As I'm reading a scientific paper from 1964, I'm wondering how researchers were able to write papers without the use of a computer. Cursive fonts, different font sizes, math, etc. are all present in these (old) papers.
There were several layers of craft technology now partly superseded by pdflatex and friends. There were author's instructions for marking up manuscripts with type font indications, there were technical typists adept at using special math symbol shapes with their modified typewriters, there were Monotype operators who ran typesetting machines from the marked up copy supplied, and so on. `
The period 1960-1980 saw a lot of technology churn as different kinds of phototypesetting machines came into use, of varying typographic quality (or lack of quality). Stabilizing this situation was one of Knuth's reasons for getting interested in typesetting: he wanted his second edition (phototypeset) of his magnum opus to look like the first (monotype hot metal set).
Added 23 Jan: And, as the thread mentioned in a comment by Gerald Edgar makes clear, there was also considerable churn on the software side.
My theoretical physics papers and thesis were written by hand, then typed on an IBM "golf ball" typewriter. The golf ball was so called because the type head was about the same size as a golf ball. The type heads were interchangeable with different fonts and symbols. Corrections were made using tippex. 40 years later, I'm still amazed at the skill of the typist.