[Cryptography] Speculation considered harmful?

[jamesd at echeque.com]

On 1/9/2018 3:00 PM, Henry Baker wrote:
> The early time-sharing systems research concluded that there is a hard limit to how much "efficiency" you can get out of shared resources, after which the competing processes damage each other more than they gain.

We are moving towards a sea of machines connected by a network - shared 
nothing message passing.

This reduces the problem to something manageable.