I hadn't worked through the biasing calculations for collector feedback common emitter amplifiers before, so I thought I'd do that for the simple one transistor NPN preamp that Dino built as his weeklyhack. But even before I got to it, I was confused by something. Check out the schematic (cribbed from Dino's post):
I was confused by R1. It's only purpose seemed to be to provide a DC offset, which was immediately removed by the action of C1. So, since I had built the circuit in LTSpice, I tried varying its value from 1 ohm to 100MEG ohms, and indeed, it had no effect on the overall gain of the amplifier. So, I removed it entirely. And the circuit continued to work just fine.
So, what's the deal? I suspect that R1 was part of a circuit that powered an electret microphone: if the Input on his schematic were an electret microphone, you'd need to provide some voltage into it to power the built in FET preamp (like you see in the diagrams on the Wikipedia page for electrets). If that power is not needed, then all that happens is the current is dissipated by the input as heat. If you cascade two of these in the naive way, you end up a resistor feeding between voltage between two DC blocking transistors, which doesn't waste a lot of power (no current flows at DC), but you end up with an extra cap and resistor.