the brain’s metabolic efficiency. I twisted my ankle running early in the trip and thereafter got little exercise beyond my short walk to work. Paul Erdos famously said a mathematician is a machine for turning caffeine into theorems; I became a machine for turning caffeine and high-fructose corn syrup into code, and I was hardly alone.
Our motivation to work hard was very simple: succeed or die. Those of us who were foreigners had some chance of escaping the country should it slide into civil war, but most of the people there were South Africans, and legitimately felt they were working to save millions of lives and quite possibly their own. The country was looking to us to orchestrate a peaceful transfer of power. While there were many means by which peace might fail, we were informed that our failure would all but assure a civil war costing an estimated 10 million lives. It would be difficult to imagine a startup with a stronger sense of purpose than the people I worked with.
The environment facilitated hard work as well. Most of us were away from home and family, in an unfamiliar city with few amenities. We lived in hotels and ate prepared food, so we had no household responsibilities. Perhaps most of all, we had the mother of all hard deadlines: the election date was set in the interim South African constitution and was absolutely unmovable.
But this effort level was not sustainable. Physical and mental breakdown under the strain of this effort was fairly obvious in me and my coworkers. Wounds healed slowly. Colds and minor illnesses were common. We all became irritable, histrionic, and more than a little paranoid. In the last days of the project, the organization began to disintegrate. Worse, we were far more impaired than we realized, as many studies of fatigue have shown, including a notable one by author John Caldwell.
Also, we made mistakes, many of which would have been avoided by people with more sleep and perspective. These mistakes added to our workload, of course. We also had a lot of near-misses. For example, I ended up in charge of the Administration Directorate operations center the first night of counting. We had to drop our connection to the outside world to reboot a crucial server, or else endure deteriorating performance. A noticeable outage would have fueled rumors of malfeasance. I had to make the decision. It was a completely freaky situation, and I was glad I’d had a decent night’s sleep about 16 hours earlier.
We were strongly motivated both by desire to succeed and fear of failure. Did fear for our own well-being reduce the number of hours we worked? Hardly. Work was the easiest way to ignore the anxiety caused by the literal anarchy and acts of terrorism around us. But while fear did not limit our work hours, it likely made those hours less productive than they otherwise would have been.
We met our deadline, of course, at the cost of an organization depleted of intellectual, emotional, and physical reserves. The organization continued to function, performing winding-up operations and finalization of results, but any bad surprise at that point could easily have
