3 ms·
It would help if the author actually understood what "positive feedback loop" means in engineering. It's not what he thinks, and social-science terms like "posi
by lutusp 14y ago
It would help if the author actually understood what "positive feedback loop" means in engineering. It's not what he thinks, and social-science terms like "positive feedback" don't translate well in the more rigorous sciences.
In strict engineering terms, feedback loops must always have a gain less than unity, and a stable, reliable system normally has negative net feedback. But this doesn't sound "right" to a social scientist, who understands these terms in an entirely different way.
- dsirijus 14y agoI've made two examples to help others discern between positive feedback loop (engineering) and positive feedback (social sciences). Where from have you drawn that conclusion? Genuine interest.
- lutusp 14y ago> I've made two examples to help others discern between positive feedback loop (engineering) and positive feedback (social sciences). I posted only because you used the entire term "positive feedback loop", which is so specific that it loses flexibility -- it can only refer to one thing and conflates the engineering and social sciences meanings. "Positive feedback", when not accompanied by "loop", can be stable. But a loop with positive feedback is generally unstable, because the output feeds the input. Here are pictures of loops with positive, unity and negative feedback: http://i.imgur.com/pxANv.png http://i.imgur.com/pxANv.png http://i.imgur.com/od4CS.png http://i.imgur.com/od4CS.png http://i.imgur.com/vC6yV.png http://i.imgur.com/vC6yV.png Very clear, and all solutions to this classic damped oscillator equation: http://i.imgur.com/h5wKA.png http://i.imgur.com/h5wKA.png Where: a = angular frequency b = damping factor Without any reference to the present exchange, this kind of language is often used by those who want to elevate an argument, make it seem more like physics -- more scientific, more deterministic, but usually in a field that doesn't have those properties.
- dsirijus 14y ago> I posted only because you used the entire term "positive feedback loop", which is so specific that it loses flexibility -- it can only refer to one thing and conflates the engineering and social sciences meanings. There's life to positive feedback loops outside engineering without any conflation. If it's more defined in one field, it doesn't mean it has no value in plenty of others. http://en.wikipedia.org/wiki/Positive_feedback_loop http://en.wikipedia.org/wiki/Positive_feedback_loop
- JoeAltmaier 14y agoSo, is there some way of controlling the gain? Hellbanning is one way but thats bang-bang (gain is 1 or 0). How about, limiting the visibility of comments to a subset of the community. If they comment/answer, then show the thread to more. If they ignore (its a rant etc) then it quietly sinks into oblivion. Just an idea.
- lutusp 14y ago> So, is there some way of controlling the gain? In engineering, yes, always. In the social sciences, not always. > How about, limiting the visibility of comments to a subset of the community. That would mean the system isn't a single system any more. Apart from making the system two or more systems, it's undemocratic and probably counterproductive. Consider the classic example -- astrophysicist Jim Peebles had an idea about the Big Bang, that there should be a residual "echo" of the original event, "audible" in the present universe, in the form of a background radiation with a specific temperature that was relatively easy to compute. Peebles realized the radiation should appear in the microwave region of the electromagnetic spectrum, and it should seem to be coming from everywhere (highly uniform). Meanwhile, physicists Penzias & Wilson over at Bell Labs had a microwave dish with an annoying noise they couldn't get rid of. They scrubbed away bird droppings, they adjusted the front end amplifier, nothing helped, and it didn't matter which way they pointed the dish -- the signal seemed to be coming from everywhere. Penzias & Wilson called Peebles and asked hum what he though the noise might be. Peebles told them, hung up the phone, and said, "We've been scooped." Penzias & Wilson won the Nobel Prize. Now that's feedback. And it only happened because there were no communications barriers.