4 ms·
What's needed for that is the U.S. Navy calculus book from WWII. The Navy needed people to understand calculus because navigation and fire control systems are d
by Animats 4y ago
What's needed for that is the U.S. Navy calculus book from WWII. The Navy needed people to understand calculus because navigation and fire control systems are devices that do integration, analytic geometry, and trigonometry. Since they still do that, with more software and fewer gears, there's a current version of the book.[1]
From the book: "The proofs of theorems shown in this section will be omitted in the interest of brevity."
You need calculus for anything that involves calculating the physics of the real world. But not as a proof system.
The same is true of geometry as taught in schools. There's an emphasis on proofs because those were a big deal when geometry was being invented. But if you need to solve geometry problems, you seldom if ever do proofs.
I've worked on proof of correctness systems, and there are times when you need heavy formalism. But it's a means to an end, not an end in itself. The end is programs that don't fail, not more proofs. I used to argue this with mathematicians when I was working on that sort of thing. There's still a tendency in software proof of correctness to fall in love with the formalism as an end in itself.
[1] https://archive.org/details/US_Navy_Training_Course_-_Pre-calculus_and_Introduction_to_Probability https://archive.org/details/US_Navy_Training_Course_-_Pre-ca...