Friday, February 13, 2009

Holy Zuck

I met the Zuck today.

It all started a few weeks ago, when I joined Stanford's CS106 section leading program. I lead a discussion with a group of seven CS106A students every Friday, help people with assignments, do some grading, and party with other section leaders. Today, a group of us went to Facebook. We walked through two of the engineering floors; even in the late afternoon, most of desks were manned, and most of the engineers didn't bother to look up as twenty students sauntered past.

Later, we filed into a small conference room. We looted the open cafe at the back of the room; one section leader pocketed three Snickers, and I grabbed a protein shake. We sat down to a terse introduction from one of Facebook's PR people: "We call him the Zuck." We waited. I looked to the door just in time to see an unassuming, incongruously dressed, boyish-looking man slide in. The room was quickly silent.

Mark Zuckerberg was wearing brand-new sneakers and designer jeans, a conventional North Face fleece clashing with the shirt and tie underneath. He would not have looked out of place at a Harvard frat party. Yet there he was, twenty-four years old and worth $1.5 billion. One Stanford CS professor had recently announced that he was leaving to work for Zuckerberg full-time. "Welcome to Facebook!"

The Zuck told his story -- how he'd hacked together a some simple apps during his sophomore year at Harvard, and how those apps later become Facebook. The first was a product of necessity. Two days before a mandatory freshman humanities class, he had been "busy building stuff", he said, and needed a crash course. Badly. It was an art history class, so he built an app that let students upload images; they could see their friend's uploads and pick a few sentences to describe each one. Afterward they'd contributed, they could see what other students had written. The professor later told him that people had scored higher on the test than on any he'd given before. Then came Thefacebook, which let users put up a picture of themselves, a few words, and poke each other. Zuckerberg said wrote in three weeks, and that three more weeks were all it took for a few hundred students to join.

Then came the Q&A.
First I asked him: "You said you built Facebook and a few other projects in your spare time at Harvard. What gave you the impetus to take all that time and just write code?"
The first part of his answer was all about hacking for it's own sake. He said that he's always like to build stuff, to hack things together quickly and to see what other people do with it. The second part was about network effects. He said he was fascinated by them. People had complex relationships to each other -- now (in 2004), with everyone constantly in touch, those were only getting more complicated, but there was no easy way to map those relationships. He realized that just as collaboration had made that art history final easier for everyone involved, it could also make relationships easier to navigate. He could use people's desire to keep in touch to map out their real-world social networks and to make their lives more transparent to themselves and to each other.

I asked him, later on, what Facebook's original killer app was. He told me: there was none. The original Facebook had no apps, of course, no status updates, no "friend feed", not even a wall. More than any social network at the time (and certainly more than Myspace), however, it emphasized existing friendships. It started only at Harvard, and so built a community of trust where people on Thefacebook knew each other, were actually friends with their online "friends", and were willing to share their lives online. That philosophy -- not any individual 'app' -- was the real "killer app", of course, and it still is.

Meeting Mark Zuckerberg was fun and surreal. Here's what I got out of it:

  • technology is cool, but the people who use it are far more important and require far more attention to detail than it does. Doing things "the right way" is good, but it's not nearly as important as most engineers take it to be. Facebook is best example of this I can think of. It started as an app written in PHP (which originally stood for "Personal Home Page", but that's another story). PHP is a scripting language, and not particularly fast. It ran on Linux, in a "LAMP" stack -- about as standard as web setups get. And it borrowed the basic ideas of a social network -- profiles, "friends", personal blurbs, and so on -- from sites like Myspace that had been around for years. It was not rocket science. It was just closely in tune with the way people connect in the real world and the way the want to communicate online.

  • nothing is impossible. Cheesy, I know, but literally nothing. Imagine you were Zuckerberg's dad, and he had told you in 2004: "I don't need to finish Harvard! I'll just drop out and become a billionaire like Bill Gates." That would have sounded ridiculous, but nonetheless: he would have been right.

  • most importantly, build it now. Facebook started on a whim and version one came out in three weeks.

Its either now, or its not at all. Make it happen! Seize the day! There are so many Facebooks that nobody's thought of... yet.