Sequences have been coming up a lot lately (have you read them?, pre-posting reqs, re-runs, and exercises). I received fantastic feedback to a recent request for advice, and the top suggestion that really solidified for me was to finish reading a set quantity of my book list and finish my "statement of non-belief" in order to have a conversation defuser concerning my deconversion.

In any case, I decided that before I read those books, I want to read the core sequences as well as How to Actually Change Your Mind. That's a healthy workload up ahead. I've actually read many of these here and there, but want to read systematically... and take notes.

Which brings me to my main point. I've been using emacs org-mode as my information manager (home and work) for some time now. I found it as a replacement to TiddlyWiki (which I still consider a close second) and it absolutely rocks. I primarily use it for intellectual property related notes at work, but my knowledge of it has advanced to using it to collect and plot data, write reports, track todos, and even track my time. It's a phenomenal program that has helped me with planning and information management a ton.

I've already used this to record notes on reading before, and I decided that I was going to use it for my notes on the sequences. As a result, I need to grab the articles and get them into org-mode. I did this pretty "kludgily" (wget plus a hand-tailored bash script that uses sed to transform the html back into characters and italics/bold into org-mode markup language). As a result, I now have org-mode files for all of Map & Territory. I thought I'd post on why I would even do this as well as to generate any discussion on whether or not this is useful for future sequences (I'll be going through those as well and will likely repeat the process).


Why in the world would you do this?

Well, first of all, it's really not that time-intensive. The bash script takes care of almost everything. The real time killer was the Intro to Bayes which had an insane amount of monospace tags for all the math. I think each article maybe took me 10min to download and clean up. Here's what we're working with after cleanup:

All articles shown collapsed (click to enlarge):

And with What do We Mean By Rationality? expanded:


Looks like something out of the 1970s? Probably because it is. You get used to looking at plain text, trust me :)

I'll be using this for notes, and many of you surely have systems for this already. One thing I like about org-mode that I find unique is that the notes and the text are in the same medium. When I read "real books" (the paper kind), I hate retyping quotes I want in my notes while I fiddle with holding the darned thing open on my lap as I type into the keyboard. Now, I can either copy/paste the quote into notes at the end of a file, or insert them inline where I want to say something. For example:

An inline note

Notes at the end:


In both cases, these notes are in their own "headlines" (shown by the preceding asterisks), which gives me some additional control over the text underneath them (more on that later).

What else? Well, for this particular article, I started a timer when I began reading and stopped it when I finished. This gives me something like this:

See the "Clock:..." line? When I'm done with this sequence, I'll have a fully filled out clock report with all the articles and the time it took me to get through them here:


That report is auto-generated, so whenever I clock in and out of a headline, I can update it and get the new summary.


I can also give myself todos. One thing that kills me about LW is browser tab self-destruction. I'll want to get through a couple posts and suddenly I have 20 tabs open with so many morsels of intellectual stimulation that I just don't want to go to bed until I've read them all, for fear of never finding them again. Not so anymore. Note all the little purple [Fn:#]s in the above pictures. I've translated all the hyperlinks into org-mode footnotes. If I'm reading along and come across one and want to know where it was going, I just press a key stroke and it transports me there. Here's a split emacs screen showing that:

In the top frame my cursor is on [Fn:3], I pressed Control-C twice and get teleported down to the Footnotes section and see it's a link to The Simple Truth. Great. No worries. Read that. What about [Fn:4] to The Conjunction Fallacy? Hmmm, seems important, but I really want to stay on task. Let's make a todo. Teleport down to the footnotes, copy the link, and add a todo:


Org-mode comes in with a built in filter to show you just your todos. You can add any number of org files to something called your "agenda list" and it will search through all of them and scrape out all the todos and present them to you. That's helpful. As an aside, I used to keep notes separate from todos because the notes were generally for longer term storage/reference, but the todos needed to be recalled and acted on. Unfortunately, separating them like this can separate the todo from its context, which means that if you keep notes and todos in separate programs, you needed to duplicate your notes in each place to recall what in the world that todo was about. Well, in org-mode, you can just add it right there and call it up at any time like so:


The bottom pane is the agenda view, which is showing me all todos for my Map & Territory org-mode file. As I add more, they'll show up. I can also add deadlines (shown under the todo headline in the top pane) to create goals for myself one when to get them done. I'm hoping that integrating org-mode into my reading will help keep me more on track rather than getting me lost in tab-land, fixated on reading stuff just because it was linked to.


New Comment