Chris Dixon likes to say that the next big thing will start out looking like a toy.

Some good examples are companies that are building networks on the web. In the beginning, many web services can look like they’re serving relatively frivolous needs. Sharing photos or music, booking reservations, writing reviews, playing games, shopping, etc. It seems that most of the popular web services currently aren’t filling critical human needs and solving critical human problems.

But what all of these companies are doing is this: they’re building huge networks of individuals connected by common interests. And the services that capture the critical mass of users in any given space are in the best position to begin to start to solve the more important problems of that space. Many problems can’t be solved on the web until a strong network is built on the web

Facebook is a good example. Today, it seems that they’re filling a somewhat unimportant need: connecting friends and providing entertainment (allowing people to essentially kill time). But by building a platform that does these few things better than anyone else, they’ve built a network that is almost impossible to match or displace. And it is this network that provides the foundation that allows web services to solve far more important problems. I would bet a lot of money that ten years from now Facebook will be solving problems that we (and the founder) haven't yet imagined.

In short, the point of this post is to say be very careful of dismissing any web service that is building a network as just a toy. It’s likely that they’re simply undershooting, or haven't fully realized, the full breadth of their users' needs and the problems their network can solve.