Page 1 of 1

An exo-kernel

Posted: Fri Jan 16, 2009 5:06 am
by quanganht
I'm going to develop a exo-kernel, which is expected to be fully functional, along with a libOS and some apps. Is there anyone wants to join me ?

Re: An exo-kernel

Posted: Fri Jan 16, 2009 7:28 am
by yemista
tell me more about this, exo-kernel

Re: An exo-kernel

Posted: Fri Jan 16, 2009 7:46 am
by quanganht
The main big idea about this kind of kernel is eliminating as much as possible abstractions from kernel. All it's work is securely binding hardware resources to applications, and let them deal with it. Everything like stdio, syscalls etc. now implemented in libOS(es), and variant between applications.

Re: An exo-kernel

Posted: Fri Jan 16, 2009 9:35 am
by Troy Martin
Wrong forum.
EDIT: Thanks mods.

Re: An exo-kernel

Posted: Fri Jan 16, 2009 9:40 am
by Solar

Re: An exo-kernel

Posted: Sun Jan 18, 2009 5:36 am
by Combuster
Sorry, writing my own (for the past years).

Re: An exo-kernel

Posted: Sun Jan 18, 2009 5:52 pm
by quanganht
I see. Maybe I have to do it on my own too

Re: An exo-kernel

Posted: Wed Jan 21, 2009 4:28 am
by quanganht
Hello? No one ??? :(

Re: An exo-kernel

Posted: Wed Jan 21, 2009 5:56 am
by Combuster
Apparently...

Not that everybody else saw that coming :wink:

Re: An exo-kernel

Posted: Wed Jan 21, 2009 9:36 pm
by Love4Boobies
I'm not sure why you find exokernel development so different from any other type of kernel interface...

Re: An exo-kernel

Posted: Thu Jan 22, 2009 12:35 am
by Wilkie
quanganht,

As a developer of an exokernel, I'd have to say that it is hard to get a group of people together with the same devotion as yourself. You get some people who are passionate about the idea, but not committed to the gritty work ahead. However, as "simple" as an exokernel sounds on the surface, it took a group of around four a period of 2 years to roll out the core of our exokernel, also slated to be fully-functional. (btw, thank you OSDev for your help on a couple of tough bugs!) Therefore, to be reasonable about completing your project, you need help!

The design and vision of the project is key to attracting contributors. Your passion must be their passion. Our vision was to write an exokernel that was forward-thinking and not legacy-driven. Perfect for an exokernel where the minimal abstraction tends to be static while the underlying technologies are ever-changing. We are admittedly broad in design, and we understand the repercussions of such a stance. But, overall, an exokernel is a very general term. What problem are you going to attempt to tackle with such power? Answer this, and maybe you will find somebody who also believes in this cause and will join you.

Unless you believe in our cause, because you can always join us. 8)

--
Wilkie

Re: An exo-kernel

Posted: Thu Jan 22, 2009 7:57 am
by quanganht
Wilkie wrote:quanganht,

As a developer of an exokernel, I'd have to say that it is hard to get a group of people together with the same devotion as yourself. You get some people who are passionate about the idea, but not committed to the gritty work ahead. However, as "simple" as an exokernel sounds on the surface, it took a group of around four a period of 2 years to roll out the core of our exokernel, also slated to be fully-functional. (btw, thank you OSDev for your help on a couple of tough bugs!) Therefore, to be reasonable about completing your project, you need help!

The design and vision of the project is key to attracting contributors. Your passion must be their passion. Our vision was to write an exokernel that was forward-thinking and not legacy-driven. Perfect for an exokernel where the minimal abstraction tends to be static while the underlying technologies are ever-changing. We are admittedly broad in design, and we understand the repercussions of such a stance. But, overall, an exokernel is a very general term. What problem are you going to attempt to tackle with such power? Answer this, and maybe you will find somebody who also believes in this cause and will join you.

Unless you believe in our cause, because you can always join us. 8)

--
Wilkie
Thank you. I'll think about that. But learning D requires more time and I don't want to waste it. But , yeah, when I feel comfortable enough with D, I will join you