On Sun, Sep 27, 2015 at 4:49 PM, Jean-Francois Simon <jfsimon1...@gmail.com> wrote: > If one had to find a hardware most difficult to compromize which one would > you take ?
None. The hardware most difficult to compromise would be hardware with nothing on it to compromise. An acceptable second place would be hardware with nothing on it worth compromising. Of course, that's not very useful. Then again, one could also say that about most computers deployed today (possibly even with some degree of accuracy - though that winds up being a matter of perspective - different people have very different concepts of "useful"). A more interesting (and perhaps - depending on perspective - more useful) concept might be compromise detection. For this you need checks and balances. Good models for this kind of thing might be "double entry bookkeeping" and/or "pain neuroscience" (the underlying abstractions are similar). Of course, that also leaves the issue of: what are you going to do in response to compromise attempts? I hope this helps... -- Raul
