3 ms·

In my experience, writing "scary" software like a compiler, an OS, a kernel driver, etc is more about bravery/daring than raw skill or intellect. There are quit

by twtw 8y ago

In my experience, writing "scary" software like a compiler, an OS, a kernel driver, etc is more about bravery/daring than raw skill or intellect. There are quite a few people that know C and C++ in high school, so it seems reasonable that at least some of them would have someone (IRL or online) nudge them in the direction of an OS and tell them "it's not as hard as you think, if you dare to try." I don't say this to diminish the accomplishment here, but to encourage others to try it too.

ptr 8y ago
I agree wholeheartedly, having written a kernel myself first year in high school (and I don’t consider myself very intelligent). Mine was even more involved — it had user space, protected memory, multitasking, file systems, etc. So much fun! What it came down to was persistence and reading the manuals.
rblion 8y ago
I don't know C but I am interested in learning. What would be a good place to learn the steps of making an OS? I just want a bird's eye view right now, just to understand the moving parts. This will also help me in the future as a manager.
- ptr 8y ago
  https://wiki.osdev.org https://wiki.osdev.org
- aswinmohanme 8y ago
  I followed this http://www.osdever.net/bkerndev/Docs/intro.htm http://www.osdever.net/bkerndev/Docs/intro.htm with some pointers from osdev wiki
  - DarmokJalad1701 8y ago
    Nice! I remember following this site back in the day. My tiny "OS" got till paged memory management and didn't really go any further. I used to use "Bochs" CPU emulator for testing.
    - necovek 8y ago
      Right, I used bochs too: it was a full emulator, none of this "let's just let the CPU really run these" things. :)
- twic 8y ago
  In Rust rather than C, but a good walk through the basics of getting an OS booted: https://os.phil-opp.com/ https://os.phil-opp.com/
- twtw 8y ago
  In addition to the resources other have mentioned, I've found this http://pages.cs.wisc.edu/~remzi/OSTEP/ http://pages.cs.wisc.edu/~remzi/OSTEP/ valuable.
aswinmohanme 8y ago
Totally agree, it was all about persistence way more than actual coding. And if you know where to look it's way easier than it sounds.
spicymaki 8y ago
Much of the willingness to attempt to write “scary” software when you are young is being ignorant of the complexity involved. That is probably why younger people are so disruptive in tech as well. Older you get the more reasons why not to do something appears.
- pjmlp 8y ago
  “They did not know it was impossible so they did it” -- Mark Twain
- Waterluvian 8y ago
  I learned programming faster when I had no clue how awful I was doing it. I just kept going and bumping into things a lot. Now learning a new language is hard because I'm trapped in thus mindset of, "surely this isn't the most time efficient way to learn this..." Only recently have I made some progress by not looking back and coding myself into a wall even if I think I can see the wall coming.
- baroffoos 8y ago
  When I was shit at programming I started a lot more ambitious projects and I ended up finishing none of them. Now I know what to expect and I end up with something finished and useful most of the time. I suppose sometimes beginners attempt things that experts think are too hard but it ends up that you can make a workable solution even if it is structured horrible.
pjmlp 8y ago
I also learned C and C++ back in the late 80's during high school, as part of the domain area we get to chose on the last three years (10 - 12). We also got introduction to UNIX via Xenix on our OS classes, got to do a Pascal syntax highlighter and a couple of other cool things. Fully agree with you, the accomplishment here is great and anyone else wondering about if they should attempt something like this, just give it a go. Don't worry about writing perfect code, rather make it work somehow, perfection will come later.
- nickpsecurity 8y ago
  "Don't worry about writing perfect code, rather make it work somehow, perfection will come later." I think your advice is true, esp for education and Worse is Better effect. I think the follow-up is wrong for two reasons: 1. Useful code tends to stick around long time getting extended more than refactored for quality. 2. Due to the methods involved, the highest, quality designs usually need to be designed for that upfront. Certain decisions will favor reaching the goal. So, my advice for high-quality outcome is to design for it with simple constructs, straight-forward composition, and plenty of docs made on the way. Gives someone a better start if they want to try to perfect it years down the line.
allenu 8y ago
I think you can make a similar statement about coding in general. For me, I got into coding because it was fun and I worked on projects (games) when I was in high school because I wanted to create something. The code I wrote was terrible but I kept at it because I really wanted that end result (a game of my own creation). If someone had told me it was hard, I probably would have laughed because it was mostly about persistence.
necovek 8y ago
I've did this myself back in 2000/01 — there were fewer resources on the web back in the day (and dial-up was harder), so it was about hunting down freely available papers, books, source code and similar (a student in Serbia can't really walk into a library and get their hands on some real literature). I actually did have a round-robin scheduler with prioritised processes and first-fit memory allocator in there as well. During the presentation, as I pulled out a 3.5" floppy disk to boot up my OS, the teachers looked at me in disbelief: you've actually got something working? If you are curious and willing to spend some quality time with your computer, this is a great way to learn about computers. Today though, I'd recommend people do it with ARM just for the kicks.

by twtw 8y ago

ptr 8y ago

I agree wholeheartedly, having written a kernel myself first year in high school (and I don’t consider myself very intelligent). Mine was even more involved — it had user space, protected memory, multitasking, file systems, etc. So much fun! What it came down to was persistence and reading the manuals.

rblion 8y ago

I don't know C but I am interested in learning. What would be a good place to learn the steps of making an OS? I just want a bird's eye view right now, just to understand the moving parts. This will also help me in the future as a manager.

ptr 8y ago
https://wiki.osdev.org https://wiki.osdev.org
aswinmohanme 8y ago
I followed this http://www.osdever.net/bkerndev/Docs/intro.htm http://www.osdever.net/bkerndev/Docs/intro.htm with some pointers from osdev wiki
- DarmokJalad1701 8y ago
  Nice! I remember following this site back in the day. My tiny "OS" got till paged memory management and didn't really go any further. I used to use "Bochs" CPU emulator for testing.
  - necovek 8y ago
    Right, I used bochs too: it was a full emulator, none of this "let's just let the CPU really run these" things. :)
twic 8y ago
In Rust rather than C, but a good walk through the basics of getting an OS booted: https://os.phil-opp.com/ https://os.phil-opp.com/
twtw 8y ago
In addition to the resources other have mentioned, I've found this http://pages.cs.wisc.edu/~remzi/OSTEP/ http://pages.cs.wisc.edu/~remzi/OSTEP/ valuable.

aswinmohanme 8y ago

Totally agree, it was all about persistence way more than actual coding. And if you know where to look it's way easier than it sounds.

spicymaki 8y ago

Much of the willingness to attempt to write “scary” software when you are young is being ignorant of the complexity involved. That is probably why younger people are so disruptive in tech as well. Older you get the more reasons why not to do something appears.

pjmlp 8y ago
“They did not know it was impossible so they did it” -- Mark Twain
Waterluvian 8y ago
I learned programming faster when I had no clue how awful I was doing it. I just kept going and bumping into things a lot. Now learning a new language is hard because I'm trapped in thus mindset of, "surely this isn't the most time efficient way to learn this..." Only recently have I made some progress by not looking back and coding myself into a wall even if I think I can see the wall coming.
baroffoos 8y ago
When I was shit at programming I started a lot more ambitious projects and I ended up finishing none of them. Now I know what to expect and I end up with something finished and useful most of the time. I suppose sometimes beginners attempt things that experts think are too hard but it ends up that you can make a workable solution even if it is structured horrible.

pjmlp 8y ago

I also learned C and C++ back in the late 80's during high school, as part of the domain area we get to chose on the last three years (10 - 12). We also got introduction to UNIX via Xenix on our OS classes, got to do a Pascal syntax highlighter and a couple of other cool things. Fully agree with you, the accomplishment here is great and anyone else wondering about if they should attempt something like this, just give it a go. Don't worry about writing perfect code, rather make it work somehow, perfection will come later.

nickpsecurity 8y ago
"Don't worry about writing perfect code, rather make it work somehow, perfection will come later." I think your advice is true, esp for education and Worse is Better effect. I think the follow-up is wrong for two reasons: 1. Useful code tends to stick around long time getting extended more than refactored for quality. 2. Due to the methods involved, the highest, quality designs usually need to be designed for that upfront. Certain decisions will favor reaching the goal. So, my advice for high-quality outcome is to design for it with simple constructs, straight-forward composition, and plenty of docs made on the way. Gives someone a better start if they want to try to perfect it years down the line.

allenu 8y ago

I think you can make a similar statement about coding in general. For me, I got into coding because it was fun and I worked on projects (games) when I was in high school because I wanted to create something. The code I wrote was terrible but I kept at it because I really wanted that end result (a game of my own creation). If someone had told me it was hard, I probably would have laughed because it was mostly about persistence.

necovek 8y ago

I've did this myself back in 2000/01 — there were fewer resources on the web back in the day (and dial-up was harder), so it was about hunting down freely available papers, books, source code and similar (a student in Serbia can't really walk into a library and get their hands on some real literature). I actually did have a round-robin scheduler with prioritised processes and first-fit memory allocator in there as well. During the presentation, as I pulled out a 3.5" floppy disk to boot up my OS, the teachers looked at me in disbelief: you've actually got something working? If you are curious and willing to spend some quality time with your computer, this is a great way to learn about computers. Today though, I'd recommend people do it with ARM just for the kicks.