The advice system that you mentioned is such a good example of this! I recently realized that I had come to rely deeply on xref navigation for going to a definition and then xref-go-back/forward to hop between different jumps. I was bummed that it wasn't a more general backward/forward capability, but then I remembered the advice system. In like 10 minutes I whipped up some quick elisp to add some :before advice around various navigation functions to call xref-push-marker-stack. Not only did it work perfectly to turn the xref marker system into a general navigation capability, I also had complete control over which navigation functions push a marker and which do not.

I think this is the difference with Emacs. Not only is it primarily a lisp environment and then secondarily a text editor, but it goes above and beyond to add general capabilities for modifying its own behavior. With most any other piece of software I can think of, you are relying upon the developers to provide you with the exact APIs to control the software. With Emacs, you control the entire lisp environment, and therefore control the editor running within it.

edit: Siri typo

> spent too much time

Time has to be spent, no matter what: you can either adapt your workflow to match the design of specialized apps, or you can choose to go against the norm and manage it on your own terms.

I vividly remember my own misery when trying to find the most optimal, ideal note-taking solution. I went through Evernote, Workflowy, Google Todos, Google Keep, Todoist, Notion, Trelo, Remember the Milk, Obsidian, and some other options.

On the other hand, I have spent a long time studying and customizing Org-mode. I can say with absolute certainty that the net positive ROI from this effort is far greater than it could ever be with any other option.

"Ignorance is bliss", right? Less tech-savvy people genuinely live in blissful ignorance of the endless possibilities, the ways and the methods available. Programmers are different - they can spot your bullshit right away, "you're holding it wrong" vibes don't sit well with them, you don't tell them "users don't know what they want", they'd get annoyed and will try to find a better way - writing scripts, hacking your app, setting up unconventional keybindings, etc.

Emacs is for programmers, for tinkerers who prefer to deal with computing on their own terms. I know the feeling - some new, shiny app comes out that makes ripples in HN threads, and you'd feel old, cranky, and left out with your not-so-shiny, brutally simple yet efficient Emacs UI, thinking that maybe this is the time when you finally have to move away from it. Then you try that shiny new feature everyone so enthusiastically talks about and think, "Meh, that's it? Is that what they're so excited about? I can't believe now I have to use this shit because everyone else does..." And then a few weeks later, someone builds an Emacs package for it.

I never regret time spent on customizing Emacs because I've gotten to the point where I know exactly how to get the most use out of that exercise. It's nothing but pure, unadulterated pragmatics. The notion that Emacs users waste their time configuring it instead of doing real work is a myth. It's like saying that the cook spends too much time sharpening the blade instead of actual cooking. A great chef knows a great deal about his knives and always keeps them very sharp, but he's not in a knife-sharpening business.