On Sat, 07 Oct 2006 15:31:53 +0200 David Kastrup <[EMAIL PROTECTED]> wrote:
> Do you have any evidence of the term "kernel" being used before, or > actually even outside of UNIX? I think it likely that UNIX was the > main culprit for the informal erosion of "operating system" which has > not really managed to make it fully back to academia (you'll find > "operating system theory" courses, but hardly "kernel theory"). I quoted this section from Ralston's "Encyclopedia of Computer Science and Engineering", 2nd Edition (Copyright © 1983 Van Nostrand Reinhold) in my answer to one of Alfred's posts: "The term /kernel/ (and sometimes /nucleus/) is applied to the set of programs in an operating system which implement the most primitive of that system's functions. The precise interpretation of kernel programs, of course, depends on the system; however, typical kernels contain programs for four types of functions: 1. /Process management/ (description elided) 2. /Memory management/ (description elided) 3. /Basic I/O control/ (description elided) 4. /Security/ (description elided) In some systems, the kernel is larger and provides for more than these classes of functions. In others, it is smaller." One could argue be that by 1983 Unix had already had such influence that the term kernel had been "backported" into descriptions of other operating systems. But even simplistic 70s OSes like CP/M consisted of a nucleus (BIOS/BDOS/CCP) and utility programs (such as PIP and FORMAT), and CP/M was never influenced by Unix. Take care, -- Stefaan A Eeckels -- Q: If ignorance is bliss, why aren't there more happy people in the world? A: Because they don't know they're ignorant. _______________________________________________ gnu-misc-discuss mailing list gnu-misc-discuss@gnu.org http://lists.gnu.org/mailman/listinfo/gnu-misc-discuss