-- Indeed, the practice goes back centuries. What I'm trying to find out is when it began to change, and where (both in what discipline and what geographical area). As I said, I recognize the logic of the upright letters. I think the "need" for the distinction is much less among pure mathematicians, as "d" is not so often used for something else, whereas in physics, I believe it has a well-defined "other" meaning and is therefore liable to confusion.
I would like to learn of books or journals published before 1950 that use the upright forms; likely areas are physics, astronomy, engineeri…

-- It sounds like the 1991 Bronstein was professionally edited. But it wasn't typeset by the "traditional" method, which was essentially extinct by the mid-1970s, if not earlier. Photocomposition, even well edited, wasn't up to the metal standards. I don't even want to consider the "cold type" methods, which were essentially glorified typewriter output.
I actually have a few engineering textbooks, mostly mechanics and fluid dynamics, from the 1950s, if I can figure out which boxes they live in. Calculus too, but that's almost certainly using the "math" conventions. I'll try to start wi…

\documentclass{beamer}
\usepackage{tikz}

\usetikzlibrary{positioning,automata,overlay-beamer-styles}
\tikzset{orange on/.style={alt=<#1>{orange}{}}}
\begin{document}
\begin{frame}{}

\begin{figure}[h]

$abbaabababa\ldots$\\
\begin{tikzpicture}[shorten >=1pt,node distance=2cm,on grid,auto]
\node[state,initial] (q_0) {$q_0$};
\node[state,accepting] (q_1) [above right=of q_0] {$q_1$};
\node[state] (q_4) [right=of q_1] {$q_4$};
\node[state] (q_2) [below right=of q_0] {$q_2$};
\node[state,accepting](q_3) [right=of q_2] {$q_3$};