zlacker

the only security paradigm that is 100% foolproof is to assume breach. Taking away any number of users freedoms, big or small, does not change that.

Old computers, before sandboxing and Windows defender and real-time protection, were more secure, because people were less likely to plug their bank account information, social security number, birth date, and home address into them.

At a certain point we have got to level with the idea that a smartphone is no longer a general purpose computer in your pocket. It's more like a cyber passport. It knows everything about you and authenticates formal activities.

>>mouse_+(OP)
Sure, but by the same logic old roads were safer because we used horses on them instead of cars and a horse won't generally plow into the oncoming lane if you fall asleep at the reigns.

It feels like this analysis really downplays some advantages making sandboxes and Windows defender and realtime protection got us in the average case (even if in the edge case someone can get hurt).

>>mouse_+(OP)
> Old computers, before sandboxing and Windows defender and real-time protection, were more secure, because people were less likely to plug their bank account information, social security number, birth date, and home address into them.

So they weren’t actually more secure – they were less secure and less useful (setting aside the questionable historical accuracy of where popular online banking sits in the timeline relative to OS security measures in that claim). Maybe if we relax the made up constraint that a change must create 100% foolproof security, we can have a more nuanced discussion about ways to improve security.