Okay, so I am curious how other people take notes from books, if they do it at all.

Goal: Attain a general working knowledge of a subject, to the point where I can answer any "standard" type problem on that topic.

Problem: Textbooks contain a lot of information, and only a fraction of it is information that I want to be able to quickly access later.

I haven't found a good workflow for quickly capturing essential information in a textbook such that it can be easily retrieved later.

I've sort of distilled this into two problems:

1. Where do you capture the information?

2. Which information do you capture?

Regarding #1, plain ol' paper works decently well, but it's not good for retrieving things you wrote weeks or months ago. On the other hand, it's a lot faster to write things down, especially if there's a lot of notation involved (e.g. matrices). Something like Obsidian is great b/c it's searchable, but it probably takes longer to typeset it all, and you can't have lots of useful note pages "in front of you" the same way you can with physical paper, especially if you want the textbook PDF open simultaneously on your laptop.

As for #2, I've often found it easy to write far too much when taking notes into some software like Obsidian. It's hard to know what to write down: every last theorem and proposition in the book? Probably not. Important definitions and theorems? How do you know what could be important later? How about proofs? Do you save past work of exercises you did, or just toss them?

Happy to hear what everyone's workflow is for this process (or, if you don't take notes at all from books, what the reasoning is behind that).