What is decided before the line
Three books about what is decided before the line: the viewpoint, the volumes that make up the form, and the distribution of dark masses.
I took up drawing seriously again this year, in a sketchbook and on location. What I took for a hobby turned out to be an engineer’s exercise, because the difficulty does not come from the hand. It comes from three decisions made before the line, often without noticing, whose consequences you then live with. Each of the books below deals with one of them.

About perception rather than technique: what the eye corrects without your noticing, and how to stop drawing what you think you see. The transfer to engineering is direct, since an architecture diagram almost always shows the system you think you have.

Hale takes drawings by the masters and redraws over them the simple solids they built on: boxes, cylinders, wedges. It is a form of reverse engineering: starting from the finished drawing, Hale reconstructs the decisions that produced it. Taking over someone else’s system involves the same work. The delivered code is the result, and what was decided before it has to be recovered.

Published in 1899, it is the most radical of the three. Dow takes the Japanese principle of notan, the arrangement of light and dark masses, and places it before line and before colour. You squint until the subject is reduced to two or three values; if that arrangement works, the drawing works, and if not, no detail will rescue it. This site applies the same idea: switching from paper to ink leaves the line unchanged and shows immediately whether the structure holds up when the ground changes.
None of these books teaches drawing, and each one read on its own leads to dry work. They describe what is decided before the line, which then assumes a great deal of practice. I know of no shortcut. I also leave aside materials, on which I spent a lot of time before understanding that they decide nothing important.