- Joined
- Aug 8, 2020
So in the world of process control, PID is generally considered the gold standard. This is how drones and guided missiles work. It's how Segways work. Limited implementations of it are use in climate control*****(usually without the D in the latter point). So let me go on a dumb rant about how this general idea could be applied to political reform:
Let's assume the general population wants the government to work a certain way, and the government we have does not reflect that. The difference between how society operates and how the people want it to operate shall be denoted as "error" (or E). Where people want it to be is the "setpoint" (or SP). How much things change based on collective action shall be called the "manipulated variable" (MV). The vector of that change is simply called "gain."
So P is for proportional gain. It's the simple idea that, the further you are to your goal, the more you push back. This sounds like it would be sufficient but there comes a point where you are so close to where you want to be that the very little push you give amounts to nothing compared to the preference to maintain the status quo. This is where most people are; they don't like how things are but it's okay enough to not push harder
I is the solution to the problems with P. The integral gain remembers past experience and fights harder the longer that error remains. This fixes the offset by fighting harder and harder until the goal is achieved. PI controllers exist, but they come with the problem of overshoot. They go well beyond the desired goal and have to bounce back and forth until reaching their desired outcome. These are your antifa types, your sovereign citizens, your accelerationists, etc. They push as hard as they can to get out of the "wrong" half of whatever political dichotomy they care about with excessive force.
That brings us to D. This is the derivative gain that acts as a counter-corrective force. It is primarily a force that takes note of the current direction the system's going and tries to flatten the curve, so to speak. They are the ones that see how slippery the slope is and actively work against any change to the current system. D's want to keep things the same no matter what for fear of change.They're happy with where they are and fear what the future may hold if the P's and I's get their way.
In the world of process control, it's all about balancing these three parameters so that you get where you want to be in the best way. Ziegler-Nichols gets you there really fast, but theres a violent overshoot and undershoot that occurs along the way. No-overshoot is a more gradual process that seamlessly leads to the desired outcome. But it takes much longer to do so.
So, in your opinion, what would be an ideal ratio of P, I and D people to get us where we want to be while minimizing human suffering?
*I obviously know people have diferent ideas about how the government should operate. And that throws a bit of a wrench in this model. But let's assume, for just a second, that it's as simple as a "common folk vs. elite" problem.
Let's assume the general population wants the government to work a certain way, and the government we have does not reflect that. The difference between how society operates and how the people want it to operate shall be denoted as "error" (or E). Where people want it to be is the "setpoint" (or SP). How much things change based on collective action shall be called the "manipulated variable" (MV). The vector of that change is simply called "gain."
So P is for proportional gain. It's the simple idea that, the further you are to your goal, the more you push back. This sounds like it would be sufficient but there comes a point where you are so close to where you want to be that the very little push you give amounts to nothing compared to the preference to maintain the status quo. This is where most people are; they don't like how things are but it's okay enough to not push harder
I is the solution to the problems with P. The integral gain remembers past experience and fights harder the longer that error remains. This fixes the offset by fighting harder and harder until the goal is achieved. PI controllers exist, but they come with the problem of overshoot. They go well beyond the desired goal and have to bounce back and forth until reaching their desired outcome. These are your antifa types, your sovereign citizens, your accelerationists, etc. They push as hard as they can to get out of the "wrong" half of whatever political dichotomy they care about with excessive force.
That brings us to D. This is the derivative gain that acts as a counter-corrective force. It is primarily a force that takes note of the current direction the system's going and tries to flatten the curve, so to speak. They are the ones that see how slippery the slope is and actively work against any change to the current system. D's want to keep things the same no matter what for fear of change.They're happy with where they are and fear what the future may hold if the P's and I's get their way.
In the world of process control, it's all about balancing these three parameters so that you get where you want to be in the best way. Ziegler-Nichols gets you there really fast, but theres a violent overshoot and undershoot that occurs along the way. No-overshoot is a more gradual process that seamlessly leads to the desired outcome. But it takes much longer to do so.
So, in your opinion, what would be an ideal ratio of P, I and D people to get us where we want to be while minimizing human suffering?
*I obviously know people have diferent ideas about how the government should operate. And that throws a bit of a wrench in this model. But let's assume, for just a second, that it's as simple as a "common folk vs. elite" problem.
