An alternative source for those interested in more on the topic should see Bourne's "Methods of Information Handling", https://archive.org/details/methodsofinfo

An alternative source for those interested in more on the topic should see Bourne's "Methods of Information Handling", https://archive.org/details/methodsofinforma0000unse/mode/2up https://archive.org/details/methodsofinforma0000unse/mode/2u... . "Edge-notched punched cards is a dead medium" should be interpreted similar to "Latin is a dead language" - it's not that no one uses edge-notched cards, but that they are no longer an effective solution. I used a Silhouette cutter to make my edge-notched cards, with a Python program to generate each card as an SVG. I still needed to push out a lot of little holes. I want to try a laser cutter next time. For selection, I used the classic solution of using knitting needles, rather than an actual McBee needle like shown here. Here's a picture of my cards, containing small molecule information: https://twitter.com/nathanbroon/status/750460860150214656 https://twitter.com/nathanbroon/status/750460860150214656 My particular interest is in random superimposed coding methods, as created by Calvin Mooers while a Master's student at MIT. The end of the video mentions superimposed coding methods, which also includes non-random superimposed coding methods. Those were used in WWII. There was a patent challenge by McBee to Mooers' claim over the difference between "random" as in arbitrary and "random" as in chosen-at-random.