This fellow has been making his own diodes, and recently managed to construct a 7-segment decoder:
http://halestrom.net/darksleep/blog/026_diode_array/

Photos and video within, previous experiments nearby. The drawback is that it's a wet diode - perhaps not entirely unlike the inside of an electrolytic capacitor, based on aluminium and baking soda.





via hackaday

Wow, that's dedication! When I was in junior high, I had visions of making my own transistors (for analog work). I think Jeri Elsworth has done that. The 4004 microprocessor was coming out about then, but it would be a few years before I would hear about a microprocessor for the first time, especially living in another country, and I was not interested in digital yet anyway.

This unlikely diode reminds me of stories I've heard about people hearing faint radio reception from strange objects such as household plumbing or (IIRC) even their own teeth!

Of course a purpose-built radio receiver includes niceties such as tuned circuits. Parasitic receivers lack these, and consequently receive all stations at once (with the strongest signals most loudly audible). The heart of the circuit is the diode, which serves to demodulate the AM signal. This fellow's "wet diodes" give us some insight into how semiconductors can be created by accident.

-- Jeff

Building your own semiconductors.. so, all won't be lost after the acopalypse! We don't have to resort to only abacuses, maybe. So, with sufficient knowledge about physics and principles it can be done from scratch. Keeping the memes around is important. I was reading up (again) on the first transistor the other day. Shockley may perhaps have received too much credit compared to the guys who did the actual work, but his great contribution was when he realized that you have to think about holes as particles in their own right, following the same laws. Not just as absence of electrons. With that insight, the rest followed.
Now we know all these things.. so in principle it should be possible to come up with something that works like diodes and transistors, even without the infrastructure we have today. Wet diodes.. I hadn't thought about that. Can he come up with a working transistor? Not yet, apparently. But maybe that's next.