http://netlab.cs.ucla.edu/wiki/files/shannon1949.pdf
Speaking as someone who is not a formally trained cryptographer, most introductory material on the topic is pretty terrible. Finally stumbling across Shannon's early work, many years after starting in security, was like a breath of fresh air.
Maxwell’s laws of electricity and magnetism are even harder. Today, “Maxwell’s Laws” refers to four equations, each quite brief. In his original paper, though, they are a system of 20 simultaneous equations with 20 variables with a dizzying amount of bookkeeping to keep track of x- y- and z- dimensions, because the curl and divergence operators hadn’t really been invented yet.
While now I tend to think about special relativity using Minkowski space (which I find to be the easiest mental construct for me), my 12-year-old self found all that stuff in Einstein's Relativity about trains and clocks very helpful in building intuition about relativity.
Just yesterday I was identifying a species of African ant. As I researched it, I found myself using taxonomic keys and descriptions published between 50 and 150 years ago - in English, French, German and Latin! Fortunately they had been digitized, and at least partially translated...
re Alexander von Humboldt: Andrea Wulf’s, The Invention of Nature[0], is an excellent overview of his work and impact.
It's quite shocking how little he is known; somewhat similar to James Clerk Maxwell.
[0] https://www.goodreads.com/book/show/23995249-the-invention-o...
I've admired Humboldt ever since I was a child - her book reawakened that interest and gave me the impetus to read some of his own writings.
There are two "new" references about "old" physics literature I particularly enjoyed: - "The Book Nobody Read: Chasing the Revolutions of Nicolaus Copernicus," by O. Gingerich, about Copernicus' book "De revolutionibus."
- "A Student’s Guide Through the Great Physics Texts," by K. Kuehn (4 volumes). It presents many original papers by Newton, Galileo, Maxwell, etc., together with study questions and exercises.
It is often presumed that because we learn new things as time goes on, and old theories are discarded as new information comes to light, that there is not much value in reading older science books. After all, isn't half of that stuff wrong? But similar to how Jack says how one can learn about different flavors of Christianity by stepping out of his own century, and thereby how much they have in common, I feel that by reading the original works of Millikan, Carnot, Gibbs, or Einstein, that you get a much better idea of the perspective that developed the theories we see as so cut-and-dried today.