> Things are never done 100% correctly Risk is rarely phrased in terms of "100% correct" because risk-management in engineering is well aware of the "swiss che

> Things are never done 100% correctly Risk is rarely phrased in terms of "100% correct" because risk-management in engineering is well aware of the "swiss cheese model" of failure. More accurately, disaster occurs not when things go less than 100% perfect, but when things start approaching 100% wrong. All of the redundant layers of safety have to fail such that catastrophic failure is the outcome. The real question is how many layers of safety is enough? Chernobyl, clearly insufficient. Fukushima, still flawed, and folks knew so at the time. The high water mark is such that none of the mistakes of Chernobyl or Fukushima would ever be made again, and we have not one, but two generations and 50 years of improved designs since Fukushima.