On Tue, 24 Mar 1998, William T Wilson wrote:
[...]
>It's a matter of which definition of OS you subscribe to. Some say that
>the OS is responsible for mediating the interactions between CPU, RAM,
>and peripherals, and providing a set of system calls for programs to
>interact with them. Linux (even the kernel by itself) provides all this.
>Others say that the OS is supposed to include all kinds of other things
>(user interface, window system, applets, programming libraries, and so
>on, which are typically provided by the distribution under Linux).
>
>The second flavor is typically adhered to by the Microsoft marketing
>department. The rest of the world follows the first definition.
Apparently, the academic definition differs from "the rest of the world".
Dr. Peter Denning (instrumental in developing modern virtual memory),
who taught my operating systems class in college, made it quite clear
that "operating systems" included the basic tools--beyond just the
kernel--necessary to operate the system. The kernel cannot exist in
a vacuum. An operating system *must* have additional basic tools, such
as some sort of user interface (how else do you control or even use the
system?), a loader/linker, hardware configuration tools (e.g. ifconfig,
setserial, tset), and a set of configuration files that control the
operation of the basic system (e.g. the contents of /etc). Clearly, a
complex (and inflexible) "kernel" could contain all of those components
and stand on its own, but Linux does not. Microsoft would have you
believe that the Microsoft "kernel" does.
Quite simply, you are incorrect in your definition of "operating system".
There is a reason the kernel is called the "kernel" and not the
"operating system". "Kernel" is not a synonym for "operating system".
--
Steve Coile
[EMAIL PROTECTED]
--
PLEASE read the Red Hat FAQ, Tips, Errata and the MAILING LIST ARCHIVES!
http://www.redhat.com/RedHat-FAQ /RedHat-Errata /RedHat-Tips /mailing-lists
To unsubscribe: mail [EMAIL PROTECTED] with
"unsubscribe" as the Subject.
