Despite Young Dave's assessment, I think it is possible that Facebook will make the leap from its predominantly young demographic to mainstream, grown-up usage. If it does so, the transition will happen soon and will not take long.
The fundamental issue is this: a good network includes everybody you would want to contact. However, a network that has literally everybody has too many people abusing their power to get your attention.
When entering data into Facebook, you're sending it on a one-way trip. Want to show somebody a video or a picture you posted to your profile? Unless they also have an account, they can't see it. Your pictures, videos and everything else is stranded in a walled garden, cut off from the rest of the web.
Like locked cell phones and copy-protected music, Facebook is on the wrong side of the open-network debate. Facebook is a sealed bubble.
I'm not so sure that Facebook is on the wrong side.
This has been a longstanding controversy. In the early 1990's, many Internet veterans regretted the fact that the Net was getting easier to use. In the days before Windows 95 and AOL, the technical competence required to get on the Net was an entry barrier that kept out the riff-raff.
Facebook began as a site that was exclusively for people with college email addresses. Many young people, including my daughters as well as my friend's son--who I call Young Dave--think of age as an important barrier to entry relative to Facebook. For them, imagining people my age using Facebook would be like a teenager thinking of her parents having sex.
My hypothesis is that Facebook is to the Internet as Craigslist is to eBay. That is, it is a service that is better known to the elite than to the masses. This improves its character in a number of subtle ways.
I see Facebook as an experiment in "filtered connections" in social networking. I think that there is a very large latent demand for such connections, and this latent demand is as strong among adults as among young people. For example, in the essay I suggest that filtered connections would help me implement a concept I am working on called Peripatetics.
I think that there are many places where one could find a dozen people interested in the sort of educational experience I sketch out. However, the cost of identifying those people is quite high. Eventually, Facebook or one of its descendants will lower that cost.