OSDev.org

Hi friends ,
This might be stupid question . But I am just curious . How many people here actually develop kernels as a part of your job ? As far as I am concerned my real job is light years away from kernel development , i do consider kernel development as a hobby . How many of you are hobbist kernel hackers ( like me ) . But at some point in my carrier I want to be a kernel developer , and if God permits , i will be .....

Finally where did it begin and what attracted you to os development ?

Truly Evil
Saint Lucifer

Hi,

Programming is purely a hobby for me and I'm happy with it staying that way - I like a job where I'm interacting with different people all the time.

The reason I got interested in OS dev was that I wanted to know more about how computers work - what happens when you push that magic button on the front of the PC? I also planned to start writing in assembly language anyway, but wasn't really satisfied with how MASM interfaced with windows libraries - I wanted to work at a lower level.

I had done quite a bit of higher-level programming before (Windows RAD), but working in an environment with no pre-written software support seemed a challenge. After quite a bit of background reading (on bonafide and MT where I joined the forums but never posted), I had a go at my first bootloader and got addicted.

Cheers,
Adam

For eight years now I'm working professionally as software engineer, with C++ being my primary language. My first round of job applications aimed for OS development, but the big players (IBM, SuSE, RedHat) didn't want a newbie, and I torpedo'ed the job interview with the small player (Met@box) when I smelled too many fish. (The company went bancrupt soon after, including charges of fraud pressed against the CEO.)

So, no pro kernel development for me. But seeing the amount of pain that can be brought to bear by the comparatively "simple" systems that earn my salary, I can't really say I'm not content.

The idea to get involved in my own OS project was born on amiga.org, where - during an agitated discussion about where Amiga Inc. / Commodore Inc. / Amiga Technologies Inc. / ... went wrong and what would have to be done - someone asked me: "So what do you suggest, writing our own OS?". The rest ist history: The project started out ambitious and died some time after, and PDCLib is what remains.

I've been working in the IT-Realm for seven years now, and I have to say, that my scope of work has greatly shifted.

In the beginning, I've been doing COBOL-hacking on a BS2000 mainframe, but then, recognizing what I'm capable of (no, I am not so modest as to deny, that I'm capable and know my stuff), the bosses have quickly considered assigning me other stuff, so I've become a kind of jack-of-all-trades in our IT-Department. You know, someone who knows a lot, has the big picture in mind and quite some software projects under his belt already. Having attended a kind of College (Viennese People might recognize the Spengergasse) and still attending a University of Applied Science (field of study IS Software-Development and Management of Software Projects) helps of course with the advancement. Nowadays I'm hardly concerned with coding anymore but rather with project management stuff, leading and tutoring people - besides being the Software Configuration manager guy & Version management administrator. It's a shift - away from implementation towards higher level stuff indeed. Just have to wait patiently till I can reap some financial rewards for this ain't a thing happening quickly at all at my working place: advancement (admittedly: I refuse to kiss butts & lick asses)

My Operating System project (BlueIllusionOS) has sprung to life six 1/2 years ago if I remember correctly. Althou' I have abandoned the website in the meantime, seeing no sense in caring for it anymore, I still do os-Deving for a pastime. These Days I'm fidgeting with Network Code (Tcp/IP Stack) and squashing the one or other big bug in that thing.

Well - gotta say that OS_Deving up to the rough thing offering most basic os services is quite an ambitious project - one learns to plan ahead, see the big picture, keep it in mind and putting together all the little elements needed to form it. An OS ain't complicated. It is complex - a thing consisting of a bunch of smaller, simple entities working together. With this knowledge and feeling, developing big SW-Projects is quite a breeze for one doesn't see one huge pile but knows by instinct: break it down and stuff the components together and there you go. ) That's the reward I've reaped from OS Deving: Knowledge and Instinct.

I do not work in OS Dev, but i do get offered related work 2-3 time a year.
So hobby OS Dev will help you a lot, if thats the direction you want to go.
Good luck.

Are you going to count building the "kernels" of simple embedded systems as OSDEVing? Or people who only build hardware drivers? If not, I think I can count the number of companies that are actively creating full-scale OSes on my fingers. The total number of people on the planet that are getting paid for doing "real" OSDEVing is probably less than ten thousand.

My job is doing hardware support for a "legacy" computer brand, and building software interfaces from the old hardware to new hardware.

I'm building my own because I can't stand the limitations/bugs/performance of current OSes anymore, but I need something to run my PCs.

I'm still in college seeking a job, so no, I'm not working in professional kernel development. Nor do I intend to, it's probably way too much for me to get into as my first job, since I still wouldn't consider myself even able to professionally program regular software yet.

Although, since I'm in computer engineering, I imagine that my ultimate job will probably be related to OS development since it will involve a good deal of embedded systems and software / hardware interfacing. I may end up writing an embedded OS one day, but that's probably going to be it.

Getting into OS development has to be hard anyway. The only real contenders where you'd actually get paid would be Windows and MacOS. Linux and other OSes don't have the market share to really consider searching for a job in them. I can't imagine that it's easy to land a job at Microsoft or Apple in OS development, and even then I doubt it pays enough to be worth the hassel.

OS devving is completely hobby work for me, and it's almost certainly going to stay that way. I know I can't write a professional grade OS, so this is all just a learning experience for me. And I must say, even just working with the emulator and writing a bootloader, I've learned plenty already.

All of you seem to be doing their degree in computers or working in an IT company. But i am doing my university in electronics ... Electronics too is interesting for me. But since 2 years i have been interested into os dev. Over two years i have gained considerable experience due to my os and am able to solve the os doubts of all my computer engineering friends who are studying with me in my college... Kernel development hence is a hobby but i want to take up a job in that and not in electronics...

I'm on the final year of my programming degree (focusing in video game development). The area I'm interested in and good at is the very low level stuff which requires me to write flexible, highly optimized, cross platform code (libraries for networking, audio, 3D, physics, AI, working with limited hardware (handheld consoles)).

I frown upon a lot of companies that are writing lazy, bloated, and unplanned code using hacked about libraries. It's an entirely different scene between the companies who write amateur code, and those companies who try to be at the forefront of their field which require optimized and innovative code. I believe the former are giving a bad name to the industry in general, and I think just because your game doesn't push technical limitations of your system is not an excuse to write bad code.

I've mentioned I like working on a low level, which is why I started my operating system. I would love to work on it more, but sadly due to it being my final year I don't have the time, and when I do have the time I feel lazy and unmotivated .

I sometimes find myself overlapping these two fields. For example, I have started work on a game framework which I've borrows the microkernel concept of things running in their own address spaces and the actual game play in a virtualised environment. In the user framework for my OS I've borrowed a lot of concepts from game libraries I've used/developed, including using an event based game loop and messaging style system.

Hi,

SandeepMathew wrote:How many people here actually develop kernels as a part of your job ?

Heh - not me. None of the jobs I've had have really involved programming.

SandeepMathew wrote:As far as I am concerned my real job is light years away from kernel development , i do consider kernel development as a hobby . How many of you are hobbist kernel hackers ( like me ) . But at some point in my carrier I want to be a kernel developer , and if God permits , i will be .....

When I was in high school I was really interested in computer programming and electronics. I decided that programming is fun because I can do what I like, but if I worked as a programmer I wouldn't be able to do what I liked and programming wouldn't be fun anymore. Because of that I tried to get into electronics but I didn't have proper qualifications, so when I was offered an electrical apprenticeship I took it (I assumed it was "close enough" and that I'd be able to move from electrical to electronics later on). My interest in electronics mostly evaporated since.

SandeepMathew wrote:Finally where did it begin and what attracted you to os development ?

I grew up on a Commodore 64, where the built-in BASIC interpreter taught me how easy programming could be; but interpreted BASIC has some severe performance problems which is what attracted me to assembly language (assembly language was the *only* way to get anything close to decent performance from a 1 MHz 8-bit CPU).

Eventually I bought a "top of the line" 80486DX2 running Windows 3.11. I set it up, turned it on, looked at the desktop and *then* wondered what on earth I was meant to do with such an expensive white box. There was no BASIC or any other programming languages, and if there was an assembler I didn't know how to use it anyway.

It didn't take me long to get hold of an assembler (A86) and some information (Peter Norton's "blue book") and start teaching myself 80x86 assembly while playing with DOS and BIOS functions (in real mode). I quickly moved on to doing some really strange (real mode, single tasking, very dodgy) operating systems; but mostly I was scratching around in the dark.

Then something amazing happened - something so ridiculously fantastic that you wouldn't believe it in a million years. It was the 1990's, and the results of some USA military research was starting to be introduced to normal citizens. Yep, say hello to the internet! Proper tools, actual programming manuals direct from Intel, mailing lists, news groups, and people who could actually understand what I was talking about! OMFG!!!

Of course my dodgy real mode single-tasking OSs got deleted. The rest is fairly obvious - natural progression if you will...


Cheers,

Brendan

Inspiring story Brendan.

I agree with what you say. It was fun when we did individual projects since while we had criteria we also had the flexibility to do whatever we wanted and full rights boast about it. Now since it's all team projects, individual productivity decreases, even though overall you achieve more than what you would by yourself (that have been times I could disagree - e.g. a 2 month long group project I could have done myself in a weekend). You have to do things in the order someone else wants to do, you have a technical director who is only doing the same amount of work as the other programmers (they don't spend time delegating work, laying out technical milestones, etc like they should, and instead make us use unproductive coding and commenting standards), and you're not able to randomly implement cool impressive things because your tech dir says it's not part of the scope, it's too complex for this project, or your skills can be better put to use on the boring stuff.

Frustrating indeed. Word of advice, don't take a management position and force your friends to follow. You might very well be the best leader, but you're likely to get a few disagreements.

In 2 weeks I have a 2 week break. Then to make things more fun, I'm going to be doing a double internship plus our major huge college final project that is likely to decide on my future.

Andrew

Hi ,
However i do programming as part of my job , but not the domain i really like ( but the job is challenging and exciting ) .

Developing in a group is fun when there is a mutual respect and friendship between the members , otherwise it is a pain in the ____ . I am missing my friend so much , in order to communicate an idea to him , sometimes all i require is a few guestures and it will be code right away

truly Evil
Saint Lucifer

me not working in kernel development but took it as a challenge as i just wanted to check that i dont loose my programming skill.
soon it started to become my hobby and now its like i need to do more in order to get my daily dose.
but my job does not allow me to work on osdev religiously.