That vision was world-changing. Prior to Facebook, the paradigm for online communicaton was email. Email was based on the old business memo. Zuckerberg saw a different way for the world to connect. The reason he was able to succeed so amazingly (unlike many other social network products that fizzled) seems to be an early decision he made to let the users of Facebook determine what it would be, and then to provide the very highest quality for those features.
Thanks to Facebook, the paradigm of communication has changed. Instead of issuing memos we communicate in social networks, with our picture next to our writing and a mixture of broadcast and address. Many children use Facebook as their primary form of online communication. Who knows where it will go next.
Facebook isn't the whole story, of course. There's Twitter and texting and other variations. When historians look back at this period, they will be able to chart the rapid evolution of communication in these still-early years of the wired world. What we do now is nothing like what we will do in ten years.
There is also an evolution in how software products get created. So many of the most successful companies are newly formed, youth-run companies because the staid old companies don't seem to be able to evolve. Successful business is moving towards a much more customer-driven model, which is what Zuckerberg created as an intuitive, naive 19 year old who hadn't been taught how to "properly" develop software. Old businesses try to move towards the new model by adding customer usabilty to their development process, but that's like putting a bandaid on a broken leg.
The fact that Zuckerberg's first social network was a university site to rate the hotness of girls is not a minor point. He developed something that he thought was cool and that he thought other people would find cool, and then he tossed it into the sea and let other people pick it up. He then swam beside them, altering it according to their use of it. You couldn't do this so easily with a product that has to be released, but part of the new business model is creating a product that is fluid and adaptive. It may be why cloud computing or something in that direction is the necessary evolution of software. Downloads and upgrades and new feature lists are all so cumbersome and annoying. In five years we may look back on them as dreadfully archaic, like typewriters.