Right at the beginning of my career, Symantec was approaching a bunch of little companies for some community outreach project they were doing that I can hardly remember.
Because they sent us a rep with severe RSI and I was so fascinated by her setup we barely talked about the project (and then my boss changed directions like he did 5 times a week).
This was just months before I briefly developed RSI myself and switched to Dvorak. The next time it resurfaced I learned about pinched nerves in the neck and started taking better care of my spine. Now I only get problems when I'm working at a properly shitty desk or in a crap chair.
Anyway.
She was using an early Wacom tablet instead of a mouse, and a pretty complicated macro system. Next job I met a guy who tried a BAT chording keyboard, never liked it, but set it up next to his regular keyboard and with a ton of programming macros for idiomatic code (define function, for loop, if/else if, etc).
Probably around that time I encountered an article about how French CS people were tut tutting about how ubiquitous computers were bad for us because we used to have to think before coding, so you thought about the problem instead of just throwing stuff at the wall to see what sticks.
From all of this and a bunch of other experiences I've sort of developed a thesis it really is worth it to spend at least 10% longer on a task thinking about ways to avoid grunt work and sources of human error. When new programmers ask me how to get better I point out that nobody is going to notice if it took them 4 hours or 4 and a half to do a task. After you think you're done ask yourself what you could do better with another 20 minutes and do it. (eventually you start doing this class of work preemptively, and you add other new things in).
And for those with RSI, this is existential. On any project there are a group of tasks for which it's more important they are done right the first time than that they are done quickly. Throw yourself onto those tasks.
Don't try to be a spaghetti hurler if you are struggling with RSI. Make yourself available to help and mentor junior people (it's amazing what you can learn about a system and about your own programming skills by observing all the bits people struggle with the most).
Michaelangelo is said to have spent a great deal of time in contemplation before starting work. We don't always get to do art but sometimes it's called for.
