Three pairs of neon lamps flicker on and off rapidly until the user presses a button, which stops the flickering and leaves only one member of each pair on. Given three pairs of lamps, there are exactly eight combinations, since two times two times two is equal to eight. At the time, there was lots of interest in Extra-Sensory Perception (ESP), so my device could be used to test if a person could demonstrate foreknowledge of the outcome of a "random" event. It could also be used to learn about binary arithmetic, at least a bit - (pun intended!)
This time, instead of using Google, I used Bing, and searched for "Ira Glickstein". (I've never heard anyone say they were going to "Bing" something.) In any case, after looking at the first two pages of references to "Ira Glickstein", I impulsively clicked on the seventh page and there it was, a link to my 1957 article in POPULAR ELECTRONICS! Someone had kindly scanned the entire issue and posted it for all the world to see! Here is a link to the .pdf file. I've reproduced the three-page article below.
As I look at my design from the perspective of a retired System Engineer with a long and creative career conceiving and designing complex avionics systems, I'm pretty impressed at how I, as a college sophomore, adapted the basic idea of an RC (Resistive/Capacitive) relaxation oscillator using neon lamps.
Looking at the circuit diagram above, it took me awhile to remember that neon lamps are basically two parallel wires, with a small gap between them, sealed within a neon-gas-filled glass tube. When the voltage difference between the two wires in the lamp is below a certain critical value, the effective resistance is nearly infinite. As the voltage builds up, a point is reached where electrons have enough energy to jump the gap. At that point, the neon gas glows, and the effective resistance drops. That causes the voltage across that gap to drop, and that lamp goes off. The capacitor in the circuit stores energy so that, as the voltage rises once more, the voltage across the gap in the other neon lamp of the pair will increase faster. Thus, the two lamps in a pair will alternately flicker on and off. When the user pushes the button, the two neon lamps of a pair are put in parallel, so they have the same voltage across them, which makes the neon lamp that was on stay on, and the one that was off stay off.
Since the three pairs of lamps and associated resistors and capacitors are bound to have slightly different parameters, they each oscillate at a slightly different frequency. Thus, when the user presses the button, each pair is likely to be in a different phase of the oscillation, such that the resultant final state is more or less "random", any of the eight possible combinations being equally likely.
However, when I learned more about statistics, I realized that there was a non-zero probability for "random" coin flips to appear non-random. For example, as you know, the probability of a "fair-coin" landing "Heads" is 50%. The probability of two "Heads" in a row is 25%, three in a row is 12.5%, four in a row is 6.25%, five in a row is 3.125%, six in a row is 1.5625% and so on.
I made use of that fact when teaching a graduate course in System Engineering by asking my students to do a two-part experiment. In the first part, they were to manually make up a list of 200 "Heads" and "Tails" that they thought was "random". In the second part, they were to actually toss a coin 200 times and record the actual "Heads" and "Tails". They were to label one of their lists "A" and the other "B", recording, but not telling me, which was made up manually and which was the actual record of 200 real coin tosses. In almost every case, I was able to tell which was which!
How did I do it? Well, given 200 actual coin tosses, the chance of getting six "Heads" or six "Tails" in a row somewhere in the sequence is almost 100%. However, when someone manually makes up a list of 200 "Heads" and "Tails" they (almost) never write down a series of five or six "Heads" or "Tails" in a row, because that does not look "random" to them! So, I'd check the two lists submitted by each student, and, if one list had a sequence of six or more "Heads" or "Tails" in a row, and the other list did not, I'd know the first list was for the real coin tosses!
I don't know if my Eight-Sided Dice POPULAR ELECTRONICS article is to blame, but, after studying Einstein's General Relativity and reading about his problem with the "Copenhagen Interpretation" of Quantum Mechanics, I came out on Einstein's side, rejecting the currently accepted view that Physics is truly random. Einstein said something like "God does not play DICE with the Universe!"
Despite the fact that Quantum Mechanics based on "Heisenberg's Uncertainty" is the most successful theory for predicting the outcome of sub-atomic experiments, I cannot shake the view that there is some currently hidden, non-random process behind all of it. So, like Einstein, I believe in Strict Causality, and, therefore, Absolute Determinism. That also requires me to believe that the Universe is both Finite and Discrete.