I pick up a lot of information from books. We have an amazing public library system, and I’m at the library at least once a week. Here’s my current workflow for reading and taking notes on books.
Finding books to read: I check the library’s new releases for interesting titles. I have a Ruby script that extracts the information and puts it into a text file. I delete anything that doesn’t interest me, and then I copy the IDs into another Ruby on Rails system that requests all the books for me. Sometimes I search for books by topic or get recommendations from other people.
Reading books: I’ve successfully weaned myself off the bad habit of folding over the corners of book pages (dogearing). Instead, I use book darts to point to the passages that I want to copy into my notes. I have only one tin of them, so that encourages me to harvest the notes from books before moving on to other books. If I don’t have my book darts handy, I use strips of paper, but they’re not as awesome.
I also keep paper and pen handy (index cards, or the small notebook I always have in my vest) so that I can take notes on ideas, questions, and other things that aren’t directly in the text of the book.
I mostly read non-fiction, so that’s easy to skim for interesting bits. I usually check the table of contents to get an overview of the book. I have no qualms about jumping straight to specific chapters and then wandering around a bit, or even picking just a few pages out of a book. I rarely use the index (and most books don’t have a good one anyway), although maybe I should check that more often.
Taking notes: I keep individual text files named after the title of the book (and sometimes authors as well). I usually include the ISBN so that I can easily look up a book later. The text files contain quotes, ideas, TODOs, and other notes.
If no one else is around and I feel like patiently dealing with speech recognition, I open Dragon Naturally Speaking and dictate the passages from the books. This helps me train Dragon’s speech model as well, which might be handy someday. If other people are around, though, I’ll just type in the segments from the book.
I usually type in my paper notes so that they’re more coherent (since I tend to write keywords instead of full thoughts). If I want to scan my index cards or notebook pages, I pop those into my Fujitsu Scansnap ix500 and scan them as JPG. I convert the JPGs to PNG and rename them with the date and title, and then I move them to a folder that gets automatically imported into Evernote. Evernote lets me search for text, and it also tries to find text even in scans of handwriting. It’s not perfect, but it’s decent, and I’ve learned to write more clearly because of that.
Looking things up: Since most of my notes are text, I can use
grep to search through them.
Sharing what I know: I sometimes include excerpts or ideas in my blog posts. When I do, I link to the blog post from my text notes as well, so I can see what I’ve digested further and shared. If I’m reading a book that I know I’ll want to share with other people later (or if the authors have asked me nicely =) ), I sometimes visually summarize the book.
Following up on ideas: I add TODOs to my Org Mode agenda. I can also schedule reminders for things. I’m a little hesitant to add my books directory to the
org-agenda-files list that Org Mode checks for TODO items (I have hundreds of book notes now!), so I’ve defined a custom agenda command that looks at just the book directory instead. Alternatively, it’s easy enough to
Planned improvements: I’m curious about the idea of a syntopicon, which I picked up from Adler and van Doren’s How to Read a Book. A syntopicon is a map of ideas across multiple books. With Org Mode’s support for indexing and links, I should be able to make something like it. I’m also looking forward to writing more about what I do with what I’m learning from books. This helps me challenge myself to learn actively instead of just letting a book flow through my brain.
If you’re curious, you can read about my past workflows: