, the people you see, is not where your updates are being directed.
Contrast this to Facebook, where nearly everyone you "follow" also follows you back. The only way for these two groups to not be the same on Facebook (followers vs. friends you follow) is to use the "hide" features of Facebook to completely block the updates of people in your friends list. From my informal survey this week I found that hardly anyone hides friend updates.
If you really look at who you follow and who follows you on both Twitter and Facebook, for most users it looks something like this:
For Facebook, your updates go out to the same people you are getting updates from. They see your updates and you see theirs. Further, given the ability to comment, "interaction" happens. This is the "social" part that Facebook makes work.
For Twitter, your updates go out to a different group than those whom you are watching. There usually is some overlap but that overlap tends to be less than you think and is effectively minimized even further if you are following any number of the popular and prolific Twitter stars. These users or feeds tend to fill your view of Twitter updates but have a very small percentage of users they follow back.
Twitter is like standing in the middle of a crowded party where everyone faces one way. You get to listen to the people in front of you but only talk to the people behind you.
Take this case of some tech blogger early adopters of Twitter. I wrote a tool this week that goes out and looks at their followers and friends and then compares the two lists to find the overlap. They look like this:
Take the case of Robert Scoble, a popular blogger, a very early adopter of Twitter, and a proponent of "following" as he has discussed in the past. Only 43% of Twitter accounts that Robert follows follow him back. Further, his list of followers is nearly 7 times his list of friends. If you follow him, chances are he is not following you back. This is even more extreme in the cases of Tom Merritt, Leo Laporte, and Molly Wood (some of my favorite webcasters/bloggers) who’s lists of followers are 70, 131 and 185 times bigger, respectively, than their friend lists.
A picture of some of the more average users from my friends and followers that I sampled looks like this:
The conclusion is pretty clear, without a lot of work to push those two circles together, Twitter is not much of a social network. Instead, it is a micro feed service which exhibits some of the characteristics of a social network and in special circumstances can be made to act like one. Twitter is more of a one way fan/feed service where only some limited interaction with your fans or people you are a fan of typically happens. Experienced users are able to hammer this model into some semblance of a social experience, but for many the "social" will be elusive. Understanding this will hopefully help users get what the service is and is not, and allow them to use it for what it does well without getting turned off by missed expectations.