OSDev.org

I like the idea, although I'm (a bit of an) OS-n00b myself.

I don't have much respect for commercial products (Like DOS/win32)

From a developers point of view, I don't like the design of DOS or Windows.. It's disgusting!

I have to say the BeOS was pretty neat looking before the company went bust, Haiku is currently doing a good job at making a 100% free alternative though.

I myself came to this forum.. not because I hate other OS's.. but because I was looking to see how many people are working on Unix-like operating systems. (Everything else... is just a toy )

Currently the only notable systems by people here are:
Mort-OS by spix.
The Spoon Microkernel by Durand Miller.

Just because I'm interested in this as a hobby, doesn't mean all other OS's are inadequate... OpenBSD rocks

@Brynet-Inc, If someone offers you a gift and you do not except it, who does the gift belong to ? .

Yes, it's called Trion and it's been around since 2003.
http://trion.sourceforge.net/

Don't know about osdev.org, but over at MT that idea has surfaced repeatedly. I don't think it can be done; the ideas of what an "ideal" OS should be like are far too varied.

You cannot even get people to agree on the points programming language, license used, micro/macrokernel, release policy, GRUB vs. custom bootloader, and top-down vs. bottom-up approach. And that's before the work has even begun...

Not to mention whether the system will be object oriented or properly organized. Or whether it contains lisp concepts like meta-evaluator, closures or continuations as first class thingies.

I think there would be a solution to get osdevers co-operate.

First thing you need is an extensible and portable but relatively simple programming language designed for assembler programming. The language should also support modularity. Then, every osdever should acknowledge the existence of others and co-operate to some point: Always when you do a concept, say... you handle IRQ modes new way or you have a different I/O interface, you could position this to a code example repository where other osdevers could then read your example in that common language and totally advance it to their own use, maybe on totally different platform!

This way we should only agree that extensible and portable but relatively simple programming language is something we all want, and we could co-operate!

I don't really think you will get people to agree on a common design and then get them to start working on it.
What might work would be to develop the basic kernel (of course how basic is up to you) and then make the source available to all for development. (Maybe getting people to join after that would be easier)

And then beat the crap out of all the other OS'es available currently.

Anyway I think this is how Linux really became what it is today.