Re:Select Code_execution from * Using SQLite; (2019)
by ngneer on Saturday June 13, 2020 @04:35PM
(#23512457)
No, there is no difference. Locks and software implement or abide by control laws. In your analogy, if you give perfect software an input and it behaves as expected, then that is the same as walking up to a locked door, trying to open it and failing. Both systems are susceptible to control bypass.
[ Parent
| on HN ]
Re:Select Code_execution from * Using SQLite; (2019)
by anyfoo on Saturday June 13, 2020 @05:21PM
(#23512844)
The point is that for some programs[1], you can prove that all inputs will lead to an output with certain properties. With a physical lock, you cannot.
For physical locks, you cannot even rigorously define the "input".
[1] Not always, unfortunately, if your language is turing complete. You might be able to prove weaker, but still very useful statements, like "no input will cause a point to be dereferenced out of bounds".
Re:Select Code_execution from * Using SQLite; (2019)
by ngneer on Sunday June 14, 2020 @02:02PM
(#23519770)
Yes, you may prove such a statement about a program. But you will not be able to prove that about the physical system where the program is running (assuming you want to run your program and not just reason about it as a mathematical object). If you need an example, consider Plundervolt.