Thus, I came to the conclusion that the designer of a new system must not only be the implementor and the first large-scale user; the designer should also write the first user manual. The separation of any of these four components would have hurt TeX significantly. If I had not participated fully in all these activities, literally hundreds of improvements would never have been made, because I would never have thought of them or perceived why they were important.
One thing I've been enjoying about working as a technical writer is that the minute-by-minute texture of the work feels right. Something about formatting text, faffing about with SVGs, trying to rewrite a sentence more clearly... it's just enjoyable in itself, and I feel at home with it.
...Working as a programmer was very much not like that. There's something in the rough vicinity of professional dev work that I do like, which I could probably label as 'iterative hobbyist tinkering with websites'. I like working on something with a strong visual component, and I like to be inside of a fast feedback loop, and I'm mostly interested in just somehow bodging through until it works. I'm not very interested in either the computer-sciencey side of programming — data structures, algorithms — or the software-engineerey side of making things run reliably at scale in a maintainable way. So maybe it's not surprising that the minute-by-minute texture of professional programming was just... kind of bad. Occasional fun bits when I got into something, but the background experience was not fun.