At the rate some social networks (and other social software) grow, by the end of the decade everyone in the world will belong. Twice.
Ridiculous, I know, but so are some of the numbers and growth rates that get tossed around. The numbers represent something, but maybe not what we'd consider usage in the real world. To a greater or lesser degree, the numbers are haunted by ghosts in the social networks.
What percentage of supposed members or users are ghosts? There's usually no good way of knowing, although the owners of social networks probably have a good idea. My unprovable guess is that ghosts represent the majority of counted users for most social software and social networks, the vast majority in some cases.Clearly, some social networks have real users. Clearly, some have substantial active user communities. Other forms of social software also matter. I have a reasonably active blog. I check in once in a while at LinkedIn. I'm a member of several traditional "social networks" (email lists) that work well.
But I'm also part of Orkut's registered-user count, even though I haven't been there in a very long time. I know people who have four or more blogs under different titles and usernames, and I've seen dozens of blogs (most created in workshops and classes) with one or two posts total. I'm even a registered Wikipedia contributor—but I've done two tiny edits and don't plan to do more.
Oh, and I have an avatar at Second Life—because I needed to join for a contest-judging assignment. Second Life makes it easy to join but actively discourages people from explicitly leaving, so I'm still a Resident—even though I've forgotten that username and never plan to return. I'm a ghost.
I mention Second Life because it's such a high profile social network, with press coverage for "coming up on X million" at a rate that would yield tens of millions in a few years. And because Clay Shirky wondered about Second Life's numbers, as recounted in a series of Many2Many posts in December 2006 and January 2007.
The December 12, 2006 post began: "Second Life is heading towards two million users. Except it isn't, really." Shirky notes the key element: "Someone who tries a social service once and bails isn't really a user any more than someone who gets a sample spoon of ice cream and walks out is a customer." He wondered how many "return customers" there are—and that's tricky.
One cut is easy enough: When Shirky checked, the startup screen showed 1.9 million "Residents," but some 691,000 had visited in the last 60 days—and that's twice the usual time limit for active use. Even that 691,000 isn't unique people—it's avatars, and a single user can have several avatars.
Shirky suspects that "Second Life is largely a ‘Try Me' virus, where reports of a strange and wonderful new thing draw the masses to log in and try it, but whose ability to retain anything but a fraction of those users is limited." He notes Pointcast, one of the classic "Try Me" situations—and I believe Orkut falls into the same category, even if it was invitational. Or maybe especially because it was invitational: You were being invited into a select circle of (a few million) "friends."
How peculiar are the numbers? One journalist blithely reported that Second Life had added a million new Residents during a period in which it had about 810,000 logins—and wasn't disturbed by those numbers. How can you add a million new users when total access from old and new users alike is 810,000? The journalist never asked.
Linden Labs (Second Life's owner) eventually did reveal "real" numbers. Of the supposed nearly two million Residents, 1.5 million "unique people" (actually unique email addresses) had logged on at least once as of the end of 2006—but just over 252,000 logged in more than a month after first trying it. A quarter of a million active "users" (not accounting for single users with multiple email addresses) isn't bad, but it's no two million.
It's certainly not just Second Life. I suspect there's a general issue of ghosts in the social software, that many cited usage numbers are vastly inflated. Take blogs. Are there more than a hundred million real blogs? Almost certainly not. Are there 55 million active blogs (Technorati's claim) actually written by people who post at least once a month? I doubt it. I'd be surprised if the number is half that high. Blogging matters, just as other social software matters, but the claims for usage may be mostly ghosts.