It's not encryption vs. safety. It's functioning software systems vs safety and there is actually no conflict because functioning systems are necessary for safety.
So really the FBI director should have phrased it as "lack of encryption vs. safety".
So really the FBI director should have phrased it as "lack of encryption vs. safety".
If all network devices were trivially penetrated by government surveillance, a hypothetical terrorist plot with a WMD has chance A of being discovered. If network devices that are impenetrable to government surveillance are used, I'm not going to argue that the same plot doesn't have < A chance of being discovered (ceteris paribus).
I happen to believe that's a good and just tradeoff to preserve the values of a democratic society, but I accept that others might think otherwise.
The US (and France, and the UK, and everyone else grappling with this that cares to pretend to be a democracy) should absolutely have a thorough, resolute discussion about our options.
[0]http://www.cdc.gov/alcohol/fact-sheets/alcohol-use.htm
[1]https://www.start.umd.edu/pubs/START_AmericanTerrorismDeaths...
Everyone knows that humans are terrible at estimating small risks of large dangers. The really scary thing is that humans are also by-and-large incapable of updating based on that knowledge.
[1]: http://www.motherjones.com/politics/2012/12/mass-shootings-m... [2]: http://www.ntsb.gov/investigations/data/Pages/AviationDataSt... [3]: https://en.wikipedia.org/wiki/List_of_motor_vehicle_deaths_i...
The claim that reinstating prohibition would prevent all alcohol-related deaths is simply not based in reality.
If you want to know what kind of black market arises when you prohibit a popular drug, you have multiple examples from recent history to look to. Such as the drug war, and the last time we tried to 'save lives' with alcohol prohibition.
The first step to making a convincing argument is acknowledging reality: the proliferation of hard encryption will alter the balance between individuals and the state.