Piotr Wozniak of SuperMemo introduced in (3.5.2) Leitner system created an interesting reading method.
The reading method Incremental Reading is a computer-assisted reading using spaced repetition mechanism. It is closely related to making extracts from reading materials. First, we import the materials you want to read on the SuperMemo system. The material is called “text.”
If you read a text and think there is an important part, you select the range and press the shortcut key to create a new text. The system shows texts in expanding intervals, according to the SM-2 algorithm of spaced repetition. When you edit a text, the system makes the interval to the shortest interval. When the system shows a text, you may or may not read. Also, if you feel bored while reading a text, you can stop reading anytime.
What happens with this system? Each imported text is what you want to read once. However, when the system shows it after a while, often we don’t feel like reading. In those case, we do not edit the text. Its interval expands. It decreases the frequency of showing the text.
On the other hand, if you are interested in the text, you read it and extract important parts. It makes new texts, and they appear frequently. So, you read important parts repeatedly.
Even if a text looks unuseful, it is a psychological burden to make an explicit decision to throw it away. My failure is to keep it shown. In the Incremental Reading, the interval expands without an explicit decision, and the frequency of unuseful text gradually decreases. Moreover, because the system does not delete the text, you can search and find it if you think it is necessary.
By repeating this process, the number of valuable extracts increases. It is related to the concept of leverage reading. Also, because the system shows texts in random order, it promotes the connection between information from multiple sources. It is related to the concept of “Serendipity from random reading.”
I think Incremental Reading is a very promising way of reading, but it is still developing. For example, there is a problem of how to deal with a text which no longer needs to be edited. Piotr Wozniak recommends to convert it into a fill-in-the-blank test, and learn it in an ordinary spaced repetition method.
However, I am not satisfied with it. For example, I introduced the quote of Francis Bacon about reading in (4.1.1.6) Gradation between “Finding” and “Assembling”. Suppose you extracted it. Is it worth to remember as you can recall it without anything? I do not think so. We can easily find detailed information about it using a search engine. It is not important to remember concrete and literal information. It is important to understand the abstract concept that there are various ways of reading books, and optimal reading way depends on the book. In the case, it is strange to make a fill-in-the-blank test. I do not know how we should do for those texts.
The mechanism of Incremental Reading that fades out unuseful texts without an explicit decision is excellent. I hope this concept goes widely known and leads to the invention of better tools(artifact) and methodology.
--- This page is auto-translated from [/nishio/(4.5.3.2) Incremental Reading](https://scrapbox.io/nishio/(4.5.3.2) Incremental Reading) using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at [@nishio_en](https://twitter.com/nishio_en). I'm very happy to spread my thought to non-Japanese readers.