I disagree with most of your comment, but explaining why will require several paragraphs of background.
You can read Maxwell's On Governors from 1868 and see that, although it explores their math in some detail, it doesn't attempt to generalize beyond braking or otherwise slowing down steam-engines: https://www.maths.ed.ac.uk/~v1ranick/papers/maxwell1.pdf
So, although devices with negative feedback had been built for centuries — and many natural or simple artificial objects can be usefully analyzed in terms of feedback — the general idea of negative feedback as such seems to date from the 1920s, and apparently Black discovered its usefulness in 1927. The idea was so outlandish at the time that his patent was initially rejected as describing a useless invention — why would you want to reduce the gain of an amplifier? It wasn't public until 1933.
After Black's invention, which upgraded FDM for long-distance telephony (the "carrier system") from three voice channels per wire up to nine, then dramatically more, and permitting long-distance circuits with a sequence of 34 repeaters, rather than, say, three.
Bennett's "A history of control engineering, 1930–1955" discusses this history in detail; find it at your favorite library.
At the same time, in 1922, Minorsky formalized PID control, which gets into negative feedback from a different angle, that of automatic ship steering.
Black's, Minorsky's, and earlier negative-feedback designs were not generalized into a general theory of negative feedback until the work of Nyquist and others in the 1930s — most prominently, Wiener!
Nowadays, explicitly designed negative feedback is far more ubiquitous than your comment suggests. By far the most common analog circuit component, for example, is an op-amp, which is almost invariably applied with negative feedback; also, though, negative feedback is central to nearly any kind of homeostasis, optimization, or stable static mechanical equilibrium. So, for example, we have negative feedback in TCP bandwidth usage, in glucemia, and in your toilet tank. Parkinsonian tremors — the opposite of your "very steady hand" — are now being analyzed as oscillations in a negative feedback system involved in human motor control, though the theory is not yet fully mainstream in neuroscience.
As for "digital devices", those date back at least to Leibniz; they are far older than analog electronics, which in turn are a couple of decades older than the concept of negative feedback.
You can read Maxwell's On Governors from 1868 and see that, although it explores their math in some detail, it doesn't attempt to generalize beyond braking or otherwise slowing down steam-engines: https://www.maths.ed.ac.uk/~v1ranick/papers/maxwell1.pdf
So, although devices with negative feedback had been built for centuries — and many natural or simple artificial objects can be usefully analyzed in terms of feedback — the general idea of negative feedback as such seems to date from the 1920s, and apparently Black discovered its usefulness in 1927. The idea was so outlandish at the time that his patent was initially rejected as describing a useless invention — why would you want to reduce the gain of an amplifier? It wasn't public until 1933.
After Black's invention, which upgraded FDM for long-distance telephony (the "carrier system") from three voice channels per wire up to nine, then dramatically more, and permitting long-distance circuits with a sequence of 34 repeaters, rather than, say, three.
Bennett's "A history of control engineering, 1930–1955" discusses this history in detail; find it at your favorite library.
At the same time, in 1922, Minorsky formalized PID control, which gets into negative feedback from a different angle, that of automatic ship steering.
Black's, Minorsky's, and earlier negative-feedback designs were not generalized into a general theory of negative feedback until the work of Nyquist and others in the 1930s — most prominently, Wiener!
Nowadays, explicitly designed negative feedback is far more ubiquitous than your comment suggests. By far the most common analog circuit component, for example, is an op-amp, which is almost invariably applied with negative feedback; also, though, negative feedback is central to nearly any kind of homeostasis, optimization, or stable static mechanical equilibrium. So, for example, we have negative feedback in TCP bandwidth usage, in glucemia, and in your toilet tank. Parkinsonian tremors — the opposite of your "very steady hand" — are now being analyzed as oscillations in a negative feedback system involved in human motor control, though the theory is not yet fully mainstream in neuroscience.
As for "digital devices", those date back at least to Leibniz; they are far older than analog electronics, which in turn are a couple of decades older than the concept of negative feedback.