For the last few months, I've been noting when I start and stop activities, at least on weekdays. I'm doing this just to get a sense of how long things take, not so I can optimize all of the texture out of my life.
There's a few surprises. Some development tasks that feel like they take forever, like adjusting a browser extension for a change in the site that it works on, actually take twenty minutes. And of course, some take days, which never surprises me.
Things that don't involve the computer that seem like no big deal because they don't require strenuous thinking sometimes take far longer than I think they will.
Just now, I set up a replacement camera that I got in the mail. I updated the firmware, hooked it up, adjusted its position. Then, I got the broken camera it replaced packaged up and repackaged it after finding out I had to include the AC adapter. I put away all the extra bits I got with the new camera, plus some SD card stuff that was on my desk. Finally, I looked up where I could ship it from.
All that took about an hour.
There's a temptation to want to excise this kind of thing from your life completely, but I'm not sure that is the right thing to do, even if you could do it. Doing that reminds me of the dominant-for-decades recommendation to optimize the consumption measurable nutrients like Vitamin A and protein instead of food in general.
That said, I'd rather automate this out rather say, programming, which, if you look at job listings lately, is the new hot thing. Of course I do not love hard debugging. But I do like having done it. (And I like having set things up to avoid it even better.) The same goes for learning how to do something I'm completely unfamiliar with, like shaders.
This thing where people just want to paper over anything they don't understand is like people that try out some martial art, decide it's too hard, and instead they'll sign up for a competition and bring a gun to it. e.g.
People "winning" Advent of Code with LLMs. These people are like states as described in Seeing Like a State — they pick metrics, then optimize metrics. This caused the Germans in the 1700s to adopt "scientific forestry," in which they optimized forests for the production of lumber by ripping out everything that wasn't a certain kind of tree. They had no idea how valuable the rest of the stuff in the forest was. People that decide that they're just going to code with LLMs, then just hope step in when they can't handle something do not understand
process knowledge.
While I'm moralizing, there are way more of this kind of person in programming than there used to be. Back at the start of the iPhone craze, there were a lot of people that showed up at Apple dev meetings that weren't saints, but they were small indie devs or people that made free stuff for universities. A couple of years later, I went to a conference at which I met some guys who were all planning to be rich via startups and also had a familiarity with cocaine, '80s Wall Street guy style. Those guys did have a respect for understanding what you made, though.