I think your reasoning about 4 is wrong. It does not look solid: in the long run, the security of all crypto goes to zero. There's a very real possibility that

I think your reasoning about 4 is wrong. It does not look solid: in the long run, the security of all crypto goes to zero. There's a very real possibility that the current system will in time be replaced by something which is intended to be better, or to fix a to-be-found flaw, but which is itself flawed in a much worse fashion. In fact, that's quite likely: under the circumstances which such a change would need to be made, there'd be a panic and quite a pressure to get it done quickly. That being said, I don't agree with the article's conclusion, either. If/when something goes wrong, it's a completely open question whether or not the reaction will be catastrophic.