Right, iterating on your workflow is a normal part of work.
> I see a similar trend with org-mode users. It's great and all on a technical level (I've used it myself in the past) but it's easy to fall into the elisp and customization rabbit hole.
I think part of the issue is that people try to jump into the middle of something which has evolved without going through that evolution themselves. In other words, they need to go to the starting line and iterate from there rather than pushing through to the finish line. I have seen some complex workflows in Org mode, but those flows were built up over years.
The software analogy would be premature optimization. Developers attempting to use tools which are meant for scaling without knowing the how or why of these tools. If you don't know why you should use something, then you might not be to the point where you need it.
I learn the most about how to tackle a problem once I stumble across the need to fix the problem. The best tools are those I find when I then go in search for the fix.
I went through multiple stages with org-mode myself:
1) Just use it naively, instead of the ad-hoc note-taking I did before.
2) Customize it heavily, including a system for citation management. I definitely wasted a lot of hours reading documentation and blog post. But I also ended up using that system for 5+ years consistently.
3) Switch to paper notebooks for calendar, tasks and note-taking.
During 2), I often felt overwhelmed by the ever-growing list of TODOs, in particular those that scheduled at regular intervals.
With paper notebooks, and manual migration of tasks from one month to the next, it's easier to say "no" to things.
But, now that I've filled up three 200-page notebooks, I do struggle to find some old notes, or even remember that they exist.
So, maybe I will revert to an org-mode based "archive" for long-term ideas.
> I often felt overwhelmed by the ever-growing list of TODOs
This is an issue with a global set of todo lists. I follow a butchered BASB type of system. I just keep all ideas in resources folders. Todo items go in those folders, but those are just ideas rather than actual todo items. Then do an occasional review to determine what I want to work on. Whatever I decide gets a folder in projects and todo items get actual deadlines. If I can't work on it, then it gets archived and maybe I'll go back to it later. The only todo items I'll look at are in those project folders and they make progress based on deadlines or they get ditched.
> I think part of the issue is that people try to jump into the middle of something which has evolved without going through that evolution themselves
This is so true. I use a small subset of Org's features, and I have grown my workflow really slowly.
I also have very minimal dotfiles, yet I consider myself an advanced Emacs and Unix user.
It doesn't need to be a timesink. Also, as you say, complicated workflows are the consequence of iterating through many years and are likely paying off the efforts multiplied by a big factor.
> I see a similar trend with org-mode users. It's great and all on a technical level (I've used it myself in the past) but it's easy to fall into the elisp and customization rabbit hole.
I think part of the issue is that people try to jump into the middle of something which has evolved without going through that evolution themselves. In other words, they need to go to the starting line and iterate from there rather than pushing through to the finish line. I have seen some complex workflows in Org mode, but those flows were built up over years.
The software analogy would be premature optimization. Developers attempting to use tools which are meant for scaling without knowing the how or why of these tools. If you don't know why you should use something, then you might not be to the point where you need it.
I learn the most about how to tackle a problem once I stumble across the need to fix the problem. The best tools are those I find when I then go in search for the fix.