Operating System design and kernel

[RedSon]

Why am I having sudden flashbacks?

[AmberJain]

Why am I having sudden flashbacks?

Well, I probably can tell answer to this question.

In the thread started by VACEPROGRAMMER (titled "Building a new OS - who wants to help?") (click here), a lot of rubbish had happened in that thread and so you can't be blamed for having "SUDDEN FLASHBACKS".
But I assure you (atleast from my side) that the mistake will not be repeated here in this thread. And more importantly you don't need to worry this time about hodge-podge that happened in VACEs' thread and amount of your effort in that thread (I say this because there are 46 times your name has appeared in Vace thread + one in top forum contributor i.e. it counts for 47 times your name appeared in page of vace's thread).

THANKS......

[RedSon]

But I assure you (atleast from my side) that the mistake will not be repeated here in this thread.

Good lets hope so, because already some of the replies to your original post in this very thread are authored by our not so good friend VACEPROGRAMMER. Don't worry he will be banned shortly.

BTW, I checked you out too ambr and so far you seem legit so at least you have that going for you.

[AmberJain]

I'm seriously interested in writing new os @ambernewlearne r

First of all sorry for the late reply.

Now let's come directly to the point. I am planning to write an OS. Below is a record of what I wish to follow. This record is just a part of my present thinking. I may change my mind in future about it's design, but for the timebeing, this is the design of OS I wish to write.

_NOTE_ It is easy to say that-"I'm writing an OS with 3D icons and a GUI with many advanced features....... ........". But come on, let's face the truth. Writing something like a OS which resembles even "Windows 95" (yeh..I know it's an old OS) will require a lot of hardwork, debugging of code and more importantly "mind power" i.e. the ability to write good code which is important for a good OS. It's easy to boast for what you cannot write (or better said-It's easy to boast for what you dont know "how to write or code"). So let's face the hard earth i.e. we cannot code a GUI OS (with advanced features) very easily and so if our luck favours (although I dont believe in luck) then in sometime in future we may take our OS to better heights than what we can imagine right now. So right now, my OS would be a simple OS with CUI.

The OS would be coded in C, C++ and Assembly language. It would be a simple OS with CUI (Character User Interface). For a look and feel of OS, you can google for "freedos" and download it to have a hand at it. Although it's look and feel would resemble FreeDOS, still it would have many unique features. The OS would be open source and free. You may call it an educational OS (but who knows it may reach greater heights than just remaining an educational OS).

Right now I'm concentrating on C and C++ programming. Then I will advance to assembly programming.
Also, one cannot write an OS just by learning programming. Instead one must also understand the Operating System Design. And therefore one must read (and understand) good books related to OS (that constitutes to reading atleast 1000 pages). I'm going to read some books as advised by JosAH in this thread while I'm am also learning C/C++/ASM. So this will take a minimum of two year on my part to read different books. This means I will start writing my OS really after atleast a period of two years.
I know there would be more to be faced but this all is a brief discussion about what I expect to face.

Also, I'm not going to call for any open source developers to contribute to OS development till I have released version 0.9 of that OS. Still if someone is willing to help, he is always welcome. I have also decided the name of OS but I'm not disclosing it now.

There would be 0.1 to 0.8 i.e. 8 versions before version 0.9. Version 0.9 would be the first to be released online for testing and better debugging by Open source developers. Version 0.9 can be thought of a beta version. Version 1.0 would be first stable release of OS.

Now, if you agree to hardwork with me and also agree to above all mentioned details, then PM me on bytes.com.

No more replies are expected in this thread as it would be a waste of time and energy on part of moderators to read, maintain and manage this thread again and again as in case of MikeJaskov(We all know this thread is going to be nowhere for atleast next two years).

THANKS.....
ambrnewlearner

[TamusJRoyce]

lately I've been pretty fond of reading ReactOS Kernel, simply because they have it available for few IDE's to open up ("make cb" for codeblocks). Makes life easy. And it's not so huge that you can't decipher and parse it like the linux kernel.

Even has a quick ability to compile into a liveCD.iso and test it with qemu or vmware.

But as for what to develop your own OS on, I would start with either vmware or qemu since you can't harm your computer using (well, harder to damage system nowdays, but maybe writing to the hard drive controller could be bad :P). Then move it to an old pc to test.

ps. If using mingw, beware of the version you use. I was using version 3.2 in college and I lost a letter grade writing the floppy disk controller software of a pre-built OS to read fat12 filesystems for the floppy disks, because mingw had a problem with it's C standard for reading files...