Almost every site I look on has at least one, and probably more, people who are going to write their own operating system. What's the point? When its finished the chances are nobody is going to use it - probably not even you.

I might not mind spending a few hours on a whim, but, mercenary sort that I am, if I was going to spend a couple of years on a project I would want to be able to sell the end product for hard cash.

Really, writing your own OS is, for me, just about having it. Just to say, "I wrote an Operating System, and no one else in the world has it" would be pretty awesome. I'm in the process of getting better at C++ and I intend to write my own OS eventually. You did have a good point when you said that no one would probably use it, but you could get it out there somehow. You don't necessarily have to get money for it.

Not only that, but if there's something wrong with it, you don't have to go to Windows/Apple tech support; you can just go in and find what's wrong with it. Granted, that would probably take a long time, but I think it's more of a bragging rights/accomplishment thing with these people, including me.

Whether educational or for hobby,
to undertake defining the os enviroment,
is a lengthy task.
There is much to learn, and to gain experience
controlling, maintianing and setting up
the system,
such experience is fundamental for making
system software, and drivers,
for the same architecture.
When writing for a kernel, or modifying it,
knowledge of the system architecture
is important,
and gives deep insight into how an OS
works, having many applications,
even in diagnosing problems,
and especially in explaining what is
going on in an OS behind the scenes.
Sometime OS development will apply
to studying operating systems to
learn to system programming,
and not particularly creating full
operating systems.
Xinu and Minix are good examples.