Facebook would like you to believe they are giving you a new platform. Anybody who's been on the Web for a while knows that Facebook is selling old wine in new bottles. And yet building Facebook apps can still make sense.
People like to talk about leveraging social networks, but here's what they leave out. Joining a social network takes five minutes (if that). Staying active in a social network is a full-time job. I know exactly what that's like, because I used to organize and promote nightclubs and raves. Going to a club takes five dollars. Being a clubber or a raver takes time. In fact it eats up your life (but in a good way).
Social networking Web sites aren't platforms. They're not circles of friends, either. They're nightclubs. Some people spend all their time at these things. Some people never go once. A new social network becomes hot, everybody who's everybody is there, and one day, suddenly, without warning, everyone is somewhere else. At most of them all you have to do to get in is show your ID.
There's a lot of money in nightclubs. You go to the Winter Music Conference in South Beach, FL, you can easily spend a hundred dollars just to get into one particular club one particular night. And it won't be the only club you go to that night. There's a lot of money in club music, club clothes, expensive vodka, designer drugs, laser lights, and really big speakers. The people who work that system successfully do very well by it.
But if you're building a Facebook app, you're building a sound system you can never take out of the club. Spending money on something which won't work anywhere else only makes sense if the payoff is immediate. It's not really an investment, because assuming any given social network will persist for any given amount of time defies history. These things have been growing first hip and then stale on a cyclical basis for five years. A New York Times article blamed various executives when Friendster went from hot to not, but fashion features a great deal of randomness. Criticizing upper management for being unable to predict effectively random phenomena is like criticizing them for being unable to defy gravity. It might make a lot of sense to leverage social networking applications for business purposes, but it definitely doesn't make a lot of sense to do so in a way that locks you into any one particular social network.
Some companies think building customized social networks is the answer. Some developers think OpenID is the answer. Personally, I think the smartest thing is to do what nightclub and rave promoters do. They go to other people's raves and other people's nightclubs and they pass out flyers for their own events. If you're building a Facebook app, you want to make sure it directs people to your own site, because one day Facebook won't be fashionable and nobody will be there. Like Myspace - nobody goes there, because it's too crowded.