Advent(ure) in System Seeing — Day 5 — Key Systems Concept: Feedback
Let’s gather some of the thoughts that have been percolating around systems. Let’s assume that you are designing the introductory page of a zine on systems, so the frame you’re working within is about the size on 5x3 index card. I’m suggesting a “zine,” because they can be playful (or serious), have graphic elements (or not), etc., but the key is the small size as constraint. At this point, we’re designing the page, jotting down ideas (and images) that will go on the page, and then drafting the page.
Some ideas for the page: a characterization of systems. A list of the key ideas in systems. Some images that illustrate your characterization. Some examples.
If that falls out quickly, pick a key concept like Boundaries, and do the same thing, drafting ideas for the design of a zine page on system boundaries. Think of analogies that suggest visuals and insights, like cell walls and containers. Why are boundaries important, and what role do they play (in systems or our conceptions of systems)? Etc.
Once again, reserve some time (in your 15–20 minutes) to step back and reflect on what you learned.
Feedback affects the ability of a system to respond to change, and whether that response is stable or unstable. It’s mathematically defined as part of control theory, and complicated systems can be modelled accurately. Complex (and chaotic) systems don’t have exact analytical solutions but the same feedback principles apply.
Feedback starts with system that has an input and an output, and a desired value of that output. The difference between the actual output and the desired output is the error. A feedback algorithm looks at the error and changes the input to hopefully improve the output. The feedback system may also anticipate a desired change in the state and blend that with the error as it exerts control.
Feedback is characterised by two values, the strength of the feedback and the delay that occurs before the feedback affects the output. The most common mistake people make is that they don’t think about or measure the delay. Since the input is changing all the time, the time it takes to apply the feedback must be much faster than the rate of change of the input and output.
The strength and direction of the feedback affects how much it changes the input. Negative feedback takes the a proportion of the error and subtracts it from the input. Positive feedback occurs when the error is added to the input, and that creates an unstable system. Too high levels of feedback may cause overshoot and “ringing”, and too low levels may take too long for the system to correct the error. If you get it just right, a perfectly damped system minimises the error without oscillating.
If the feedback takes too long to take effect it acts like an input that is inappropriate to the current state of the system, and will have an effect that is sometimes positive and sometimes negative, and the system becomes unstable.
In the steering example, the inertia of a car is a stock of kinetic energy that depends upon the mass and speed of the car, and higher inertia slows the response of the car to inputs.
