This study is aimed at reducing idle time in large computer systems. Unfortunately, the paper contains too much jargon and is obviously meant for a specialized few. The study seems to ignore the fact that large systems use multiprocessing and multitasking, allowing many tasks to run simultaneously; in contrast, small systems accept idle time as a given.
The section on idle time scheduling introduces the preemptive interval that follows each foreground action during which no action takes place. Isn’t this a waste of time, even though slight? The discussion that follows leaves me in the dark. If work can be done during this period, how can it be called idle time? The next section, “Implementation,” uses a state diagram to display transitions that may occur between idle, foreground, background, and preemptive (when the system stops to check things out) tasks. In summary, this paper lacks clarity and, moreover, quantitative results.