>I suppose what I'm bitching about (and I've probably just faceplanted
>by stepping into that particular pothole - it's my turn, I guess) is that
>there seems to be no part of the threat model where risk is
>acceptible.  I mean, going all the way back to hand-wired
>electromechanical processors just to be able to bootstrap back to
>silicon and losing 20-30 years of technical advancement?

>Somewhere, we went way off course.  There is a saying: "Perfect is the
>enemy of working."  I think that's where we as a group have lost our way.

>The threats are known.  The risks are known.  Let's act.

I agree with that.  I think it's better that we get 50% of the population to use encrypted phones, where the encryption isn't truly known to be perfect, than to get 1% of the population to use perfect encryption.  Verifying the last little bit of doubt is going to cost a rather large amount of money.  Raising the demand for crypto phones to 50% represents a huge market, which will be satisfied, and the profits for that market will pay for the next generation of closer-to-perfect phones.

              Jim Bell