Well, lessee, we ought to be able to tell him sumthin' to give him some reassurance, right?
Voltage is an "across" variable, only makes sense when measured between two places, like temperature. (Saying something is "10 degrees" makes no sense unless you say what your reference is.)
Current is a "through" variable, it's measured flowing through a surface (like the cross section of a wire, or an imaginary sphere around an antenna), like heat.
When heat flows through a material, a "temperature drop" across the material forms, hotter in some places than others--unless it's a perfect conductor.
When current flows through a material, a "voltage drop" across the material forms, higher in some places than in others--unless it's a perfect electrical conductor.
In "linear" materials, the through and across variables are linearly proportional to each other.
In nonlinear, (like transistors) the relationship is not linear and may not even be known exactly.
For "perfect" material, either I = 0, no matter what voltage is impressed across the material (an insulator--you can't make current go through it no matter how hard you try), or V = 0, no matter how much current you shove through it (a super-conductor--you can't make a voltage difference appear across it).
Semi-conductors conduct under certain conditions, and act as insulators (poor conductors) when those conditions are not met. Semiconductors are "on-the-edge" of conductors: you can imagine that it might be possible to "do something" to make them better, or worse, conductors. Indeed. It's called "doping."
There's two ways to "dope": add more 'electron sources' than the s/c normally possesses (N-type doping) or add more 'places for existing electrons to jump into', called "holes" (P-type). Doped s/c are more conductive than non-doped.
The first bit of magic discovered at Bell Labs was that forming a "PN-junction" by putting an N-type next to a P-type resulted in a device that allowed current flow in one direction, but opposed current trying to flow the opposite direction. Same device behaves like a short in one direction, but like an insulator (open-circuit) in the opposite. Huh? Exactly. Something new. Something not natural. Something which just might be of some practical value. This is the famous diode. (They were already doing this with vacuum tubes, so they were actively trying to figure out how to do it with semiconductors to match behaviors they already had.)
The next bit of magic was to form TWO junctions (PNP or NPN). The three leads coming out of the relatively thicker ends and the thinnish middle were called collector, emitter, and base. typically, one of the leads is "shared" between the input voltage and the output voltage. (Remember that they are across variables.) Right away they discovered that, the input dc voltages and currents (the "biases") can be set "just so" to get different ac performances with ac signals.
In one "mode" where the transistor device is said to be in its "active" region, small (which means reasonably ranged) signals (can be voltage or current, depending on how the device is configured) are reproduced as larger output signals. In those cases, one can talk of the amplification or "gain" of the transistor.
In another mode, where the transistor is said to be biased at saturation or cutoff, it's behavior is in more of a slam-bang mode of operation: it slams on when even a small signal appears, and slams off when it disappears. That is to say, the output voltage (or current, depending) is either present and large or absent and small. In such a case, you could think of the device in that configuration as "determining whether there will be a large output signal, or not, with a small input signal"--another name for a relay... We say the transistor is biased in "switching-mode."
Sometimes you really want to "switch" something off-and-on with your transistor, but a third bit of magic came with the realization that the presence and absence of an output voltage represented two unique states: ON and OFF (called "positive logic," or OFF and ON for "negative logic"). Two unique states: two bits: Boolean arithmetic: digital computers. We'll interpret ONs as binary 1s, and OFFs as binary 0s. If you want to send me a message, code it in the pattern of output voltages of your transistor switch.
The genius of Boolean (2-state) computing is that we DON'T CARE what the SIZE of the voltage is, only WHETHER the voltage is. (That's a bit of a misrepresentation in real life because thresholds are established such that devices will accept anything above a certain voltage to be one thing and anything below a lower voltage to be the other. Anything in between is "never-supposed-to-happen" and they'll complain.
So "how-to bias" depends on what you want to use it for. There's guides.
Field-effect transistors are a little bit more mysterious, but if you can understand how different types of biasing predispose a BJT to work one way or another, you'll be able to grasp the behavior of FETs without trouble. In fact, FETs are easier in the sense that their behavior approaches ideal more closely than BJTs. (In the idea case, one ought to be able to use the device as a "switch" without it drawing power, but BJTs require significantly higher bias currents than FETs. By rethinking the structure, the power requirements were driven way down in the latter.)
Another useful way to think about electricity is to realize that it is sometimes used for POWER, and sometimes for INFORMATION, and increasingly for BOTH (those home networks which use your power line?). We saw that the output state of a switch could be used to turn something ON, or to communicate a bit of information.
(Someone else jump in here and give insights that have helped you.)
Living embodiment of Howe's Law.