OSDev.org

Hello everyone,

I am looking for people to co-author a (probably self-proclaimed) UNIX-like operating system with me. It's just for sh*ts and giggles. I also want it to be completely free of any code written by others not in the eventual group of devs, for example a shell like bash or ksh will not be used in a possible release. All that will be written from scratch. I also won't be a leader, or a spokesperson, everything will be discussed in group.

Things that aren't welcome in a possible co-author:
- Supports monolithic kernels
- Loves the video of Richard Stallman (awesome guy btw) eating something he peeled off of his foot

Things that ARE welcome in a possible co-author:
- Supports microkernels
- Loves reincarnation servers (MINIX 3 for life, seriously)
- A knowledge of C and Assembly (preferably the Intel syntax)

I don't expect the eventual flood of developers that Linus Torvalds got, nor do I want one, because, bloat.

The eventual license doesn't matter, although I lean towards the GPL in general.

Greetings,
SoLDMG

Read: http://wiki.osdev.org/Beginner_Mistakes

Especially regarding recruiting.

I read the Beginner Mistakes many times. I know what I'm doing, although I appreciate the concern.

Could you let us know what particular attributes you have that make you a good co-ordinator for such a large-scale project. That way we can judge whether it is worth consideration. After all, I wouldn't want to waste my time on a project that is doomed to fail from the start.

I'm guessing the first thing that you will be working on, to be independent of all code not written by your group, will be an assembler, C compiler (do you plan to use C or some other language?), and a linker. Can we take it that you have already written these essential utilities?

I look forward to hearing about your progress to date.

SoLDMG wrote:I read the Beginner Mistakes many times. I know what I'm doing, although I appreciate the concern.

Could you give your response to those four points in Beginner's Mistakes, just so that we can be confident that you know what you are doing?

1. If you have no established codebase, people will not join because they can see you lack experience and expect the project to fail.
Where can we view your established codebase?

2. If you lack a (worked out) design, people will not join you because they can't see how your OS is more interesting than their own design.
Where can we see your worked out design?

3. If your reputation doesn't precede you, especially the more experienced people will be very wary of you and lack the trust to join.
Can you give any references to establish your reputation?

4. If you don't have project management skills, the few rare people that do join will quit shortly because they are discussing stuff and do not get to code.
What background do you have in project management?

iansjack wrote:Could you let us know what particular attributes you have that make you a good co-ordinator for such a large-scale project. That way we can judge whether it is worth consideration. After all, I wouldn't want to waste my time on a project that is doomed to fail from the start.

- I don't really have any attributes that make me a good co-ordinator, which is why I don't want to be a leader of sorts, and why I want everything to be in a group so people can discuss and possibly work together on things, so everything evolves naturally instead of having some dictator screaming at people to get stuff done and not to complain. I do however will step up to lead the project into this direction in the beginning.

iansjack wrote:I'm guessing the first thing that you will be working on, to be independent of all code not written by your group, will be an assembler, C compiler (do you plan to use C or some other language?), and a linker. Can we take it that you have already written these essential utilities?

- I have not yet written/completed these utilities, although I did some work on an assembler which was able to assemble for a toy VM of a friend quite some time ago (the source code has been lost in forgotten Dropbox accounts). The experience I gained will help in this project though. I do plan to use C wherever possible in both the kernel and userland, although the bootsector and such will be written in Assembly because they have to be written in it.

1. If you have no established codebase, people will not join because they can see you lack experience and expect the project to fail.
Where can we view your established codebase?
- I can link to some prototype bootcode that I have on Dropbox that could be put up for download if you want to see what I've done. I can't at the moment though because I'm on my phone.

2. If you lack a (worked out) design, people will not join you because they can't see how your OS is more interesting than their own design.
Where can we see your worked out design?
- The design is to be discussed in the early stages of development. In the meantime work can be done on the utilities like a C compiler or userland commands.

3. If your reputation doesn't precede you, especially the more experienced people will be very wary of you and lack the trust to join.
Can you give any references to establish your reputation?
- No, sorry.

4. If you don't have project management skills, the few rare people that do join will quit shortly because they are discussing stuff and do not get to code.
What background do you have in project management?
- None.

I know my ways aren't the best, but I just want to have some fun doing what I like to do alongside people who also like what I like. Which is screwing around doing low-level programming to accomplish stuff on the computer.

So basically, you just made all the mistakes.

Good luck then, you'll need it.

Thank you. I think that answers my concerns.

I don't think I'll be joining your team as you appear to me to have made four classic mistakes when it comes to recruiting.

Combuster wrote:So basically, you just made all the mistakes.

Good luck then, you'll need it.

Thanks.

iansjack wrote:Thank you. I think that answers my concerns.

I don't think I'll be joining your team as you appear to me to have made four classic mistakes when it comes to recruiting.

Okay then. I hope you'll find a different project to work on (if you were looking for one).

If anyone wishes to follow my progress, check out https://github.com/SoLDMG.

SoLDMG wrote:UNIX-like operating system with me

......

- Supports monolithic kernels

......

- Supports microkernels

.....

- Loves reincarnation servers (MINIX 3 for life, seriously)

operating system != kernel.

if i would write an os on x86, i would probably not even use protected mode, to alow me to just put together hardware access and memory management in a few hundred line kernel within days, to let me focus on the most important things, like on operating system itself (paint, freecell, web browser, cat dresser for little girls v 0.5, pantsu remover)

Geri wrote:if i would write an os on x86, i would probably not even use protected mode, to alow me to just put together hardware access and memory management in a few hundred line kernel within days, to let me focus on the most important things, like on operating system itself (paint, freecell, web browser, cat dresser for little girls v 0.5, pantsu remover)

It's fairly simple to enter identity mapped protected mode (even long mode) and do everything else just a simply as real mode. Put your drivers into libraries. Have a working OS in days.