Now this is a really clever approach, imo. Very worth watching:
http://www.youtube.com/watch?v=bDsIFspVzfIUpdate: I would say there might be one possible flaw and that is that one would need to apply some sort of probability to the various squares. Sticking with his 2 columns; what if the probability of column B - bottom square is infinitesimally small. What if GCC is occurring (row 2) and we do nothing (column 2) but the extreme outcome case he presents is extremely unlikely and that a much milder outcome is most likely. Then it would not make sense to pick column 2 anymore. This would be the classic risk vs. cost of protection trade-off. If one argues that the potential consequences are so catastrophic that even a small probability event needs to be catered for then we could imagine a very long list of such events that would need to be attended to and then we would end up spending ourselves into oblivion.
In the End, we will remember not the words of our enemies, but the silence of our friends - Martin Luther King, Jr.
