In learning, choosing the right learning sources is the first step to success. A well-written article will get you to the basic idea from its first paragraph or even a sentence. Incremental reading is best suited for articles written in hypertext or in an encyclopedic manner. Ideally, each sentence you read has a contribution to your knowledge and is not useless without the sentences that follow.
Imagine that you would like to learn a few things about Gamal Abdel Nasser. You will, for example, import to SuperMemo an article about Nasser from Wikipedia. In the first sentence you will find out that "Gamal Abdel Nasser (1918 - 1970) was the second President of Egypt". If you are new to Nasser, you may be happy to just know he was the Egyptian president and safely jump to reading other articles. Thus you may delay the encounter with the historic role of Nasser and economize some time to finding out, for example, who Shimon Peres is. When you see the Nasser article for the second time, you might find that "He was followed by after President Muhammad Naguib and can be considered one of the most important Arab leaders in history". This piece of knowledge is also self-contained and you can patiently wait for your third encounter with Nasser. When you return the next time, you may conclude that another piece about Nasser is of lower priority: "Nasser was born in Alexandria". You can schedule the review of that piece in 2-3 years. Perhaps your interest in Nasser or in Alexandria will grow to the point that this knowledge will become relevant. If not, you can always dismiss or delete such an extract. Alternatively, you can skip a few paragraphs and extract a more important sentence: "In 1952, Nasser led the military coup against King Farouk I of Egypt". Even if your read individual sentences about Nasser in intervals lasting months, your knowledge will progressively expand and will become increasingly consolidated (esp. if you employ cloze deletions, which are mandatory for longer intervals).
Naturally, not all texts are are so well-suited for incremental reading. For example, a research paper may throw at you a detailed description of methods and leave results and conclusions for the end. In such cases, you may extract the abstract and delay the body of the paper by a period in which you believe the abstract will have been sufficiently processed. Then, if you are still interested in the article, you can schedule the methods well into the future (you will or will not read the methods depending on the conclusions of the article). You can schedule the results and the discussion into a less remote point in time, and proceed with reading the conclusions.
The hardest texts may not be suitable to reading in increments. For example, a piece of software code may need to be analyzed in its entirety before it reveals any useful meaning. In such cases, when the text (here the code) comes up in the incremental reading process, analyze it and verbalize your conclusions. The conclusions can then be processed incrementally. You will generate individual questions depending on which pieces of knowledge you consider important and which become volatile. The original computer code can be still retained in your collection as reference only.
When learning at the university, you do many courses in parallel. That's a macro version of incremental reading. Many people love to zap TV channels and play a chaotic version of incremental video with their TV set. Zapping may not be a recommended way of learning, but it won't leave your mind blank. Another example can be seen in people who have a habit of reading a few novels in parallel. Their limit on the number of novels comes from the limits of human memory. There is a breaking point beyond which a novel, if read in bursts separated by longer intervals, cannot be followed due to fading memories. Incremental reading is based on SuperMemo, and by definition is far less limited by your forgetful memory. The number of articles in the process can reach a hundred thousands, and given basic skills, you won't get confused.
|last reprioritisation on||suggested re-reading day|
|started reading on||finished reading on|