>“Work it out” is generally a demand to invent new mathematics, but sadly, mathematics doesn’t work like that The article invalidates a somewhat reasonable poi

>“Work it out” is generally a demand to invent new mathematics, but sadly, mathematics doesn’t work like that The article invalidates a somewhat reasonable point by saying this sort of thing. A cryptographic end-user application is not "mathematics". It's a piece of software, running on a piece of hardware, and there is no platonic, infallible security going on. This is the sort of 'no' people utter if they have ideological objections disguised as technological ones. In reality any system, including cryptographic ones exist on a curve. Differentiated systems for access exist. The honest criticism would be that a system with a backdoor is less secure, but it's certainly possible to enable privileged access to third parties while excluding others, it's just riskier. But risk in reality also exists with any encrypted application, because it runs with keys stored on a phone, not in some untouchable maths dimension. How much you move between access and security is absolutely a question of policy and architecture, not some theoretically impossible thing.