[FoRK] Why Fukushima made me stop worrying and love nuclear power
sdw at lig.net
Fri Mar 25 14:33:33 PDT 2011
On 3/25/11 12:42 PM, Gregory Alan Bolcer wrote:
> In simple terms, if someone doesn't use the lock correctly, you can design another iteration (and another, and another) to reduce
> forgetting. Thus all products in the world incorporate some type of risk tradeoff. Fukushima argued that the circumstances
> following the tsunami were 1-in-1000 year faults. Obviously they were wrong.
A "1-in-1000 year fault" is a probability estimate. They may not have been wrong, just unlucky. We'll have the first complete
sample in the set of Z trials in another 1000-n years... Clearly, they should have designed for a multiple of the expected worse
case instead of a few percent above. I think I heard that they built a 30-ft wall to contain waves expected up to 20-25 ft. They
should have been at 60 or 90 feet, or had failsafes that could survive that level.
> You can't use the term "perfect" in your definition; that makes it circular. Perfect would imply there are no further design
> changes to reduce failures by stupidity.
> Again, I'm not arguing *all* possible, just that it can be incrementally changed. I'm arguing induction, you are arguing limits.
More information about the FoRK