Annotating in Moon Reader

Motivation

I wish to use my annotations in a few specific ways. I want to generate flashcardsAn important part of learning is testing. There is ample literature on the topic. An accessible and useful introduction is Make It Stick by Peter Brown et. al.

without continuously having to refer to the book. I also want to have a record of what I thought of a book as I was reading through itShort of using a tape recorder and transcribing it later, I don’t see a clean approach to truly capture what I was thinking as I read through a book. We’ll have to settle for annotations.

, along with an export of useful passages or quotes. I want to keep these exported annotations in my Obsidian Vault.

What do I highlight and why?

I’ve come across a few annotation workflows on the web but for now, I’ve settled on a few key categories. I use a separate color for each which makes for easy processing. I try not to get too feature specific in an effort to keep my setup portable.When I say portable, I’m trying to say that I want to keep my workflow intact when I switch to a different app or device This is why I am hesitant to use features like underlines and other annotation types provided by Moon+ Reader.

Outlines and passages

For writing flashcards, I find it critical to have the context of chapter and section headings. The easiest way to extract these from the book is to just highlight them.The reason I highlight headers is because when going through Moon+ Readers exports, it doesn’t give you context about the chapter. Instead, it gives us a section number.

I tag headings with #, sub headings with ## and so on.There’s a minor headache with this. Some books use problematic font sizes without any additional styling to distinguish section levels. This makes it difficult to figure out whether I’m at an H2 or at an H3 level. One way I’ve found around this is to disable styling for font sizes and headings with the “Disable CSS styles” option in the “Miscellaneous” settings.

The other thing I need for my flashcards are the salient points from a chapter itself. I just highlight these directly. If I can think of a good question on the flyI usually can’t.

, I put it in a note for that highlight.

All of these highlights are in gray, like in the screenshot below:

Vocabulary

I also highlight new words I come across in yellow. I usually highlight the entire sentence or enough for context when I encounter it later as word meanings vary depending on their usage. I add the word itself as a note.

Process Changes

I originally used the gray highlights to directly generate flashcards. This entailed the following:

• Mark up the paragraph that serves as the answer
• Write the question as a note
• Parse these later to generate a deck of flashcards

After a while, I began to realize that I was too eager to make flashcards. This led to me focusing on trivia instead of the bigger picture being presented by the author. It also turned out to be quite difficult to come up with a good question that covered a point the author was trying to make while deciphering the point itself.Essentially, poor multitasking. To quote a relevant passage from How To Read A Book:

We do want to stress, however, that most people, even many quite good readers, are unaware of the value of inspectional reading. They start a book on page one and plow steadily through it, without even reading the table of contents. They are thus faced with the task of achieving a superficial knowledge of the book at the same time that they are trying to understand it. That compounds the difficulty.

I therefore switched to creating outlines. I could then write unit tests for the outline itself to ensure I had “good coverage”. Depending on the coverage percent I desired, I could write as many or as few tests as necessary. For writing these tests, I would start off with the four classic questions that Mortimer Adler trained us to askSee: Adler, Mortimer J., and Charles Van Doren. How to read a book: The classic guide to intelligent reading. Simon and Schuster, 2014.

:

• What is it about as a whole?
• What is being said and how?
• Is it true in part or whole?
• What of it?

Of course, it is not necessary that these be asked for every section.

With these out of the way, I could then write other tests to reinforce other key ideas.One could also just try writing a fixed number of flashcards for each chapter. That way, you could limit yourself from going overboard, while at the same time, forcing yourself to think deeply about what the core ideas presented in the chapter were.

Examples of how I write flashcards will remain outside the scope of this article.

Exporting Annotations

I’ll briefly touch on this here but more detail can be found in the Moon Reader article. Exporting annotations is somewhat problematic in my workflow. While exporting is relatively straightforwardExport list to file from the Bookmarks menu.

, I need to sync highlights often so I can continue making flashcards without waiting until I’ve completed the book. This leads to me exporting fairly often and I’d like to optimize this soon. As a reminder, Readwise automatically syncs highlights as well.

Parsing Annotations

I now convert my annotations into simple pandoc markdown divs, since pandoc markdown has excellent interoperability. I use a Templater script.Templater is an Obsidian plugin that allows arbitrary JS execution. I’m also currently in the process of rewriting this as a plugin so more folks can make use of their exports.

It parses the Moon+ Reader export. So a single “highlight block” from a .mrexpt file:

#
100
Alice's Adventures in Wonderland
/sdcard/Books/MoonReader/Alice_en.epub
/sdcard/books/moonreader/alice_en.epub
1
0
685
133
1996532479
1632421631476


very much out of the way to hear the Rabbit say to itself,
0
0
0

will turn into:

:::

> very much out of the way to hear the Rabbit say to itself,

:::

Down the line, I will split each chapter into its own file to avoid having a single large file. I’m also thinking about how best to retain other metadata from the export, such as timestamp, color, section number, etc.I’m leaning towards storing it in a database or a CSV but I’m open to suggestions.

Using my annotations

With my parsed outline in place, I push the headers to another file with a script.I surprisingly couldn’t find a plugin that does this but perhaps I didn’t search thoroughly enough.

So in my “tests” file, I have:

# Chapter 1: Down the Rabbit Hole

# Chapter 2: The Pool of Tears

In a non-fiction book, I’d have the subsections outlined as well:

# 11: Agreeing or Disagreeing with an Author

## The Third Stage of Analytical Reading

## Judging the Author's Completeness

## Judging the Author's Soundness

## Prejudice and Judgement

With this, I can begin writing my flashcards. There’s plenty of other stuff I plan on doing with my annotations but that’s currently in the works. As always, feel free to drop me a message if you have any suggestions or criticisms.