BJTs can work good as amplifiers because their function is linear, the only factor is gain and it has no exponent, and this function is suitable for amplification. So traditional transistors replaced lamps because they had same linearity sort of speaking. But mosfets can not do that on same level. One of the implications of the curve function of mosfets is that it can not work as amplifier of large signal. Because the function is curved (coz of the exponent) it distorts the amplification. Draw a graph with xoy and a curved function and see that a certain variation on X produces different outputs on Y, depending where the X variation takes place. Meaning some levels get more amplifications, others less so, which means amplification distorts the original signal. So when we want amplification of a signal, we want it to keep the original signal integrity, but only to amplify it across the board, amplification has to be same at all levels, which means we do not want a curved function, but a linear one instead.
Maybe sometimes are used as amplifiers but only on *small signal* where it can ignore the curved function distortion on the account of it being small because the signal variation was small to start with. So when are used as amplifiers its small signal only, and where application would accept *some* level of distortion. So generally speaking mosfets are not used as ampifiers, again except small signal amplification but thats rather rare - as a matter of opinion, my own.
What i see largely used as "definition" its current device or voltage device, depending on input type. That matters because our uC are voltage level outputs so uC outputs voltage, therefore when pairing devices we want to pair it also with a device which takes voltage for input, so they will speak the same "language". Say an uC output of 5v goes to a transistor base, but the base would require 50mA to get to the operating point, which the uC may not provide, that would be a case where the two devices speak different language. So if we need a switch to a uC pin, better to use a mosfet because they can "understand" each other better.
Otherwise to know what a device is, have to know its elementary function like that, and check the implications of it. Any device has an some input and some output, and "a device that amplifies and/or switches the flow of current" thats some unprofessional generalization that could actually suit almost any device out there. Including my wall mounted light switches, some old crt monitors or closer to the subject, old school vacuum tube, which are not transistors by any means, they were actually the devices which the transistors replaced. And there are more similarities between tubes and bjts in that their elementary function are almost the same, except tubes are still better for amplification so thats why a class A+ audio amplifier still has to use vacuum tube specifically for the amplification function. If it uses a bjt for amplification function, then its less than class A, it may be class D audio amplifier instead.
Maybe sometimes are used as amplifiers but only on *small signal* where it can ignore the curved function distortion on the account of it being small because the signal variation was small to start with. So when are used as amplifiers its small signal only, and where application would accept *some* level of distortion. So generally speaking mosfets are not used as ampifiers, again except small signal amplification but thats rather rare - as a matter of opinion, my own.
What i see largely used as "definition" its current device or voltage device, depending on input type. That matters because our uC are voltage level outputs so uC outputs voltage, therefore when pairing devices we want to pair it also with a device which takes voltage for input, so they will speak the same "language". Say an uC output of 5v goes to a transistor base, but the base would require 50mA to get to the operating point, which the uC may not provide, that would be a case where the two devices speak different language. So if we need a switch to a uC pin, better to use a mosfet because they can "understand" each other better.
Otherwise to know what a device is, have to know its elementary function like that, and check the implications of it. Any device has an some input and some output, and "a device that amplifies and/or switches the flow of current" thats some unprofessional generalization that could actually suit almost any device out there. Including my wall mounted light switches, some old crt monitors or closer to the subject, old school vacuum tube, which are not transistors by any means, they were actually the devices which the transistors replaced. And there are more similarities between tubes and bjts in that their elementary function are almost the same, except tubes are still better for amplification so thats why a class A+ audio amplifier still has to use vacuum tube specifically for the amplification function. If it uses a bjt for amplification function, then its less than class A, it may be class D audio amplifier instead.