2. After I’ve completed a book, I’ll go back and type out the notes into OneNote, I usually have my raw notes (what I’ve underlined), my synthesized notes (the best of the raw notes), and then finally some kind of concept map for the overarching ideas.
3. To completed notes I add a “#review” tag into the note and then every two weeks I’ll do spaced repetitions on the notes. At some point I’ll add a tag like “#long-term-review” and then only look at it every 6 months.
4. Use physical books and use a pen, not highlighter. Physical books allow you to read faster and a hundred other things that would be too long to get into here, and pens are better because you can underline, circle words, and write in the margin all with one device.
5. Learn how to distinguish good books from bad books. Bad books tend to have very “markety” like “How Anybody Can Be Creative!” while good books have more boring titles like “Reflections on Lateral Thinking” by XYZ University Press. Also University Press’s are almost always a good buy. You will never find those books unless you look, consider the average marketing budget for a University Press book is about $5,000, but they’re one of the best untapped resources for knowledge in the world currently. Amazing that they aren’t more popular.
In fairness, I'm bad at being consistent. But once you have a mind map or similar (generally with themes or topics as title) you have an exportable resource.
Once you have an exportable resource you can release the kraken (let slip the dogs of war?) and do whatever you want with it. Things like spaced repetition, reminders of things you haven't looked at or thought about over time, totally random, etc.
It's cool.
A "random" link from the past, or a book passage, or a note tree, or even just exploring a random path, might directly solve a current problem or push you to think in a different direction, or jog a thought pattern you didn't know you had.
I'm currently looking for a new solution that has the ability to add arbitrary back-ends to retrieve data, probably Electron-based.
Now I am not saying that this is the most efficient way to cram for a test. But I don't think I have any tests of the kind that can be crammed for coming up soon so I don't let the methods that are good for test cramming shape the way I read...but it took me some time to get to this point.
Good luck.