OSDev.org

I was working on my OS with someone, now they quit and I am continuing alone.

OS developing is not fun nor easy with one person.

So basically I'm looking for someone else to work with.

Here's where I am:
Using a basic GRUB Bootloader
Have Basic Video driver (text, cursor movement)

I know this is close to the beginning of developing however,
if someone else is also here that would be awesome.

If you would like to be my partner or make a Dev team contact me (pm, or post)

Thanks

Move this post to Announcements, Test Requests, & Job openings

done.

I'd be glad to contribute once I have a significant amount of code in my OS (I just started writing the kernel this weekend), but I won't be able to be part of the dev team per se. Exams are starting and I'll have essays and such going on for the next two and a half weeks.

GeniusCobyWalker wrote: OS developing is not fun nor easy with one person.

But... but... nobody arguing about your design decisions!

Wilkie wrote:

GeniusCobyWalker wrote: OS developing is not fun nor easy with one person.

But... but... nobody arguing about your design decisions!

That's the main reason I don't have a partner.

Yeah, I find that at least kernel level programming is best done by one person. Once you have a working kernel, you can assign different people to write different userland programs pretty easily. If you use a microkernel design, you can also do a similar thing with drivers and servers, so if you have lots of friends who can program, you can probably get one of those up and running faster even with the extra complexity.

So is it not helpful to have a partner at the beginning?
(Basically a 1 person job?)

GeniusCobyWalker wrote:So is it not helpful to have a partner at the beginning?

I'd say it is very helpful. It is the perfect thing for some peer programming. Especially when debugging seriously complicated problems

Not that too many things can be done in parallel and completely independent of one another at the early stages. But you need that second pair of eyes.

Well, it's undoubtedly useful to have another person for debugging, which is usually most of the time spent, but for the general design and initial implementation, I think it's better for there to be one person. In a kernel, things are very dependent on each other, and state is everything, so you need to understand every part of the codebase and all of its subtleties to work on the core services and routines. When you write that core, you already know how it works, and there are no misunderstandings. Of course, that part doesn't make up that much code, so I guess it is probably better to have multiple people. Also, I still know my kernel like the back of my hand - it's only 900 SLOC (but fully functional) - so maybe my view is limited.

NickJohnson wrote:Well, it's undoubtedly useful to have another person for debugging, which is usually most of the time spent, but for the general design and initial implementation, I think it's better for there to be one person. In a kernel, things are very dependent on each other, and state is everything, so you need to understand every part of the codebase and all of its subtleties to work on the core services and routines. When you write that core, you already know how it works, and there are no misunderstandings. Of course, that part doesn't make up that much code, so I guess it is probably better to have multiple people. Also, I still know my kernel like the back of my hand - it's only 900 SLOC (but fully functional) - so maybe my view is limited.

When my group initially fell apart, I found myself as the sole developer. I am a very strong programmer with years of experience, but I am also a complete moron! This is only found to be true after leading in design decisions and not being opposed. You need someone to oppose you. I spent some time after that to regroup the best of the best and I will push them to help me. I lost a lot of ground thinking in the mindset of a programmer of one. There is no such thing.

After the kernel has been written, all of the supporting components will undoubtedly be written by other people. It just has to be this way. You will need multiple people who can understand this codebase to aid in this explosion of development. Unless you document heavily, but it is still a stretch... and who does that? You should probably be documenting a bit more yourself for this exact reason.

But these things are something you just have to trust me on, or make the same mistake.

I´m also near the beginning, so I´d like to join you. But I also write an OS, and I don´t have time for two OS-projects
btw I´m also searching for people interested in helping me.

If you want to take a look at my OS: http://darkinsanity.netne.net/nerdpole/ ... rdPole.img and http://sourceforge.net/projects/frostkernel

My OS already has Software-Multitasking and Paging, and I´m working on a Floppy-driver at the moment.

darkinsanity wrote:btw I´m also searching for people interested in helping me.

It is almost like we need a social networking site just for kernel devs. MyKernelSpace.com? Too much pun?

Wilkie wrote:

darkinsanity wrote:btw I´m also searching for people interested in helping me.

It is almost like we need a social networking site just for kernel devs. MyKernelSpace.com? Too much pun?

Cant, thats taken already. I cant read any of it though

Hey, www.sourcebook.org isn't taken! And I betcha we could get another .org cheap!