KolibriOS - an x86 operating system written in FASM

[the.weavster]

KolibriOS is an x86 operating system written in FASM, it's tiny and blisteringly fast. The developers missed out on being included in the Google summer of code so they've gone for a kickstarter to try and raise the funds themselves.

If any of you have a few pounds to spare please give it a look: KolibriOS Summer Of Code

I'm not associated in any way, I just tried KolibriOS after spotting it on distrowatch and it was so fast I found myself wishing it was more complete so it was properly usable.

[es_91]

Thx for this interesting stuff.

[Kiffi]

here is a instruction how to set up a virtualbox with KolibriOS:

http://wiki.kolibrios.org/wiki/Setting_up_virtualbox

Greetings ... Kiffi

[Nituvious]

I love homebrew operating systems! Thanks for sharing.

Another cool one is ReactOS. It's borderline useless right now but it is a very neat toy.

[tj1010]

It takes a few hours tops to code and test a kernel and bootloader, and write it to an MBR or overwrite MBR. If something has only core functionality and almost no hardware support, it's pretty much useless to anyone. BSD and Linux were created for people who wanted custom operating systems with stability, functionality for any environment, and hardware support.

This is only obvious when you install an OS and literally can't do anything with it.. The remote potential for stability and basic productivity years from now doesn't cut it and can't be properly developed and tested in a time-span that makes it sensible given what already exists..

I seen ReactOS in some banner on tuts4you. They basically made it load some PE with subsystem emulation, so it won't be just like the other homebrew OSs with no software.

[MachineCode]

you install an OS and literally can't do anything with it

That's the problem with all these homebrew OSes like KolibriOS: they're nice to look at, but will never go anywhere or become accepted by the masses. They're just made for bragging rights or to show off their coders' skills. Big deal.

ReactOS is different because its aim is to run the Windows software that you already own, so it has great potential, and indeed it runs of a lot of Windows software already. That, coupled with the fact that a lot of people still hate Microsoft, gives it a bright future, because they get to run their Windows software without Microsoft "owning" them.

[LuCiFeR[SD]]

ReactOS is different because its aim is to run the Windows software that you already own, so it has great potential, and indeed it runs of a lot of Windows software already. That, coupled with the fact that a lot of people still hate Microsoft, gives it a bright future, because they get to run their Windows software without Microsoft "owning" them.

they've had "potential" for more years than I care to remember... and have stagnated for pretty much the whole of that time. You are better off running your choice of *nix based OS and wine.

[tj1010]

IMO it probably is impressive if you don't know how easy it is to program directly with hardware compared to API structures and ISO specs..

If you want to do something to impress people who actually have a clue, you need to do algorithm or security development, which isn't something even a talented person typically does on a lunch break..

[MachineCode]

Lucifer, yeah, I've given up waiting for ReactOS too. But my point was that it's a better project than starting an OS from scratch that has no ready software base. Yes, kudos to KolibriOS and all that, but it won't go anywhere at all. I've seen lots of alternative OSes come and go over the years, but at least ReactOS is still plodding along... slowly.

[LuCiFeR[SD]]

IMO it probably is impressive if you don't know how easy it is to program directly with hardware compared to API structures and ISO specs..

If you want to do something to impress people who actually have a clue, you need to do algorithm or security development, which isn't something even a talented person typically does on a lunch break..

If you want to impress people, you put your money where your mouth is. It really is that simple. As for security development being impressive.... no, no it really isn't. If I wanted to be impressed by a bunch of sweaty men waving their willies around... I'd go watch The full monty

Lucifer, yeah, I've given up waiting for ReactOS too. But my point was that it's a better project than starting an OS from scratch that has no ready software base. Yes, kudos to KolibriOS and all that, but it won't go anywhere at all. I've seen lots of alternative OSes come and go over the years, but at least ReactOS is still plodding along... slowly.

Don't worry MC, I understood... and my apologies if what I wrote "felt" like I was trying to take a pop at you... that was not my intention . I agree with you hehe.

[Thorium]

IMO it probably is impressive if you don't know how easy it is to program directly with hardware compared to API structures and ISO specs..

If you want to do something to impress people who actually have a clue, you need to do algorithm or security development, which isn't something even a talented person typically does on a lunch break..

Obviously you didnt tried coding a OS yourself.

Coding a simple kernel is easy. Coding a OS with all the features of KolibriOS (MenuetOS) is a huge undertaking. Try code a OS with preemptive multitasking, GUI using VESA, USB-, webcam- and printer-support, TCP/IP stack, irc-, http-, nntp- and tftp- clients, and file system access for different media and formats. While designing your own ABI, driver system and executable format, thats not based on another OS.

And then you know doing that is a ton of work and cant be compared with using API like winsock or GDI. Actualy using API is like taking a holiday in comparision.

Sure the OS will never be big, but it is a impressiv work.

[tj1010]

IMO it probably is impressive if you don't know how easy it is to program directly with hardware compared to API structures and ISO specs..

If you want to do something to impress people who actually have a clue, you need to do algorithm or security development, which isn't something even a talented person typically does on a lunch break..

Obviously you didnt tried coding a OS yourself.

Coding a simple kernel is easy. Coding a OS with all the features of KolibriOS (MenuetOS) is a huge undertaking. Try code a OS with preemptive multitasking, GUI using VESA, USB-, webcam- and printer-support, TCP/IP stack, irc-, http-, nntp- and tftp- clients, and file system access for different media and formats. While designing your own ABI, driver system and executable format, thats not based on another OS.

And then you know doing that is a ton of work and cant be compared with using API like winsock or GDI. Actualy using API is like taking a holiday in comparision.

Sure the OS will never be big, but it is a impressiv work.

Thanks for the information. It's nice to be schooled on computer science by someone with so many accomplishments in the field of programming, or even computers in general..

I think I'll stick to respecting criticism from people who actually work in the field or have finished something..

[Demivec]

Obviously you didnt tried coding a OS yourself.

Thanks for the information. It's nice to be schooled on computer science by someone with so many accomplishments in the field of programming, or even computers in general..

I think I'll stick to respecting criticism from people who actually work in the field or have finished something..

@tj1010: So, have you tried coding an OS yourself?

[tj1010]

Obviously you didnt tried coding a OS yourself.

Thanks for the information. It's nice to be schooled on computer science by someone with so many accomplishments in the field of programming, or even computers in general..

I think I'll stick to respecting criticism from people who actually work in the field or have finished something..

@tj1010: So, have you tried coding an OS yourself?

I've done RTOSs, and application OSs using OSI. I did it when I was a student learning about architecture and memory management.

Given the complexities of hardware interfaces, and their obscurities, these OS projects are like carving a wooden boat out of a wooden boat, with a dull knife. This is why Linux and BSD exists...

[Thorium]

IMO it probably is impressive if you don't know how easy it is to program directly with hardware compared to API structures and ISO specs..

If you want to do something to impress people who actually have a clue, you need to do algorithm or security development, which isn't something even a talented person typically does on a lunch break..

Obviously you didnt tried coding a OS yourself.

Coding a simple kernel is easy. Coding a OS with all the features of KolibriOS (MenuetOS) is a huge undertaking. Try code a OS with preemptive multitasking, GUI using VESA, USB-, webcam- and printer-support, TCP/IP stack, irc-, http-, nntp- and tftp- clients, and file system access for different media and formats. While designing your own ABI, driver system and executable format, thats not based on another OS.

And then you know doing that is a ton of work and cant be compared with using API like winsock or GDI. Actualy using API is like taking a holiday in comparision.

Sure the OS will never be big, but it is a impressiv work.

Thanks for the information. It's nice to be schooled on computer science by someone with so many accomplishments in the field of programming, or even computers in general..

I think I'll stick to respecting criticism from people who actually work in the field or have finished something..

You seem to know me well.

It doesnt matters who tells you that you write wrong stuff. Its still wrong and just a result of beeing jealouse.

No need to reply you are on my ignore list now.