Interesting that he started out with sources categorized by country, but as he proceeded the categories disappeared and he constructed his non-linear groupings and visualizations. Also noteworthy, that by the time he was actually writing, he had read, or at least reviewed, his source material at least three times. By the time he was typing at the computer, there was never really the proverbial “blank page”. The manuscript was already there, it just needed to be applied to a new form.

Side note: I like his shirts. Reminds me of the professor I had in college for my graphic design class. He had good taste in clothes, too.