[Condensed excerpts from] Chapter 1:
First Principles, Mental Models, and a Framework for Real-Time Actions
MAJOR UPDATE TO PREVIEW CHAPTERS 1, 2 3 WILL APPEAR SOON
Chapter 1 begins by developing the concept of real-time per se from first principles, such as latency, timeliness, and predictability.
First principles are the fundamental concepts or assumptions on which a theory, system, or method is based, analogous to axioms. An example of a first principle for real-time is that an action may have a completion time constraint which is part of the action’s logic.
“Reasoning by first principles is one of the best ways to develop mental models
that are rare and useful.
Put another way, forcing yourself to look at the fundamental facts of a situation
can help you develop your own perspective on how to solve problems
rather than defaulting to way the rest of the world thinks.”
—James Clear, Mental Models
Chapter 1 next formulates mental models. Mental models facilitate reasoning about abstractions which are built on the first principles.
Examples of mental models include
- A syntactic construct to express the completion time constraint for an action—e.g.: a deadline; or a time/utility function
- Specifying timeliness as an application-specific semantic interpretation of an action’s completion time with respect to its time constraint—e.g.: an action which misses its deadline constitutes a failure; or an action’s completion time with respect to its time/utility function has some effects(s) on how satisfactory the action is
- The kinds of uncertainties in the system and its operational environment which may affect an action’s timeliness and predictability of timeliness, and how—e.g.: everything is static so there are no uncertainties; or time-constrained actions are periodic with presumed worst case operation durations; or certain action properties such as arrivals and operation durations and shared resource conflicts are intrinsically epistemically uncertain.
Those models are then placed in a conceptual framework to compose a coherent viewpoint of real-time systems in general—e.g., a taxonomy or ontology of “real-time,” or timeliness and predictability of timeliness as qualities of service.
This logical exposition is a dramatic departure from how the traditional special case static real-time viewpoint is based primarily on ad hoc fragmented and loosely structured lore.