Hi,

I am an absolute newbie in this field so please excuse me if you find
my question silly/obnoxious. We already have many excellent micro-
kernels floating around (say L4 for instance) then what is the point
in continuing with our own version of the same?

Hi,
We no doubt have a lot many micro kernels around. But each vary with
their goal. Some may want performance while some may look at
security...here in minix, we wanted reliability the most...as
Tanenbaum says...we want our computer to be like our TV in all
respect, i.e. in terms of reliability...(other factors also keeping in
mind, like security and performance...). Hence we have a new kind of
micro kernel...Self recovering is another beautiful feature of this
micro kernel...I hope this clarifies our doubt (why one more micro
kernel?) a bit...more on the differences and similarities with other
micro kernels can be found in the papers listed in minix.org...

Thanks
Srinu.