How control-friendly is a computing system? And how control-friendly could it be?
To date, not that much. Improvements are possible, but some system-theoretically grounded re-design is necessary. We discuss the matter based on our experience, and as a consequence, we come to distilling some design clues and research directions.
