View source for Incremental learning

__TOC__

<blockquote>
<p class="quote">''Our cultures don't encourage us to think much about learning. Instead we regard it as something that just happens to us. But learning must itself consist of sets of skills we grow ourselves; we start with only some of them and slowly grow the rest. Why don't more people keep on learning more and better learning skills? Because it's not rewarded right away, its payoff has a long delay''.</p>
<p class="author">Marvin Minsky</p>
</blockquote>

= What is incremental learning? =

This article describes the fastest avenue towards rock-solid lifetime knowledge: '''Incremental Learning'''.

Incremental learning is the fastest and the most comprehensive way of learning available to students at the moment of writing (2013).

Incremental learning is a consolidation of computer-based techniques that accelerate and optimize the process of learning from all conceivable material available in electronic form, and not only.

Currently, [http://www.super-memo.com/supermemo17.html SuperMemo] is the only software that implements incremental learning. In SuperMemo, the student feeds the program with all forms of learning material and/or data (texts, pictures, videos, sounds, etc.). Those learning materials are then gradually converted into durable knowledge that can last a lifetime.

<div class="bs-callout bs-callout-warning">
'''Incremental learning helps the student convert all forms of learning material into durable and lasting memories.'''
</div>

In incremental learning, the student usually remembers 95% of his or her top priority material. That knowledge is relatively stable and lasts in student's memory as long as the process continues, and well beyond.

<div class="bs-callout bs-callout-warning">
'''Incremental learning easily ensures 95% recall of top-priority learning material for lifetime''' (as long as the student ensures a regular review along the prescription provided by the program).
</div>

The cost of high [[Glossary:Retention|knowledge retention]] is very small when compared with various traditional learning methods. For example, in learning a language, the [http://super-memory.com/english/ae.htm vocabulary of an educated native speaker] can be retained in SuperMemo at the cost of 20 minutes per day in the first years of the process, and mere minutes in later years (assuming the original set is acquired in portions spread over 4 years in 30-50 min. sessions).

<div class="bs-callout bs-callout-warning">
'''Incremental learning ensures high recall at a fraction of the cost in time (as compared to textbook learning).'''
</div>

The incremental learning derives its name from the incremental nature of the learning process. In incremental learning, all facets of knowledge receive a regular treatment, and there is a regular inflow of new knowledge that builds upon the past knowledge. In incremental learning, the student sits in the driving seat and determines which knowledge should be mastered. He or she determines when this happens, with what degree of detail, at what [[Glossary:Priority|priority]], and at what desired degree of [[Glossary:Retention|recall/retention]]. For example, in a single session, the student may learn a few facts of geography, discover a few rules of healthy lifestyle, figure out a few statistical formulas, read a couple of paragraphs from a friend's blog, process a few minutes of his home video collection, annotate a few family pictures, watch a few pieces from his YouTube video collection, and read a few articles in subjects related to a forthcoming exam. In other words, all areas of knowledge keep growing in parallel in proportion to interests and importance.

Typical learning at school puts an emphasis on a few areas of knowledge and neglects all the remaining areas. A medical student may spend a few months mastering anatomy, while gradually forgetting his biochemistry material in the meantime (or the other way round). At the same time, he or she will not find time to study important issues of the day that will always depend on a given person in a given context. With blinkers imposed by the heavy load of school material, the student may never find time, for example, to figure out what incremental learning is. Narrow horizons and narrow perspectives only make it harder to further rationalize the selection of the learning material.

<div class="bs-callout bs-callout-warning">
'''Incremental learning is the opposite of the irrational school system learning in which a heavy focus is put on just a few areas of knowledge in a semester (at the cost of other, equally important, areas of learning).'''
</div>

== General outline of incremental learning ==

In incremental learning, you acquire and maintain knowledge using the following steps:

* importing knowledge from various electronic and non-electronic sources (e.g. articles on the web, YouTube videos, music files, pictures from your camera, e-mails, scanned paper notes, etc.)
* [[#Priority queue|prioritizing knowledge]] for incremental processing (e.g. high [[Glossary:Priority|priority]] for physics, low [[Glossary:Priority|priority]] for movie trivia, etc.). Incremental approach means processing knowledge in small bits and in small steps
* gradually converting the learning materials into lasting knowledge in your memory. This conversion may also produce an easily searchable and well-annotated computer media archive that does not even need to be part of the learning process
* expanding creatively upon the acquired knowledge (e.g. in the process of [[#Incremental writing|incremental writing]], [[#Incremental problem solving|problem solving]], etc.)

<div class="bs-callout bs-callout-warning">
'''With incremental learning, you can consolidate all sources of knowledge, and convert information into lifetime memories at the chosen cost in time, and along strictly defined goals and priorities.'''
</div>

== Components of incremental learning ==

Incremental learning tools differ substantially for various forms of learning material, media, and goals. Here are the main components of incremental learning:

* [[#Incremental reading|incremental reading]]
* [[#Visual learning|incremental image learning]] (or visual learning)
* [[#Incremental video|incremental video]]
* [[#Incremental audio|incremental audio]]
* [[#Incremental mail processing|incremental mail processing]]
* incremental creative elaboration (incl. [[#Incremental problem solving|incremental problem solving]], [[#Incremental writing|incremental writing]], etc.)

With the rich toolset offered by incremental learning, all reading, learning, viewing, archiving, and annotation functions can be delegated to SuperMemo. This goes far beyond standard learning and includes personal notes, home videos, lectures available in audio and video formats, YouTube material, family photo-albums, diaries, audio files, scanned paper materials, etc.

The oldest, most popular, and the most mature component of incremental learning is '''incremental reading'''. We will use incremental reading as the comprehensive introduction to other forms of incremental learning.

== The value of interruption in learning ==

In incremental learning, we often quickly move from one subject to another. Such interruptions may occur many times during a single learning day. When people first learn about this incremental methodology they immediately ask ''"Why interrupt? Aren't these the prime principles of human endeavour to be thorough, persist, and do things right to the end?"''

The 3 main advantages of interruption in learning are:

* '''improving memory''': spaced learning has long been proven dramatically more efficient than conglomerate learning massed in time
* '''improving learning choices/priorities''': unless the learning material has been pre-selected by a higher authority, student's own choices require prioritization, which in turn requires preview. Previewing is a form of interruption. Regular interruption allows of prioritizing on the go
* '''improving attention''': whenever attention declines, change of the subject is the simplest remedy other than taking a definite break from learning

As for the disadvantages ... there are none! Simply put: '''interruption is optional!''' It is true that incremental learning may lead to "learning impatience" and "craving interruption", however, these have never been proven detrimental beyond showing that once you employ incremental learning, you may never want to go back to traditional "book at a time" learning. Nevertheless, you should not forget that schools are incremental too. Just on a slightly moderate scale. Schools employ interruption when kids move from geography to physics, or when they close the books for the day.

Once the art of incremental learning is mastered, the advantages go far beyond the advantages of the interruption or [[Glossary:Spaced_repetition|spaced repetition]]. Here is a shortlist (for a detailed discussion see: [[#Advantages of incremental reading|Advantages of incremental learning]]).

* '''massive learning''' - you learn  more than you thought your memory can hold 
* '''95% knowledge retention''' - you nearly eliminate the problem of forgetting
* '''lifetime memories''' - your memories will last for life (as long as you stick with the regular review schedule based on [[Glossary:Spaced_repetition|spaced repetition]]) 
* '''comprehensive learning''' on all fronts (rather than the school-like focus on 2-4 majors)
* '''better understanding''' of the studied subject is assisted by moderation in consuming details, and easy inclusion of explanatory material (e.g. from dictionaries and/or encyclopedias)
* '''better consolidation of the knowledge structure''' by incremental approach, interrupted learning, spacing, and slotting in of the new knowledge. Contrary to popular belief, incremental learning helps you keep the big picture in your mind
* '''better attention''' by focusing on a single issue at a time without ever missing a detail, and by remedying attention deficits with a constant change of the learning material
* '''creativity''' - by encountering different subjects in unpredictable sequences, your creativity soars. You can employ it, for example, in the process of [[#Incremental_problem_solving|incremental problem solving]] or [[#Incremental_writing|incremental writing]] (this article was written using incremental writing tools in SuperMemo)
* '''battling chaos''' - it is easier to resolve contradictions in SuperMemo, e.g. when processing new research with contradictory claims and findings. Unlike your memory in "real life" where you keep oscillating between contradictions, SuperMemo does not tolerate information discrepancy. Contradictory material converges up to a point when you realize you need to decide on the nature of the truth
* '''all knowledge is well prioritized'''
* '''all knowledge is easily searchable'''
* '''all knowledge is quantifiable''' (size, retention, workload, etc.)
* '''stresslessness''' - nothing frees your mind for learning efficiently as the sense that no detail will ever be missed and you can focus on a single problem at a time while delegating other problems to later
* '''fun''' - once you master incremental learning, it can truly be the best part of your day with few other earthly pleasures giving you as much satisfaction as new useful knowledge

In short, with incremental learning you learn fast, you acquire massive loads of knowledge, retain memories for life, remember almost all that you have learned, understand things better, develop harmoniously in all directions, enhance your creativity, and all that while having incredible fun! If that sounds too good to be true, please read more below or just give it a solid try. For a detailed explanation see: [[#Advantages of incremental learning|Advantages of incremental learning]].

== Interruption is not a problem ==

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 [http://en.wikipedia.org/wiki/Gamal_Abdel_Nasser Gamal Abdel Nasser]. You will, for example, import to SuperMemo an article about Nasser from [http://en.wikipedia.org Wikipedia]. In the first sentence you will find out that &quot;''Gamal Abdel Nasser (1918 - 1970) was the second President of Egypt''&quot;. 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 [http://en.wikipedia.org/wiki/Shimon_Peres Shimon Peres] is. When you see the Nasser article for the second time, you might find that &quot;''He was followed by after President Muhammad Naguib and can be considered one of the most important Arab leaders in history''&quot;. 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: &quot;''Nasser was born in Alexandria''&quot;. You can schedule the review of that piece in 2-3 years. Perhaps your interest in Nasser or in [http://en.wikipedia.org/wiki/Alexandria Alexandria] will grow to the point that this knowledge will become relevant. If not, you can always [[Glossary:Dismiss|dismiss]] or delete such an [[Glossary:Extract|extract]]. Alternatively, you can skip a few paragraphs and [[Glossary:Extract|extract]] a more important sentence: &quot;''In 1952, Nasser led the military coup against King Farouk I of Egypt''&quot;. Even if your read individual sentences about Nasser in [[Glossary:Interval|intervals]] lasting months, your knowledge will progressively expand and will become increasingly consolidated (esp. if you employ [[Glossary:Cloze_deletion|cloze deletions]], which are mandatory for longer [[Glossary:Interval|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 [[Glossary:Extract|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 [[Glossary:Collection|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 [[Glossary:Interval|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.

== Complexity of incremental learning ==

Unlike classic SuperMemo, incremental learning requires quite a lot of experience and training before it becomes effective. However, your investment will be returned manifold once you become proficient with the method.

Incremental learning is a consolidation of technologies that have been in development for nearly 3 decades. It is still in the process of maturing and it is still pretty complex. It requires skills that take months to develop. It requires your own strategies that may mature over years. Moreover, incremental learning requires the mastery of SuperMemo, which has been optimized for professional use. As such it is not beginner friendly.

Users complain that SuperMemo has a steep learning curve. They are right. SuperMemo has been optimized to make a life of a pro easy. It makes life of beginners hard because it does not ever compromise the learning efficiency for sleekness or marketing value. Take the [[#Priority queue|priority queue]] as an example. Nearly everyone asks why the articles of highest value have a [[Glossary:Priority|priority]] of 0% rather than the obvious 100%. They ask: ''"Why is SuperMemo always keeping things upside down?"''. They got a point. However, no pro user would ever swap the ease of typing 1, 2, 3 for his top priority material as opposed to 99, 98 or 97. Those dilemmas slow down the adoption of SuperMemo. However, once you become a pro, you will appreciate this approach and will more likely become a lifelong devotee.

= Incremental reading =

{{:Incremental reading}}

= References =

{{:References}}

= Additional skills in incremental learning =

== Recognizing unsuitable texts ==

Best articles for [[Glossary:Incremental_reading|incremental reading]] are fact-rich and context-rich. You need to develop your own rules for selecting quality reading material. Nothing can substitute for your own experience. You will learn to identify texts that are hard to process and yield lower reading efficiency.

Let's consider two extremes:

* '''Wikipedia''': [http://en.wikipedia.org Wikipedia] is great because it is crowd-sourced and many authors edit only a small section of a text in total abstraction of the rest of the article. This is why even a small portion of the text (as in [[Glossary:Incremental_reading|incremental reading]]) will usually contain all relevant context
* '''Fiction''': Fiction is best read linearly and you can just use a paper bookmark in a paper book to do as well as you would do with [[Glossary:Incremental_reading|incremental reading]]

Many articles fall in between [http://en.wikipedia.org Wikipedia] and fiction extremes. Where [http://en.wikipedia.org Wikipedia] would say &quot;the http protocol&quot;, a typical article might just say &quot;the protocol&quot; (if &quot;the&quot; is clear from preceding passages). Some text are peppered with &quot;as I mentioned in the previous chapter&quot;, or &quot;go to the next section&quot;, or &quot;the three points explained in the previous chapter&quot;, etc. Abuse of working memory in text writing makes [[Glossary:Incremental_reading|incremental reading]] difficult. This means that if authors use a lot of &quot;referential ambiguity&quot;, the texts are not good for [[Glossary:Incremental_reading|incremental reading]]. Example ambiguity keywords: &quot;the&quot;, &quot;they&quot;, &quot;he&quot;, &quot;it&quot;, etc.

Many articles are also filled with irrelevant chaff ([http://super-memory.com super-memory.com] isn't impervious to that weakness). Too much beating about the bush without clearly stating the conclusions that are most important to the reader. &quot;Speculative philosophy&quot; might be a good inspirational read, but probably not for [[Glossary:Incremental_reading|incremental reading]]. You need to decide.

Some narratives should just best be read passively. They may be a compilation of facts that are generally obvious. In such cases you can just read and [[Glossary:Dismiss|dismiss]]. Or you can read and schedule another review in a month or in a year (if you worry you miss something important). Or you can try to write, in your own words, a sentence or two on what new things you have learned from the narrative. Your sentence would shortly extract the quintessence from an otherwise lengthy passage. If it is meaningful and quintessential, you shall find little trouble with locating keywords suitable for [[Glossary:Cloze_deletion|clozing]].

You need to develop your own rules for deciding which articles are good for reading. The chief rule might be: import anything that looks interesting, start reading, and if you recognize tell-tale problems, just
deprioritize or delete or fish for a few highlights and then delete.

Remember that you need to differentiate between [[#Example: Unsuitable texts|unsuitable texts]] and [[#Handling incomprehensible articles|difficult texts]].

== Handling incomprehensible articles ==

In [[Glossary:Incremental_reading|incremental reading]], you will often encounter material that is difficult to understand. You will need to develop analytical skills that will help you identify the reasons for the difficulties. If the culprit is the author, delete the article. If you need to digest other pieces of your [[Glossary:Collection|collection]] first, delay the article. If you need more knowledge, delay the article and import more knowledge that will be needed to boost understanding. Do not forget that '''some texts make an inherently poor material for [[Glossary:Incremental_reading|incremental reading]]''' (e.g. descriptions of scientific experiments, mathematical derivations, programming examples in source code, case studies, etc.). In such cases, use traditional methods of thorough analysis, summarize results of your analysis, and use SuperMemo to keep track of your own summaries. See: [[#Recognizing unsuitable texts|Recognizing unsuitable texts]] (incl. [[#Example: Unsuitable texts|example]]).

This is how you can approach complexity in [[Glossary:Incremental_reading|incremental reading]]:

# Start reading the article from the top. Once you find a difficult fragment, analyze it, and diagnose the reasons for your comprehension problems
# If the rest of the article does not depend much on the difficult fragment, extract it, and keep on reading
# If the rest of the article cannot be understood without understanding the difficult fragment choose one of the following:
#* if you need more knowledge to understand the fragment: postpone the article ('''[[Element menu#Learning|Learning]] : [[Element menu#Reschedule|Reschedule]]''' from the [[element menu]] or ''Ctrl+J'')
#* if the fragment is hopelessly intricate and leaves no hope for the future (e.g. because of wrong grammar, wording, formulation, logic, etc.), delete the article, and search for alternative material
# If you decide to postpone the article with ''Ctrl+J'', decide what new knowledge you will need before getting back to the difficult fragment. List dictionary entries, encyclopedia articles, articles on the net that you will need to process before going any further. Schedule the search for materials as separate [[Glossary:Topic|topics]] or try to search for new knowledge instantly
# Estimate the earliest time when you hope you will be able to understand the difficult article and use the appropriate [[Glossary:Interval|interval]] with ''Ctrl+J''. If the article includes high priority knowledge, it is always better to err on the safe side and provide a too early review

== Randomizing repetitions ==

You can execute [[Glossary:Outstanding_material|outstanding repetitions]] in a [[Glossary:Subset|subset]]. If you would like to use a random sequence, follow these steps:

# open the [[Glossary:Subset|element subset]] in the [[browser]]
# choose '''[[Browser menu#Random|Random]] : Randomize browser''' on the [[browser menu]] (''Shift+Ctrl+F11'') 
# choose '''[[Browser menu#Learn|Learn]]''' on the [[browser menu]] (''Ctrl+L'')

== Read toolbar ==

{{:Read_toolbar}}

== One memory, one action ==

In [[Glossary:Incremental_reading|incremental reading]], you achieve highest efficiency if your process knowledge in small steps separated in time. This way you can accomplish a good memory effect at little processing effort. However, many users fall into traps of inefficiency where too little or too much work is done on a subject in a single review.

=== Futile review ===

Futile review is an example of insufficient work done (one action, zero memory). Futile review is born in this mental scenario: you see a [[Glossary:Topic|topic]] and think: ''I am in no mood for this material now. Let's execute '''Next repetition'''.'' This is wrong! You must take action or you will loop into wasting time and learning little! It is a cardinal sin to execute processing operations without actually learning anything. When a [[Glossary:Topic|topic]] arrives, you may have a dozen of excuses: "''I do not like this one. Let's do it tomorrow''". Or "''I am too sleepy for this one''". Or "''This one will take too much time.''" If you find yourself in a loop and constantly rescheduling the same [[Glossary:Topic|topic]], or spending time rescheduling a number of [[Glossary:Topic|topics]], you are hurting the efficiency of learning! This is the time that could be spent on more productive steps.

If you do not want [[Glossary:Incremental_reading|incremental reading]] to become a waste of time, you must '''always take some action''' when you see a [[Glossary:Topic|topic]]. For example:

* read a sentence and delete it, or mark as <span class="ignore">ignore</span>, or extract it
* if the article is not vital for your further progress, set low [[Glossary:Priority|priority]] (e.g. 99%)
* for the material that you won't have time for in a month or more, set the [[Glossary:Interval|interval]] (e.g. to 333 days)

In other words, either make tiny inroads into the text, or mark it clearly as low priority or to-be-done later. Consider also '''[[Element menu#Delete element|Delete]]''' or '''[[Element menu#Done|Done!]]'''

=== Item perfectionism ===

Another facet of the same problem is taking too many actions on a single piece of information. It is highly inefficient to work on colors, fonts, pictures, [[Glossary:Priority|priorities]], etc. during a single repetition. All those actions can be spread over time. Naturally, setting the right [[Glossary:Priority|priority]] is one of the most important steps. Perhaps a piece of information is not important enough to ever squeeze through your crowded learning. If so, you will save tons of time on not doing colors and styles.

Here also you should remember: '''one memory, one action'''. Each time you review a piece of information, you are allowed to do an edit, font change, template change, concept group change, etc. However, unless you can do all your actions in a single burst of machine gun keyboard strikes, or unless some actions are associated with learning new things, you should rather limit your actions to a single step per single repetition.

'''One memory, one action''' rule demands that '''every operation in incremental learning should leave a trace in your memory'''. It also says that '''one operation on a piece of data is better than two operations at the same time'''

=== Example: incremental item structuring ===

The following [[Glossary:Item|item]] may look like a violation of the [https://supermemo.guru/wiki/20_rules_of_knowledge_formulation 20 rules]:

<blockquote>
'''Question''': Inflammation is produced by [https://en.wikipedia.org/wiki/Eicosanoid eicosanoids] and [https://en.wikipedia.org/wiki/Cytokine cytokines]. Eicosanoids include (1) [https://en.wikipedia.org/wiki/Prostaglandin prostaglandins]  that produce [https://en.wikipedia.org/wiki/Fever fever] and [https://en.wikipedia.org/wiki/Vasodilator vasodilation], and (2) <span class="cloze">[...]</span> that attract certain [https://en.wikipedia.org/wiki/White_blood_cell white blood cells]<br>
'''Answer''': [https://en.wikipedia.org/wiki/Leukotriene leukotrienes]
</blockquote>

However, the [https://www.super-memory.com/articles/20rules.htm 20 rules] were written in 1999 for classical SuperMemo. Incremental learning is incremental across the spectrum of rules and principles. In particular, formulating [[Glossary:Item|items]] and building understanding are incremental too.

According to the '''one memory, one action''' principle, the presented [[Glossary:Item|item]] will assume its final shape some time in the undefined future (or never, if its [[Glossary:Priority|priority]] is not high enough). It must be processed incrementally due to the following factors:
# incremental build up of comprehension, and
# incremental reformulation that requires time.

The [https://www.super-memory.com/articles/20rules.htm 20 rules] say &quot;'''Do not memorize until you understand'''&quot;, however, understanding is also an incremental process. Converting this [[Glossary:Item|item]] to plain &quot;''What eicosanoids attract white blood cells?''&quot; might make sense only if the student fully understands and remembers the hierarchy of inflammation factors, and involved eicosanoids. If this is not the case, carrying the context in the shape of this complex [[Glossary:Item|item]] is a form of transitional stage between a [[Glossary:Topic|topic]] and an [[Glossary:Item|item]]. The [[Glossary:Item|item]] still asks the question vital for [[Glossary:Active_recall|active recall]]. However, it also makes sure that full context is provided until the rest of the knowledge structure is firmly established in student's mind. In [[Glossary:Incremental_reading|incremental reading]], the order of reading is often chaotic, the understanding is incremental, and the effort to build a solid knowledge structure is gradual too.

In addition to the incremental buildup of comprehension, extensive edits of [[Glossary:Item|items]] are costly (esp. a total rewrite of the item to a plain question). In fact, one of the main advantages of [[Glossary:Incremental_reading|incremental reading]] is the minimum need for typing. This is why we use electronic sources in the first place (instead of just books that offer no disadvantage when entire [[Glossary:Item|items]] are typed in anyway). This is why an important efficiency principle in incremental learning is to minimize edits by complying with the '''one memory, one action''' principle.

Let us see how the presented [[Glossary:Item|item]] might evolve in successive repetitions. Note that all edit steps my proceed only with solidification of related knowledge (i.e. a single repetition may actually bring no edits at all). The execution of those steps will also be somewhat dependent on item [[Glossary:Priority|priority]]. High priority items will receive more exposure, more processing and will demand better formulation quality.

==== Take 1: original complex item ====

<div class="center">
[[File:SuperMemo_Commander.jpg|800px|SuperMemo: A cloze deletion whose formulation is improved incrementally over many repetitions in accordance with the principle: "one memory, one action"]]
<blockquote>'''''Figure''': A [[Glossary:Cloze_deletion|cloze deletion]] related to inflammation, with a formulation that seems to violate the [https://www.super-memory.com/articles/20rules.htm 20 rules]. This cloze will be improved incrementally over many repetitions along the principle: "'''one memory, one action'''"''</blockquote>
</div>

==== Take 2: moving a clue to the answer field ====

<blockquote>
'''Question''': Inflammation is produced by [http://en.wikipedia.org/wiki/Eicosanoid eicosanoids] and [http://en.wikipedia.org/wiki/Cytokine cytokines]. Eicosanoids include (1) [http://en.wikipedia.org/wiki/Prostaglandin prostaglandins] that produce [http://en.wikipedia.org/wiki/Fever fever] and [http://en.wikipedia.org/wiki/Vasodilator vasodilation], and (2) <span class="cloze">[...]</span><br>
'''Answer''': [http://en.wikipedia.org/wiki/Leukotriene leukotrienes] (that attract [http://en.wikipedia.org/wiki/White_blood_cell white blood cells])
</blockquote>

==== Take 3: removing the prelude ====

<blockquote>
'''Question''': Inflammation: Eicosanoids include (1) [http://en.wikipedia.org/wiki/Prostaglandin prostaglandins] that produce [http://en.wikipedia.org/wiki/Fever fever] and [http://en.wikipedia.org/wiki/Vasodilator vasodilation], and (2) <span class="cloze">[...]</span><br>
'''Answer''': [http://en.wikipedia.org/wiki/Leukotriene leukotrienes] (that attract [http://en.wikipedia.org/wiki/White_blood_cell white blood cells])
</blockquote>

==== Take 4: bare bones item ====

<blockquote>
'''Question''': Inflammation: Eicosanoids include (1) [http://en.wikipedia.org/wiki/Prostaglandin prostaglandins], and (2) <span class="cloze">[...]</span><br>
'''Answer''': [http://en.wikipedia.org/wiki/Leukotriene leukotrienes]
</blockquote>

==== Further reading ==== 
* [https://supermemopedia.com/wiki/Dismembering_complex_topics_before_or_after_learning_the_structure_of_knowledge Doubts about incremental knowledge structuring]
* [https://supermemo.guru/wiki/Knowledge_darwinism Knowledge Darwinism]

== Learning lists ==

Lists and sets are difficult to remember. It is hard to remember the whole set of countries that belong to the European Union. When learning lists, you should rather decompose the problem into smaller subproblems.

Let us consider an example in which you want to memorize the entire sequence of letters in the alphabet. It won't be very effective if you use the following item:

<blockquote>
'''Question''': What is the sequence of letters in the alphabet?<br>
'''Answer''': A, B, C, D, E, F, G, H, I, J, K, L, M, N, O, P, Q, R, S, T, U, V, W, X, Y, Z
</blockquote>

You will notice that you frequently stumble on parts of the sequence and need to stop repetitions just to exercise the entire sequence in the traditional way (like we all learn poems by rote).

However, you can approach this in a way that guarantees quick effects:

<blockquote>
'''Question''': What is the sequence of letters in the alphabet between A and E?<br>
'''Answer''': A, B, C, D, E
</blockquote>

<blockquote>
'''Question''': What is the sequence of letters in the alphabet between D and H?<br>
'''Answer''': D, E, F, G, H
</blockquote>

<blockquote>
'''Question''': What is the sequence of letters in the alphabet between G and K?<br>
'''Answer''': G, H, I, J, K
</blockquote>

<blockquote>etc. etc.</blockquote>

After 2-3 weeks of repetitions, you may take on an extra task of recalling the whole sequence after each repetition of these simplified [[Glossary:Item|items]]. This will make sure you can recite the entire alphabet quickly. You will also frequently rehearse that parts of the sequence that are harder for your memory (e.g. ''V, W, X, Y, Z'') as opposed to those that are much simpler (e.g. ''A, B, C, D, E'').

=== Practical problems with memorizing lists ===

You may develop a good methodology for memorizing lists in SuperMemo. However, you may later discover that the memorized lists are not very useful in real life, or worse, your memory may fail you when you undergo a baptism of fire. You already know that list should not form an answer to a question. In a list A, B, C, SuperMemo needs to separately understand your difficulties with linking A and B, B and C, etc. You can use [[Glossary:Cloze_deletion|cloze deletion]] to learn lists using multiple [[Glossary:Item|items]]. For example:

'''Parent template [[Glossary:Topic|topic]]/[[Glossary:Extract|extract]]:'''
<blockquote>A B C</blockquote>

[[Glossary:Item|Items]] generated with [[Glossary:Cloze_deletion|cloze deletion]]:

'''Item 1:'''
<blockquote>
'''Question''': <span class="cloze">[...]</span> B C<br>
'''Answer''': A
</blockquote>

'''Item 2:'''
<blockquote>
'''Question''': A <span class="cloze">[...]</span> C<br>
'''Answer''': B
</blockquote>

'''Item 3:'''
<blockquote>
'''Question''': A B <span class="cloze">[...]</span><br>
'''Answer''': C
</blockquote>

It is important to know that '''Item 1''' above may make you fail to answer with A to the question C if you only learn to answer '''Item 1''' by understanding the association of B with A. In such cases, you will need even more work by formulating [[Glossary:Item|items]]: A-B (where A is the question and B is the answer), A-C, B-A, B-C, C-A, and C-B. Although you will get six [[Glossary:Item|items]] instead of one, you knowledge is likely to be more solid and you may actually spend less time on repetitions of those multiple [[Glossary:Item|items]] than on repetitions of the conglomerate A-B-C item

=== Lists can often be ignored ===

In a majority of cases, we do not need to learn lists at all. The perception that a list is worth memorizing is often a reflection of a bad habit we bring from school where lists are a frequent feature at exams.

A user of SuperMemo asked: &quot;''Could you please help me with extracting [[Glossary:Item|items]] from the following text? I am really not sure where to mark the boundaries of [[Glossary:Extract|extracts]] and how to use [[Glossary:Cloze_deletion|cloze deletion]]?''&quot;:

<blockquote>
<p>''Changing Rates of Mental Illness''</p>
<p>''Mental illness is becoming an increasing problem for two reasons. First, increases in life expectancy have brought increased numbers of certain chronic mental illnesses. For example, because more people are living into old age, more people are suffering from dementia. Second, a number of studies provide evidence that rates of depression are rising throughout the world''</p>
</blockquote>

This fragment is difficult to process because it is an enumeration (a list) that forms one large logical structure. However, for understanding the subject, you do not really need to remember how many factors affect mental illness. You primarily need to remember the relationship between the cause and the effect. If you ignore the enumeration, you can simply produce the following [[Glossary:Topic|topics]] that will each be easy to process further:

<blockquote>
* ''Mental illness is becoming an increasing problem''
* ''Increases in life expectancy have brought increased numbers of certain chronic mental illnesses''
* ''Because more people are living into old age, more people are suffering from dementia''
* ''Rates of depression are rising throughout the world''
</blockquote>

If you believe that you cannot live without the enumeration, you can first extract the facts listed above, and then simplify the enumeration by deleting all superfluous information:

<blockquote>
''Mental illness is increasing for 2 reasons:''
# ''increases in life expectancy have brought increased chronic illnesses''
# ''rates of depression are rising''
</blockquote>

=== Using Decompose ===

Little known but very useful function in SuperMemo can help you tackle lists.

If you are learning the names of fruits, you might see a text like this:

<blockquote>''Examples of fruits are apples, oranges, pears, cherries, banana...''</blockquote>

You do not want to memorize the whole list of fruits, esp. that the longer it gets the more likely you are to fail. There is no definite or complete list of fruits that would make a good memorization target. All you really want to know is that an apple is a fruit, or an orange is a fruit, etc. The list itself is rather useless. You can keep it as a reference in SuperMemo, but memorizing it would be a waste of time.

You can use '''[[Component menu#Decompose|Decompose]]''' to quickly achieve your learning goals in reference to fruits. Start from modifying the [[Glossary:Topic|topic]]:

<blockquote>''Apples, oranges, pears, cherries, banana ... are fruits.''</blockquote>

Generate a template [[Glossary:Cloze_deletion|cloze]]:
<blockquote>
'''Question''': Apples, oranges, pears, cherries, banana are <span class="cloze">[fruits/vegetables]</span><br>
'''Answer''': fruits
</blockquote>

Convert that [[Glossary:Cloze_deletion|cloze]] to a version that can easily be decomposed by placing list members into braces using slashes as separators:
<blockquote>
'''Question''': {apples/oranges/pears/cherries/banana} are <span class="cloze">[fruits/vegetables]</span><br>
'''Answer''': fruits
</blockquote>

You can now treat the question field with '''[[Component menu#Reading|Reading]] : [[Component menu#Decompose|Decompose]]''' on the [[component menu]] to get a series of [[Glossary:Cloze_deletion|clozes]] like these:
<blockquote>
'''Question''': apples are <span class="cloze">[fruits/vegetables]</span><br>
'''Answer''': fruits
</blockquote>

<blockquote>
'''Question''': oranges are <span class="cloze">[fruits/vegetables]</span><br>
'''Answer''': fruits
</blockquote>

<blockquote>
'''Question''': pears are <span class="cloze">[fruits/vegetables]</span><br>
'''Answer''': fruits
</blockquote>

<blockquote>etc.</blockquote>

<div class="center">
[[File:Decompose-ready cloze deletion.jpg|800px|SuperMemo: An exemplary cloze ready for Decompose]]
<blockquote>'''''Figure''': An exemplary [[Glossary:Cloze_deletion|cloze]] ready for '''[[Component menu#Reading|Reading]] : [[Component menu#Decompose|Decompose]]''' on the [[component menu]].</blockquote>
</div>

== Thorough preview of articles ==

It is hard to formulate a simple algorithm for deciding when to preview the whole article before reading, and/or when to extract the most important fragments during a preview. After all, the article can also be read incrementally in a linear sequence. This is a case a multi-criterial optimization where many factors must be taken into account and the ultimate decision will depend on your own preferences.

Here are some criteria:

* if you need this knowledge today, you should start with a quick preview and extracting mission-critical fragments
* if this knowledge is very important and your learning process overflows with articles to read, extract-preview will increase your exposure to the article
* if the article is not very interesting, line-at-a-time reading will be equivalent to assigning a lower [[Glossary:Priority|priority]] (you will just read a sentence once per week or once per month and you may never finish the article unless it gets more relevant or interesting)
* if you believe the article contains very important pieces in its body, you may want to quickly locate these and extract them for separate (more detailed) reading
* if your reviews occur in very long [[Glossary:Interval|intervals]] as a result of slow reading, you may opt for shortening the interval or running a preview of the most important sections instead
* if you are reading texts from your e-mail [[Glossary:Tasklist|tasklist]], preview is highly recommended: not all people start their messages with the most important points and you certainly would not want to delay locating paragraphs requiring immediate action with weeks of delay
* if you import from [http://en.wikipedia.org Wikipedia], high priority articles can be split automatically (with '''Split: Split the article''' in the '''[[SuperMemo Commander|Commander]]'''), and you can assign [[Glossary:Priority|priorities]] to individual subsections

In summary, these are the most important incentives for the whole-article preview:
* high priority of the material
* indication of higher-priority fragments buried in a lower-priority text
* high interest

== Unified store of knowledge ==

SuperMemo should conglomerate all your knowledge and your learning material. Ideally, you should just keep a single body of knowledge in a single [[Glossary:Collection|collection]]. That main collection should be enriched daily by new imports of articles and multimedia from various sources. Keeping your knowledge in a single [[Glossary:Collection|collection]] is vital for [[Subset learning|subset review]], searching for knowledge and references, searching for pictures, statistics, progress monitoring, import defaults, etc. SuperMemo provides a rich set of tools for handling disparate areas of knowledge in a single continuous learning process. However, in certain situations, you may create separate [[Glossary:Collection|collections]] for various purposes. There are two main reasons for keeping [[Glossary:Collection|collections]] separate:
# collections that are not part of your lifelong learning, and
# collections that need special treatment due to various technical limitations imposed by SuperMemo or current technology.

Here are some examples of situations that justify creating [[Glossary:Collection|collections]] separate from your main body of knowledge:

* project collections ([[Creativity and problem solving in SuperMemo#Incremental_problem_solving|problem solving]], [[Glossary:Tasklist|tasklists]], creative elaboration): those [[Glossary:Collection|collections]] serve a specific project, not your general learning. After the project is complete, they may either be archived or integrated with the main body of knowledge (depending on their nature).
* [[Creativity and problem solving in SuperMemo#Incremental_writing|incremental writing]] projects, blogs, etc.
* [[Creativity and problem solving in SuperMemo#Incremental_problem_solving|incremental problem solving]] [[Glossary:Collection|collection]] dedicated to a specific problem where you want to focus entirely on a single issue
* picture [[Glossary:Collection|collections]] (public: family albums; private: family picture collections, etc.): if your picture database goes into hundreds of thousands of pictures, you will definitely want to keep it separate for easy backup and dedicated processing, which will often be of different nature than learning
* YouTube video collections: learning with [http://youtube.com YouTube] is quite different from [[Glossary:Incremental_reading|incremental reading]], you cannot play your favorite music while learning with videos, you may give up learning when you have no access to the Internet, etc.
* video file collections (e.g. home video): those [[Glossary:Collection|collections]] will usually call for different processing. In addition, they quickly grow into terabytes in size, which discourages frequent backups
* diaries and journals
* logs
* jukebox: incremental learning process can add an extra level of fun to experiencing music. Only after playing with a jukebox.kno for a few weeks you will truly understand how important [[Glossary:Priority|priorities]] and [[Glossary:Interval|intervals]] are in the perception of music!
* mail: you may want to separate business from family&amp;friends and from the mail related to the current project
* [[Glossary:Tasklist|tasklists]] and charts
* paper notes
* kid's [[Glossary:Collection|collection]]: if your kids are too young to do SuperMemo on their own
* notes: on occasion you may prefer to keep your notes in SuperMemo rather than a word processor (e.g. easier startup, collection switching, better search, etc.)
* audio archive, and more

== Flow of knowledge ==

The figure below roughly illustrates the flow of knowledge in time depending on knowledge difficulty:

<div class="center">
[[File:Repetitions_graph.png|800px|SuperMemo: Time flow of knowledge in the learning process]]
{{:Image:Repetitions_graph.png}}
</div>

== Incremental reading of PDF articles ==

PDF is a proprietary format. SuperMemo does not support PDF natively. This has always made PDF materials harder to process than ordinary HTML text imported from the web.

There are 4 approaches that are most often used to process PDF incrementally. You will need to see which one is best for your particular material. It may happen that you will need to resort to mixed strategies and use different approaches to different texts. The 4 options are:

* '''[[#PDF to HTML converters|Converters]]''': using PDF to HTML converters to generate HTML text that can be read in SuperMemo
* '''[[#PDF with visual learning|Pictures]]''': using page snapshots (e.g. with '''Print Screen''') and employing [[visual learning]]
* '''[[#PDF copy and paste|Manual]]''': using copy and paste to copy PDF to SuperMemo page by page (or picture by picture)
* '''[[#Incremental PDF|Incremental]]''': using an incremental approach to read-copy-and-paste while working with PDF in Adobe Reader called from SuperMemo

Conversion to HTML is most convenient and least expensive. However, some converters and/or some PDFs produce HTML that is quite different than the original, and/or difficult to process in SuperMemo (e.g. requiring extra filtering, or extra manual formatting). Page snapshots are a fast way to read and import pages that are difficult to convert or are read only (e.g. manuals that require a specific page layout). Copy and paste approach is best for articles that can easily be selected in their entirety and which do not contain too many pictures. Finally, the incremental approach is most natural for SuperMemo, however, instead of using [[Glossary:Read-point|read-points]], the student needs to make a note where he or she stopped reading the text.

For more solutions, see: [https://supermemopedia.com/wiki/PDF PDF at SuperMemopedia].

=== PDF to HTML converters ===

As SuperMemo cannot host PDF files natively for [[Glossary:Incremental_reading|incremental reading]], the most promising long-term solution is the use of PDF to HTML converters. However, all converters have their limitations. The languages defined by HTML and PDF are not equivalent. Not all expressions of PDF can be expressed in HTML. This is why you need to check a few converters and see which one is best at processing your type of learning material.

==== Conversion ====

Here is a short list of converters as compiled at [https://supermemopedia.com SuperMemoPedia]:

* [https://supermemopedia.com/wiki/PDF_to_HTML_converters PDF to HTML converters]

If you are not sure which one to choose, you might use this one: [http://www.pdfonline.com/ PDF online] (it keeps a limit of 2 MB on the size of files to convert).

==== Processing ====

Once texts are imported to SuperMemo, they may need further processing:

* filter problematic code with ''F6'' (e.g. bulletizing tables)
* import pictures (some pictures are text-like constructs that need to be snapped with copy&amp;paste)
* split into manageable portions (e.g. with the help of '''Split: Split the article''' in the [[SuperMemo Commander|'''Commander''']])
* convert portions of text to plain text with ''Ctrl+Shift+F12'' (in cases of most unfriendly formatting)

==== Using MS Word ====

Some students use MS Word as an intermediary stage in conversion to HTML. However, as much as the conversion of PDF to HTML is never complete and accurate, there are differences between PDF and MS Word that make life
difficult. At the moment of writing, there aren't any tools that would reliably convert PDF to MS Word format. Pasting from Adobe Reader to MS Word yields poor results (e.g. texts in columns lose or mix portions of texts). If you own MS Word, you can give it a try, however, using converters or [[#Incremental reading of PDF articles|other methods]] will probably work better.

==== Using OCR ====

Some users swear by [https://en.wikipedia.org/wiki/Optical_character_recognition OCR]. See: [https://supermemopedia.com/wiki/PDF_and_OCR Using OCR to convert PDF for SuperMemo].

=== PDF with visual learning ===

You can convert PDF to pictures and employ the tools of [[visual learning]].

Here are the steps:

# convert PDF to pictures either by saving as [http://en.wikipedia.org/wiki/JPEG JPEG], or by using page snapshots with '''Print Screen'''
# [[Visual learning#Pasting pictures from the clipboard|paste pictures]] to SuperMemo or [[Visual learning#Importing pictures from a local drive|import pictures from a local drive]]
# as pictures are not searchable, it is important to copy&amp;paste a part of article (if possible) that could later be used in [[Subset learning#Search and review|search and review]]. For example, abstracts of scholarly articles with necessary reference tags (<span class="reference">#Title</span>, <span class="reference">#Author</span>, <span class="reference">#Source</span>, etc.)
# use [[visual learning]] to process the imported materials. Pictures will need to be trimmed (e.g. (1) ''Alt+click'', (2) marquee selection of the interesting part, (3) press ''Esc'', and (4) choose ''Permanently cut/crop the zoomed/trimmed image file''). Use visual [[Glossary:Extract|extracts]] (''Alt+X'') to focus on smaller portions of imported articles
# annotate the pictures or write down summaries to further process most important portions with [[Glossary:Incremental_reading|incremental reading]]. Without the use of [[Glossary:Cloze_deletion|cloze deletion]], imported material may quickly be forgotten.

This method of reading PDF is analogous to [[#Incremental reading of paper articles|incremental reading of paper materials]].

See more at [https://supermemopedia.com SuperMemoPedia]: [https://supermemopedia.com/wiki/PDF_and_Visual_Learning PDF and Visual Learning].

=== PDF copy and paste ===

If [[#PDF to HTML converters|conversion of PDF to HTML]] is not satisfactory, you may need to resort to old plain copy&amp;paste. Some PDF texts paste whole chapters pretty well with little extra text/code (e.g. page headings). Some PDFs respond well to ''Ctrl+A'' (for selecting all text) and '''Copy with Formatting''' (right click). Usually, Adobe Reader will not let you do multi-page selections that would paste to SuperMemo fast and nicely. You may even get annoyed with your attempts to copy a multi-column portion of a single page. You will get one column selected, middle of the second, and the ending of the third. Other double-column texts will paste as a mix of both columns per line. Sometimes pasting multi-column texts is not possible. This means that you will often need to go from page to page, or even column to column, and repeat copy&amp;paste. That can take ages. Similarly, pictures do not copy as part of HTML. So you need to copy&amp;paste pictures separately. Here again, when Adobe Reader will not let you copy pictures easily, you can just use '''Print Screen''', paste to SuperMemo and trim&amp;crop in SuperMemo (see: [[#PDF with visual learning|Visual learning]]). Tables are usually best processed in the form of pictures as they usually import poorly to HTML. Some PDF documents do not even allow of selecting texts. In those cases you will need to resort to [[#PDF with visual learning|PDF visual learning]] too. Sometimes you may just give up and read the article straight without bothering to do it incrementally, or look for alternative texts, or just give up reading altogether. You can also read the article using traditional methods, and copy to SuperMemo only the portions that you would normally extract within SuperMemo.

After pasting text from Adobe Reader to SuperMemo, it will often be jagged, disrupted, wrongly formatted, etc. Converting short passages to plain text often dramatically improves readability. The shortcut is ''Ctrl+Shift+F12''.

For more, see: [https://supermemopedia.com/wiki/PDF_Copy_and_Paste PDF Copy and Paste]

=== Incremental PDF ===

If you want to work incrementally with PDF articles, you can use the following strategy:

# Save your PDF files to a dedicated folder
# Import them regularly with '''[[File menu|File]] : [[File menu#Import|Import]] : Files and folders'''
# Define a PDF [[Glossary:Template|template]] with a large [[HTML component]] and a tiny binary [[Glossary:Component|component]] for holding PDF (use that template as the default for the import [[Glossary:Concept_group|concept group]], or apply it after import to all PDF articles)
# When reading is scheduled for the PDF element, click the binary [[Glossary:Component|component]] and jump to the page where you last finished reading (to automatically go back to the last viewed pages, in Adobe Reader, check '''Edit : Preferences : Documents : Restore last view settings when reopening documents''')
# Paste all important fragments of the PDF document to the [[HTML component]] (if texts cannot be selected, use the '''Print Screen''' option, paste the picture, and use [[visual learning]] tools to extract individual fragments)
# Paste figures to SuperMemo by clicking a selected picture, right-clicking, and choosing '''Copy Image''' in Adobe Reader or '''Print Screen'''. Use [[visual learning]] tools to process larger pictures (trimming, zooming, extracting, etc.)
# Process the pasted texts in parallel with reading the PDF document (esp. if they are very important). Alternatively, start [[Glossary:Incremental_reading|incremental reading]] in HTML only when you finish reading the document in PDF

== Incremental reading of paper articles ==

If you have many notes taken from paper journals, or you must read paper articles, you can use a few methods to employ [[Glossary:Incremental_reading|incremental reading]] in that process. Working with paper will never be as effective as working with electronic material. However, you can still triple your performance with the benefits of incremental learning.

Here is an exemplary algorithm for processing paper notes:

# before getting down to processing, search the net! Many noteworthy articles have already been published online. That saves lots of time
# if the text is of general nature, you might find a better equivalent (e.g. at [http://en.wikipedia.org Wikipedia])
# create a dedicated ''PaperNotes.kno'' [[Glossary:Collection|collection]] for easy backup. It might grow to a huge size! You can use a dedicated processing time slot for your paper notes
# very short notes or very important notes, you can type by hand. The advantage of typing is that it can shorten the texts, give them more meaning, and have them instantly searchable in the [[Glossary:Collection|collection]]
# the rest of texts you can import as pictures or process with [https://en.wikipedia.org/wiki/Optical_character_recognition OCR] (e.g. use a pen scanner to pick the most valuable pieces from the book)
# use a digital camera to quickly snap paper pages. A scanner yields better quality but is too slow
# copy the pics to the hard disk
# use '''[[File menu|File]] : [[File menu#Import|Import]] : Files and folders''' to import all pictures at once
# use [[visual learning]] and the tools of incremental learning to prioritize and process the imported material (the '''Extract''' option is particularly useful)

A user of SuperMemo wrote a few words about his experience with [https://en.wikipedia.org/wiki/Optical_character_recognition OCR] in [http://www.super-memory.com/articles/users/zmuda.htm Studying Law with SuperMemo].

See also: 
* [https://supermemopedia.com/wiki/I_have_lots_of_notes_from_paper_journals Paper notes in SuperMemo]
* [http://supermemoadventures.blogspot.com/2010/04/how-to-incrementally-read-anything.html How to incrementally read anything]

= Philosophy of incremental learning =

== Incremental learning is not for everyone ==

Incremental learning is powerful. Even though we strongly believe it might spell a revolution in learning, its present incarnation will reach only a small fraction of society. This is why:

* Most of people will never arrive to this website. A great part of this planet do not have access to the Internet. Of those who are privileged enough to be able to use [http://google.com Google] or [http://en.wikipedia.org Wikipedia] on a regular basis, most are too busy with work, family and other pursuits. The rest are busy with [http://facebook.com Facebook], [http://twitter.com Twitter], games and other &quot;blessings&quot; of the Internet.
* A significant fraction of the users of the Internet realize that the web offers incredible self-help opportunities. However, a substantial portion of those will suffer from a cognitive bias known as the '''illusion of knowledge'''. This bias does not spare highly intelligent individuals. Under the illusion of knowledge, we are less motivated to seek more learning. With less learning, the learn drive diminishes with age providing a negative feedback of inaction
* Users of the Internet who are highly interested in learning quickly discover that their memory can be improved, and that their learning can be advanced to a new level with some help from the net or software. However, the language barrier is another big sieve. The best materials about learning techniques have been published in English. Language divisions result in a global inequality in the access to information. You can read about SuperMemo in a couple of major languages. Perhaps you might learn about [[Glossary:Incremental_reading|incremental reading]] too. The article you are reading now will take months or years to percolate to non-English areas of the net!
* For anyone interested in learning, it does not take long to find out about ''[[Glossary:Spaced_repetition|spaced repetition]]''. The term was quite exotic when first used in the context of SuperMemo in the mid 1990s. In those days it could only be found in a couple of scientific publications. However, it was SuperMemo that helped popularize the concept, which now has entered the popular culture. Those who seek better memory will quickly be informed that [[Glossary:Spaced_repetition|spaced repetition]] is the way. They will get their introduction to [[Glossary:Spaced_repetition|spaced repetition]] from [http://super-memory.com/english/down.htm SuperMemo freeware], or an array of other excellent free spaced repetition applications, some of which use older SuperMemo algorithms. The fact that the newest SuperMemo is a commercial product will prevent many potentially excellent students from ever going in the direction of incremental learning. This must be made clear then. At the moment of writing, [http://www.freewarefiles.com/SuperMemo_program_13849.html SuperMemo 2004 is freeware] and supports [[Glossary:Incremental_reading|incremental reading]]. It might be missing the [[Glossary:Priority_queue|priority queue]], but this will matter only much later in the process when you build a [[Glossary:Collection|collection]] of unmanageable size. In [http://www.freewarefiles.com/SuperMemo_program_13849.html SuperMemo 2004] you can do '''Import''', '''Extract''' and '''Cloze'''. That's the core of [[Glossary:Incremental_reading|incremental reading]]!
* Among users of SuperMemo, only a fraction will ever become interested in a complex concept of incremental learning (probably still below 5%). A typical newcomer is more likely to drop out altogether than to move on to the next level. The chief culprit is the incompatibility of SuperMemo with the human nature and the modern lifestyle. Stressful and busy lives do not help in forging new paths and acquiring new habits. It is not just that humans like to live in their comfort zones. It is also that the pressures of the modern dog-eat-dog society are pretty effective in stifling higher inspiration, lifelong learning, creative pursuits beyond one's job, etc. Those issues are inherent (human nature), or systemic. As such they cannot be easily remedied.
* Some users might hold a relatively positive opinion about [[Glossary:Incremental_reading|incremental reading]], but will excuse their lack of commitment by technological limitations of SuperMemo. For example, &quot;''unless you give me a Mac version, I won't go there''&quot;, &quot;''I hate Internet Explorer''&quot;, &quot;''Your interface comes from Windows 98 era''&quot; or &quot;''I will come back if you have a mobile version''&quot;. Incidentally, there is an inherent incompatibility between mobility and incremental learning. Mobility correlates with multitasking and low attention (interaction with others, social media notifications, arriving mail, phone calls, browser pop-ups, etc.). The power of incremental learning lies in balancing creativity with attention. As such, incremental learning is inherently immobile (unless you understand mobility as a transfer between two or more peaceful and static environments). In addition, without a standard-size keyboard, [[Glossary:Incremental_reading|incremental reading]] might feel like paddling up the creek.
* Among those users of SuperMemo who get down to reading about incremental learning, a large proportion will come with a preconceived goal: proving that incremental learning is not really worth the investment of time. &quot;Negative reading&quot; is more likely to occur in a bad state of mind: low-spirits, irritation, or reading outside one's [http://super-memory.com/articles/sleep.htm#Optimizing_the_timing_of_brainwork optimum circadian brainwork bracket]. This preconception may be largely subconscious and based on a defense mechanism: proving that incremental learning is a weak concept is easy, should save one a great deal of investment in mastering the concept, and prevent the need to leave one's comfort zone. Discovering that incremental learning is a must will simply augur long hours of plodding through the documentation that is notorious for its poor readability index. The presented article is harder to read than 70% of articles at [http://en.wikipedia.org Wikipedia]. The ultimate judgement is strongly correlated with your present state of mind. You are more likely to say &quot;''Incremental learning is full of fluff''&quot; or &quot;''Incremental learning goes against human nature''&quot; than to say &quot;''I have discovered a new great technology that will change my life''&quot;. You might be just seeking affirmation of your preset frame of mind: &quot;''Incremental learning must be a hoax or at least a waste of time''&quot;, or simply &quot;''Incremental learning is not of much use in my particular profession''&quot;. It is easy to find the desired confirmation. Imperfections in this article will provide a rich material for dismissal. A perfection is not achievable. Prejudiced mind will always find a way to misinterpret. Naturally, if you are enthusiastic about incremental learning at your first encounter, this is a very good predictor of your ultimate success.
* Having passed all this rich obstacle course, you are now bound to be in a tiny group of people who arrived at this point. The good news is that you might be part of a special elite. People with fertile minds who are open to new challenges, motivated well enough and self-disciplined enough. Naturally, you might have equally well come here exhausted with your failure in learning, in hope of finding a remedy to your recent bad exam results. Whatever the answer, the road is still very long. Even with a brisk mind and great enthusiasm, you may still drop out at later stages.
* When reading about incremental learning, you might hit another major obstacle: &quot;''This is more convoluted than I thought!''&quot;, or &quot;''The whole idea of [[Priority queue|prioritization]] makes me nauseous''&quot;, or &quot;''The toolset may be rich but is it natural?''&quot;. You can get discouraged either by incomplete understanding of the concept, or the incompatibility of the concept with your personality.
* If you have already decided to try incremental learning, and you have [http://super-memo.com/supermemo17.html SuperMemo 17] installed, you may get an unpleasant shock. [http://super-memo.com/supermemo17.html SuperMemo 17] is not user friendly! It has been optimized for pro users and only simplified enough at lower levels ('''[[File menu|File]] : [[Levels|Level]]''') to make it somewhat palatable for a novice. You can overcome this impression only with a firm belief: incremental learning works and it is worth the investment of my time. After a year, you may love the solutions you hated in SuperMemo at first. User unfriendliness and complexity will make many give up in the first days or weeks. Those dropouts rarely return! We are guilty here too. New solutions are never perfectly encapsulated and/or interfaced. You will get your neat iPad and Android version only when incremental learning is more popular, which is not anytime soon.
* The dropout rate in the first weeks is still high. Some of the main problems: importing unsuitable text, difficulty in parsing the text with [[Glossary:Extract|extracts]], losing context of [[Glossary:Extract|extracts]], comprehension problems (not related to SuperMemo), [[Glossary:Cloze_deletion|clozing]] texts that are too complex, [http://www.super-memory.com/articles/20rules.htm difficulty in formulating items], excessive time spent on formatting and [[Glossary:Template|templates]], [[Glossary:Overload|learning overload]], confusion, problems with SuperMemo (e.g. layouts), lack of sense of progress, etc. Without clear gratification, many users come to the conclusion that the cost is greater than the benefits. They are not wrong. It is not easy to get a return on your investment in a month or even longer. Few users start their learning swiftly on a wave of solid enthusiasm!
* Many of the advantages of incremental learning rely on the size of your learning [[Glossary:Collection|collection]]. The value of [[Subset learning#Search and review|search&amp;review]] increases with the size of the [[Glossary:Collection|collection]]. At the beginning of the process, when [[Glossary:Collection|collection]] is small, and as a result knowledge-poor, you may feel like laying the first brick of a large pyramid on a vast boring desert. The greatest fun can be found at the top of the pyramid, when you can see the extent of your knowledge in a good perspective. This metaphor should not imply your pyramid will ever stop growing. You keep adding stone blocks to the sides. The pyramid keeps growing high and wide for lifetime.
* The extent of failure does not only cover the dropouts and the never-has-beens. Even long-term users may enter a never-ending struggle with [http://www.super-memory.com/articles/20rules.htm knowledge formulation], complex material, [[Glossary:Priority|priorities]], [[Glossary:Overload|overload]], etc. The later you start, the longer the transition from pain to fun. Old habits die hard. There is one clear litmus test for success: once incremental learning is fun, you know you do things right. If you have not reached the level of fun after a couple of months of trying hard, you need to re-think your strategies. For example, you may need to start from reading this article cover-to-cover again.
* If you happen to be a happy user of incremental learning, your knowledge will keep growing fast, and your hunger for learning will keep increasing at an ever faster rate. This makes you a part of a very tiny elite. This also imposes an obligation. It is time to help others reach your level!

== Big picture in incremental learning ==

A frequent reservation voiced by skeptical observers is that dismembering texts into little units will result in an inevitable loss of the ability to see things from a distance in their entirety. The question is: ''Does incremental learning produce a loss of the big picture?'' What the skeptics fail to appreciate is the power of [[Glossary:Spaced_repetition|spaced repetition]] that stands behind SuperMemo. The SuperMemo method ensures high [[Glossary:Retention|retention]] of once-mastered knowledge. This means that there is minimum disintegration of the coherence of knowledge once it is understood and well modeled in student's mind.

'''The main advantage of SuperMemo is that you convert lots of disparate pieces of information into a solid model of reality that lives in your memory'''. All these pieces can be dispersed randomly in your [[Glossary:Collection|collection]] like pieces of a jigsaw puzzle, however, they fit into a coherent entirety that stays firmly intact in your mind. In other words, incremental learning is reductionist at the level of knowledge processing, but is holistic at the level of memories stored in your brain.

The big picture worry comes from the fact that in traditional education, students rely heavily on their short-term memory. They cram a single subject intensely before an exam only to forget most of that knowledge in the following months unless the subject is reviewed in later courses. It is true that short-term memory can act as a glue that holds the big picture in memory. However, it is the short-term memory and the weakest long-term memories that are gone first. With that in mind, it is only natural that you might worry that if you start intermingling more than a few courses, you will end up with chaos and confusion. It is not the case with incremental learning. New memories, once established, are gradually reinforced to permanently enter the long-term storage. All incremental learning does is to help you focus on a small portion of the material at any given moment.

<div class="bs-callout bs-callout-warning">
'''In traditional (unspaced) education, the big picture is maintained with the glue of short-term memories. As such it is volatile, and subject to forgetting. In incremental learning, the big picture has a lasting value!'''
</div>

Worrying about the big picture in [[Glossary:Incremental_reading|incremental reading]] is as if you worried that you might forget the structure of your family just because you meet too many members at the same time at a big family occasion, or that you might forget your name when focusing too much on spelling its individual letters. Once the big picture resides in your long-term memory, you can play with details to your heart's content. Today, tomorrow, or in a couple of years.

== Incremental learning can make you smarter ==

{{:Incremental learning can make you smarter}}

== Knowledge that makes you smarter ==

=== Smart and dumb learning ===

Not all knowledge is of great value. Memorizing all tributaries of [http://en.wikipedia.org/wiki/Amazon_River Amazon] may take precious time from other areas of learning. Unless you are an expert on [http://en.wikipedia.org/wiki/Amazon_River Amazon], [http://en.wikipedia.org/wiki/Rain_forest rain forest], geography, Indian tribes or related areas, you would rather not want to begin your incremental learning experience from that Amazon exercise.

Your learning efforts must be based on high applicability of newly acquired skills and knowledge. If you memorize the whole phone book, your problem solving ability will increase only slightly (mostly through the beneficial effect of memory training on the health of your brain). On the other hand, a simple formula for [http://en.wikipedia.org/wiki/Expected_payoff expected payoff] may affect all decisions you make in problem solving and in life in general. It can, for example, save you years of wasted investment in lottery tickets. Millions of people are enticed with huge lottery jackpots, yet they would never agree to give up their whole income for life in order to get it back at retirement in one-off payment, which is a frequent probabilistic payoff equivalent of taking part in lotteries. Using the terminology defined above, you will find most benefit in mastering and understanding highly abstract rules of logical thinking and decision making.

To accomplish smart learning, you will need to constantly pay utmost attention to what material you decide to study. You must avoid short-term gratification at the cost of long-term learning. It may be great fun to learn all Roman emperors and details of their interesting lives and rule. However, unless you study with a big picture in mind (e.g. in an attempt to understand why civilizations thrive or fall), you may benefit less than by struggling through less entertaining but highly applicable formulas of [http://en.wikipedia.org/wiki/Operation_research operation research], which might, for example help you optimize your diet, investment, daily schedule, etc. In other words, '''you cannot be guided by just the fun of learning but by your goals and needs'''. In time, you will learn to see the link between long-term learning and long-term benefits. You will simply condition yourself to love beneficial learning. Hard study material can still provide instant gratification.

While you focus on your goals, you cannot forget about the overall context of human life. You cannot dig solely into studying car engines only because this happens to be your profession. This would put you at risk of developing a tunnel vision. You might spend years improving a liquid fuel engine efficiency while others would leap years by getting involved in lithium batteries or hydrogen engines. One of the main reasons for which companies go bankrupt is that their leadership fails to spot the change. As corporate darwinism eliminates short-sighted teams, future society will witness more and more intellectual darwinism. To understand the trends and the future, you need to study human nature, economics, sociology, history, neurophysiology, mathematics, computing sciences, and more. The more you lick, the stronger your [http://super-memory.com/articles/genius.htm#Predicting%20the%20future predictive powers] and your problem solving capacity and the creative strength.

Should we learn &quot;trivia&quot; such as &quot;''Which year was the Internet born?''&quot;. '''The concept of trivia is highly relative!''' To a child in a kindergarten, the birth of the Internet is rather meaningless. At this stage of development, the child may find it difficult to grasp the concept of the Internet itself. Many parents will wait until the primary school before showing a web browser to their child. The value of putting the date on the birth of the Internet probably develops only in the context of an effort to understand the history of technological development. In this context, 1969 may be as important as the name of [http://www.super-memory.com/articles/genius.htm#Johannes%20Gutenberg Gutenberg]. Only when multiple events of the 1960s and the 1970s dovetail together, the commissioning of ARPANET becomes meaningful. When we figure out that we landed the man on the moon before making the first connection via the net, 1969 looms larger. If we dig deeper, we may find it inspiring to know that when Charley Kline tried to log in on October 29, 1969, the network crashed as he typed the letter ''G''. Imagine, you work on commissioning a major installation that you have worked on for several years. You know that the installation implements revolutionary concepts yet it keeps on crashing. You are about to lose heart. This may not necessarily be an emotional event, after all you also need to apply probability to deciding when to give up blind-alley pursuits even after years of investment. The juxtaposition of the small letter ''G'' and the groundbreaking concept of the interconnected world will help you see the big picture. If your concept is great enough, you will go on through another 100 crashes in hope of diagnosing the reason. That perseverance may seem to have little to do with &quot;being smart&quot;, but it will make you into a winner (given a bit of luck).

Listen to other people's advice and valuations. The younger you are the more you should listen. In the end though, '''it must be you who determines the criteria for sifting golden knowledge from trivia'''. Only you can measure the value of knowledge in the light of your own goals.

Remember that not all knowledge can easily be formulated in a declarative manner. Remember then to use the power of your own neural networks: solve problems, practice your skills, compute, abstract, associate, etc. You and others may not be able to see or verbalize some rules but your brain will extract them in the course of practice. Once the rules have been developed, try to formulate them and write them down. This can be of benefit to you and others.

=== Knowledge that assists problem solving ===

Problem solving in artificial intelligence can be used to mimic those mental faculties of a human being that we associate with intelligence and creativity. There are many approaches to problem solving such as deduction, induction, abduction, reasoning by analogy or probabilistic methods, neural networks approach, searching in state space, and more. And even though some of these can generate false conclusions or uncertain responses, they can all be expressed by means of deductive methods such as those used in theorem proving in mathematics. Deduction proceeds from axioms or premises towards true formulas or assertions through logical derivation using valid inference rules. The better the selection of assertions and the selection of inference rules, the broader the reasoning capacity of an expert system or human brain. Rich inferential knowledge stored in our memory makes us fast in thinking, creative, intelligent, efficient in problem solving, etc. Yet we need factual knowledge as the raw processing material for derivation. Even the creative vein and inspiration can be expressed in terms of multithreaded derivation and backtracking well formalized in logic programming. No magic fluids or reflections of the soul are involved here. Just a plain network of firing neurons recovering the encoded patterns of facts and rules. See also: [http://super-memory.com/articles/genius.htm Roots of creativity and genius].

To boost your problem solving capacity you need to master a great deal of inferential knowledge. If you learn about specific classes of problems to solve, you may need to choose between memorizing the outcome of inference or its individual steps. Both equip you with a formula for transforming knowledge into new quality. Learning derivation steps is costly. At the same time it can equip you with highly abstract derivation rules. You need to balance cost and benefit. We cannot blindly assume that it is better to derive answers to problems than to simply memorize the answers. In problems with multiple instances, we will tend to learn the derivation. For example, it is easier to learn how to derive the third power of X than to memorize all useful pairs: argument-result. On the other hand, most of people memorize the multiplication table at the time when they could still derive the result from bit-operations on binary numbers. Memorizing some 40 or so combinations seems more practicable, esp. that derivation always takes time and we make use of the multiplication table thousands of times in a lifetime. Here memorization saves hours and days in the lifetime perspective. Brain armed with rich factual and inferential knowledge will associate the most remote ideas and derive the most unexpected conclusions. Through a conscious control over this rich reasoning process, we have built the present civilization.

=== Knowledge that assists creativity ===

Creativity is based on association of ideas. It is helpful to be born with a creative brain that keeps spewing new ideas in all circumstances. However, creativity can also be assisted. When you master a great deal of facts and rules in areas adjacent to the problem that you try to creatively solve or elaborate upon, your creativity gets enhanced. If you employ incremental learning, the association of random ideas will be a norm throughout the process.

All innovation in the history of science and technology is based on association. Forgetful memory can be helped with external sources of knowledge and each major scientific breakthrough is based on a series of smaller steps, many of which will be based on hours of search, trial-and-error and experimentation. The brain armed with extensive knowledge is more likely to come up with the needed associations. As in problem solving, inferential knowledge is also particularly valuable in creative efforts. The rules that you know, help you derive new truths, and associate these with all your current knowledge.

Two ideas do not come together to produce a great invention unless they sit in the same head. They either have to be called up at the same time or one has to come unexpectedly from the outside:

* [http://en.wikipedia.org/wiki/Archimedes Archimedes] stepped into a bath and immediately associated the fact with all he knew about weight and buoyancy. Thousands of Greeks at his time failed to make the link. This was not solely because of their less lively minds. It was mostly because of their lack of understanding of [http://en.wikipedia.org/wiki/Hydrostatics hydrostatics]
* [http://en.wikipedia.org/wiki/Isaac_Newton Isaac Newton] is said to have been hit by an apple that produced an immediate use of his knowledge of physics and mathematics to enhance his three famous laws of motion published in ''[http://en.wikipedia.org/wiki/Philosophi%C3%A6_Naturalis_Principia_Mathematica Principia]'' by the laws of gravity. Even though the story cannot be verified, it makes for a good example of creative association 
* [http://en.wikipedia.org/wiki/James_Watt James Watt] is rumored to have watched a kettle boil to combine this inspiration with his knowledge of engineering to improve upon the [http://en.wikipedia.org/wiki/Newcomen_steam_engine Newcomen steam engine]. This association changed the industry upside down in the course of the hundred years that followed

SuperMemo can help you be creative not only by combining various pieces of knowledge in your mind. It can help you generate new ideas while making repetitions! Creative associations do not come from the mere existence of two pieces of knowledge in your brain. Those two pieces of knowledge have to light up at the same time. Only this way can your brain make a connection. Strangely, a repetition related to genetically modified tomato can light up memories related to tomato juice, cucumbers, genetic disorders, take-home dinner, flu or even the silent [http://en.wikipedia.org/wiki/Mars_Polar_Lander Mars Polar Lander]. If your [[Glossary:Collection|collection]] combines knowledge pertaining to different subject domains, the stream of new ideas and unexpected associations coming to your mind may surprise you!

== Incremental learning vs. human progress ==

Incremental learning has a potential to boost your knowledge and boost mankind's potential to solve major problems of the day. However, in the long perspective, the role of human learning will be lessened by the development of artificial intelligence.

=== Knowledge and history ===

Throughout the ages, knowledge was the cornerstone of human progress. From [http://en.wikipedia.org/wiki/Stone_Age Stone Age] to [http://en.wikipedia.org/wiki/Information_Age Information Age], in pain, we have built a tiny oasis of civilization in ruthless expanses of the evolving universe. The history of the mankind is made of billions of individual lives that keep on sparking and fading. Born of self-preservation instincts imprinted by evolution, history books paint a picture of a constant string of wars, conflict of interest, loss and gain of influences, lust for power and submission to weaknesses of human nature. On the other hand, the greatest achievement of the evolution, the rational mind, kept on contributing to new findings, discoveries, technologies and philosophies. Progress has always hinged on discovering new truths and preserving them for posterity in form of stories, solutions, tools, books, and other carriers of information. Knowledge is the basis of human power, yet it constantly struggles with two forces that regularly diminish it: death and forgetting. We can preserve knowledge in books and other forms of information storage. However, this knowledge translates to value only then when it is used by the creative power of the human brain. The limitations of the human brain will remain a bottleneck of progress for many years to come. We will develop artificially intelligent knowledge processors not earlier than in a decade or two.

=== Knowledge and death ===

Death poses an ageless challenge to educating new generations. Years of hard work needed to gain knowledge on professorial level are obliterated in a single act of death. Newborns need to go through years of education before they are able to access, read, and comprehend this text. They all have to struggle with basic literacy skills, lessons of safe sex and teen pregnancy, lessons on superiority of altruism over egoism, the difference between wise and not-so-wise choices, existential questions, etc. Although constant reeducation may contribute to gaining a fresh perspective in each generation, it is also painfully wasteful. As yet, there is no efficient remedy to the death of knowledge. All we can do is to attach more weight to healthy lifestyle and health research. Those two promote longevity of knowledge in a single generation.

=== Knowledge and forgetting ===

Forgetting is a natural process that makes it possible to efficiently use the limited memory space of the brain. We forget to dispose of knowledge deemed less important in order to make space for knowledge of higher importance. Currently we have only a limited control over what we remember and what we forget. Today, the most important tool that we can use to prevent forgetting is practice. We can minimize time needed for practice by using [[Glossary:Spaced_repetition|spaced repetition]] (i.e. learning technique based on computing optimum [[Glossary:Interval|intervals]] between repetitions). '''Spaced repetition is the key to maximizing knowledge within a single human lifetime'''.

=== Immortal knowledge ===

Artificial intelligence is our best hope for approaching immortal knowledge. It can nearly eliminate the problem of death (except for the heat death of the universe or factors that we cannot foresee today, e.g. the collapse of space, and the like). Artificial intelligence can also eliminate the problem of forgetting (at least within the bounds of the available storage). Some forgetting is needed to shape crisp associative knowledge, but that is a welcome phenomenon that leads to more applicable knowledge. Despite great hopes we might have about artificial intelligence, the best path towards immortal knowledge must still rely today on the use of the human brain. This is why we believe that incremental learning is so vital for further human progress.

== Incremental reading is an extension of traditional book reading ==

Someone remarked: &quot;''If Gutenberg was a blessing then [[Glossary:Incremental_reading|incremental reading]] might be a curse!''&quot;. Is incremental reading an attack on traditional books? If you read in pieces and with endless interruption, does it not destroy the storyline?

Whether [[Glossary:Incremental_reading|incremental reading]] is a curse or a blessing depends on the way it is employed. There is no sharp transition between traditional reading and incremental reading. In the simplest case, you can use [[Glossary:Incremental_reading|incremental reading]] exactly in the same way as you would read a book. Partitioning of texts and interruptions are not compulsory. You can read the entire text from top to the bottom without a single interruption. This would make sense if you needed a storyline for context, and did not want to bother with committing it to long-term memory. If you do take breaks or skip portions of texts or change the natural sequence of reading, it all happens in situations that have their counterparts in the world of books:

* '''Interruptions''': you can read an entire book in a single evening, but very often you take breaks and read a book in portions on a daily basis. You may even put it back on a shelf in a busy period and return after a few weeks
* '''Multiple subjects''': it is not unusual to read more than one book in parallel. You may pick one that matches your mood for a given evening, or read portions of different books on the same day
* '''Changing the sequence of reading''': many readers are tempted to peek at the end of the book for its ending. Or browse back to earlier chapters to check some details that failed to register in prior reading
* '''Deleting portions of the book''': deleting portions of the text in [[Glossary:Incremental_reading|incremental reading]] is analogous to dropping them from your memory through forgetting once you read a book. The book is still on the shelf, the original electronic article may still exist in your archive, but your [[Glossary:Collection|collection]] or your memory retain only highlights that may get sparser as weeks go by. Years after the original reading you may retain only single quotes or golden thoughts. The rest is gone

In other words, in extreme cases, there may be no difference between traditional and [[Glossary:Incremental_reading|incremental reading]]. Gutenberg's blessing is safe. If you believe interruptions or multiplicity of subjects are beneficial, you can employ them at greater ease that it is the case with book reading. At the other extreme, you may wish to take on thousands of independent articles, make interruptions a norm, focus your reading only on portions that you deem most important, etc.

A rule of the thumb is: use traditional reading when you read stories or you read for enjoyment. Use [[Glossary:Incremental_reading|incremental reading]] to process learning material, textbooks, notes, or scientific literature that you need to remember for life (or at least for many months).

== Knowledge acquisition: Areas of optimization ==

There are seven main areas where the learning process can be enhanced. Incremental learning plays a role in each, however, the first two areas will benefit only through your growing experience, and ability to quantify your progress:

# '''access to knowledge''': with the advent of the world wide web, we all have information literally at our fingertips. Searching for knowledge with [http://google.com Google] is fast, easy and accurate. [http://en.wikipedia.org Wikipedia] is the biggest encyclopedia in human history. Access to knowledge is no longer a bottleneck of human learning. It virtually became a non issue!  Web's quality and role will continue increasing exponentially as more and more people appreciate its potential and contribute to its growth. There are still many complementary sources of information that compete successfully with the Internet. However, it is only a question of time before you will be safely able to rely on the Internet as your sole source of information. The great benefit of reading the web, as opposed to reading the books, is that the hypertext nature of the web enforces a very compact and usually self-explaining nature of individual articles. A jump to a randomly selected page in an average book will leave you confused due to the context-dependence of the material. On the other hand, it is less likely the same confusion will trouble you in a random jump to a selected page of an equivalent material placed on the web. Web authors usually put more effort to add contexts to the page (at least in the form of hyperlinks). In other words, it is easier to build quality knowledge by reading single pages of the web than by reading single pages of paper books. We are getting closer to the ideals of incremental life-long learning as opposed to thorough-review learning which for many ends with the end of school years. In the busy days of modern society, few can afford a thorough review of their rusty knowledge in individual fields. It is much easier to fix the gaps incrementally: today an article on the structure of the atom, tomorrow an article on a healthy diet, etc. And all that strictly adjusted to individual's interests and professional priorities
# '''selecting knowledge''': you will face the need to fill the gaps in your knowledge in many more areas than your time permits or your memory makes possible. You can ask SuperMemo to help you scrupulously note down and prioritize all areas of knowledge that need an enhancement! In SuperMemo, you are the master of what you learn and what you neglect. Your ability to select valuable information will grow in proportion to the acquired knowledge. Incremental learning helps you quantify your knowledge and extrapolate into the future. This will help you be more selective in choosing your learning materials.
# '''reading''': this is the first stage where knowledge makes an actual intimate contact with the brain. Traditionally, it is streamed into memory in a more or less linear manner (i.e. paragraph after paragraph). Incremental learning helps you delinearize this process and optimize reading by enhancing '''knowledge selection and prioritization''' concurrently with reading. For example, you should be able to say &quot;''This paragraph can be processed later''&quot; or &quot;''This paragraph requires utmost attention right now''&quot; or &quot;''This paragraph can be skipped for good even if I decide to read the article again''&quot; or &quot;''I want to read this paragraph again in three days and in more detail''&quot; or &quot;''I want to mark this paragraph with lower priority and come to it only after all higher priority paragraphs have been processed''&quot;, etc. 
# '''representing knowledge''': the way in which you present knowledge will affect comprehension and [[Glossary:Retention|retention]] (i.e. how well you remember). Things that are simple are easier to understand. Things that are simple are also easier to remember. Many people do not realize the degree to which simplicity can affect learning. Many people doubt that even the most complex material can be presented in a very simple way. Einstein noticed that &quot;''it should be possible to explain the laws of physics to a barmaid''&quot;
# '''remembering knowledge''': SuperMemo helps you eliminate the problem of forgetting. You will base all learning on [[Glossary:Spaced_repetition|spaced repetition]] that SuperMemo pioneered almost 30 years ago. SuperMemo produces immense savings in time by scheduling review of the studied material only then when the review is necessary (see: [[SuperMemo|Introduction to SuperMemo]])
# '''life cycle of knowledge''': knowledge in your [[Glossary:Collection|collection]] and in your memory keeps on evolving and maturing. This will involve continual rewording, reprioritizing, and re-associating pieces of knowledge. You will often give up portions of knowledge that become outdated or lose their high-priority status. You will apply the rules of knowledge representation that will make knowledge easy to remember. Your knowledge will also become more associative in time. In other words, it will become a more suitable ground for making intelligent choices 
# '''using knowledge''': knowledge translates to value when it is properly used. In the long-run, skills discussed in this article will indirectly help you become more creative and skillful in using your own knowledge. Not surprisingly, your skills needed to efficiently use knowledge are also part of knowledge itself and tend to grow spontaneously as your knowledge increases

== Cost-Benefit Analysis ==

=== Cost of knowledge ===

Each piece of knowledge stored in SuperMemo adds to the total cost of learning in terms of time. If knowledge is [http://super-memory.com/articles/20rules.htm well-formulated], we can estimate that a single element will be repeated from 8-20 times in a lifetime. If a single repetition time falls in the range from 3 seconds to 15 seconds, we could conclude that the expected lifetime cost of a single item ranges from 24 seconds to 5 minutes. SuperMemo has a statistical measure called '''[[Statistics#Cost|Cost]]'''. It measures the total time cost of a single memorized element. In a well-managed long-term learning process, this cost is usually estimated at 2-3 minutes/item in a 10-year bracket. This is then more than the theoretical prediction due to the fact that each collection contains a small subset of so-called [[Glossary:Leech|leeches]] which dramatically increase the average cost/item (note that [[Glossary:Leech|leeches]] can easily be [[Leeches|detected and removed]] from the learning process).
The rational criterion for deciding if a piece of knowledge should be memorized by means of SuperMemo is to judge the benefits of having the given piece constantly available in memory. If the benefits do not seem to add up to more than 10 minutes, the student might simply not add an item to SuperMemo. With a dose of practice, this analysis becomes a semi-automatic process, and should painlessly blend with your life

<div class="bs-callout bs-callout-danger">
The cost-benefit criterion is: '''if costs of not knowing a piece of knowledge is greater than the cost of repetitions in a given period of time, add this piece of knowledge to SuperMemo''' (otherwise do not add
it, or relegate it to lower priority).
</div>

=== Cost of items ===

Cost of a single [http://super-memory.com/articles/20rules.htm well-structured] [[Glossary:Item|item]] in lifetime ranges roughly from 24 seconds to 5 minutes. At any given moment it may be approximated for a 30-year bracket, by multiplying '''[[Element data#Repetitions|Repetitions]] : Future''' (in [[Element data|'''Element data''' window]]) by '''Avg time''' (in [[Statistics|learning statistics]]). SuperMemo displays this value in the [[Element data|'''Element data''' window]]. If your [[Glossary:Item|item]] is ill-structured (i.e. difficult to remember), this cost may bloat! To eliminate ill-structured [[Glossary:Item|items]] use [[Leeches|leech analysis]].

=== Benefit ===

Each piece of knowledge in your [[Glossary:Collection|collections]] should be associated with a tangible benefit. Only '''you''' can accurately guess the value. For example, the value of memorizing the opening hours of your gym in a given time might be approximated by multiplying (1) the probability you will choose wrong hours by (2) the time-cost of missing the gym. For example, if you believe that the probability is 25% and the cost of choosing wrong hours is 40 minutes, the cost of knowing the opening hours is around 10 minutes. In such a case, memorizing the opening hours will be cheaper than missing the gym. However, if the probability is sufficiently low or the time loss sufficiently small, you should not add opening hours to SuperMemo. For example, if the probability is 10% and time loss is 3 minutes, you are not likely to recover your learning investment! There are naturally less clear-cut cases in-between. If you are not sure if you should add a piece of knowledge to SuperMemo, add it with sufficiently low [[Glossary:Priority|priority]]. With overflow tools (e.g. [[Glossary:Auto-postpone|auto-postpone]], [[Glossary:Auto-sort|auto-sort]], etc.), you can drag along less valuable pieces of knowledge at little cost.

=== Priorities ===

You will quickly realize that there are awfully many pieces of information that pass the cost-benefit criterion. You cannot ever hope to learn all this information. This is actually optimistic. This means that with a good selection of knowledge, you can gain far more than just a few minutes on memorizing the gym hours. You may gain untold hours by learning things that transform your life. Your best tool in making sure you always focus on the most beneficial material is to use the [[Glossary:Priority_queue|priority queue]]. If you have too many [[Glossary:Item|items]] to remember, those of lower [[Glossary:Priority|priority]] will get a less meticulous treatment and will carry a higher probability of forgetting. However, this also means that they will benefit from the [[Glossary:Spacing_effect|spacing effect]] and you will learn more of such [[Glossary:Item|items]] at higher speed (despite lower [[Glossary:Retention|retention]]).

== Selecting knowledge ==

The [http://en.wikipedia.org/wiki/Library_of_Congress Library of Congress] holds 10 terabytes of printed materials. Global knowledge resources can be measured in [http://en.wikipedia.org/wiki/Petabyte petabytes] (1000<sup>5</sup> bytes). All digital information has already reached [http://en.wikipedia.org/wiki/Zettabyte zettabytes] (1000<sup>7</sup> bytes). Only a micro-fraction of those resources can be mastered by an individual in a single human lifetime. Even a single copy of the ''[http://britannica.com Encyclopaedia Britannica]'' goes in detail far beyond what a single human being can encompass in a lifetime! The actual speed of learning and lifetime learning limits can be measured with SuperMemo (see: [http://super-memory.com/articles/theory.htm Theoretical aspects of SuperMemo]).

The microscopic capacity of the human brain has not prevented it from building the present civilization as we know it. The human power comes from:

# '''collective effort''' - a billion heads is more than one
# '''specialization of labor''' - all collective tasks are subject to top-down functional decomposition and a single brain usually only needs to process a fraction of information at a time 
# '''knowledge selection skills''' - the associative power of the human brain combined with the selective nature of forgetting help us retain memories that are actually most useful in problem solving

Incremental learning is a great way to combat forgetting. However, forgetting plays an important role in our lives. It runs a valuable garbage collection on knowledge we acquire daily. If the power of forgetting is diminished (as it is the case in incremental learning), your responsibilities in the area of selecting knowledge increase manifold!

<div class="bs-callout bs-callout-warning">
'''SuperMemo will help you eliminate forgetting! At the same time, it will increase your responsibility for selecting knowledge that is truly important and applicable. If used without care and attention, SuperMemo may actually waste your time by helping you remember reels of garbage trivia'''
</div>

A piece of information that occupies just several bytes of your hard disk may carry a relative value that my translate to a net gain of millions of dollars as well as a net loss of millions of dollars. It may also carry no value whatsoever. For example, a sentence written in French &quot;''SuperMemo vous aide a mémoriser et apprendre diverses informations comme une langue, des chiffres, etc.''&quot; may be of nearly zero value for someone who does not know French. At the same time, an [[Glossary:Item|item]] related to a [http://en.wikipedia.org/wiki/Heimlich_maneuver#Abdominal_thrusts Heimlich maneuver] can save the life of a family member. We know that the expected payoff equals the value of the payoff multiplied by its probability. Therefore, the [http://www.nlm.nih.gov/medlineplus/ency/article/000047.htm low probability of a family member choking] and the probability of actual successful application of the maneuver make the value of &quot;Heimlich item&quot; a fraction of the value of the human life. At the same time, even minor errors in medical knowledge of a physician can actually cost somebody's life and carry substantial negative value!

<div class="bs-callout bs-callout-warning">
'''Frequently, you will find more benefit in memorizing the three best things you have learnt today than in memorizing a whole monothematic article to the last detail!'''
</div>

== Readers are leaders ==

You must have heard it from your parents or teachers that &quot;''Readers are leaders''&quot;. [http://en.wikipedia.org/wiki/Harry_truman Harry Truman] added a pessimistic twist to that claim: &quot;''Not every reader is a leader, but every leader is a reader''&quot;. That qualification tells you that reading is not a panacea, and that making reading smart is vital. With the advent of the web, [http://en.wikipedia.org Wikipedia], blogosphere, and other sources of knowledge, access to information is easier than ever. This also means that information becomes more overwhelming. There are always too many things to read. And it is hard to select from a myriad of essential sources. This is no longer a world with a hundred books of which you needed to pick five. Now you literally have thousands of menu options for a single meal. Old bookworm dilemmas are now even more painful. Search the web for solutions for book selection, and you will literally find dozens of blogs with advice. However, most of the advice is based on volatile skills that are hard to master. You need to become an expert in book selection, in speed-reading, in skimming, in skim-or-read decisions, in highlight-and-review strategies, in margin notes and in underlines. For traditional massive reading mastery (without SuperMemo), you would need to go for the extreme: reading in the toilet, reading on the train, reading on your mobile phone, making use of your dead time, etc. You would need to ruthlessly control your
emotions with &quot;not to read&quot; decisions, or &quot;not to finish&quot; decisions, or &quot;not to re-read&quot; decisions, etc.

If you master incremental learning, you will not solve the problem of information overload. However, you will systematically address all issues that pester massive readers. Once you embark on the only rational reading road, you will free your mind from all the dilemmas and stresses involved in reading. Incremental learning helps you focus on learning itself!

As for the bathroom time, incremental learning can free that for creative thinking. After all, you also want to ponder your own ideas, not only process other people's output.

<div class="bs-callout bs-callout-warning">
'''[[Glossary:Incremental_reading|Incremental reading]] solves all major dilemmas of massive readers! Incremental readers are leaders.'''
</div>

== Does incremental reading slow down mastering complex subjects? ==

Could [[Glossary:Incremental_reading|incremental reading]] slow down or hamper learning complex subjects such as [http://en.wikipedia.org/wiki/Quantum_mechanics quantum mechanics]? Is complex knowledge based on just &quot;remembering passages&quot; or does it go well beyond it. Any form of interruption or delay while working on many things in parallel can potentially slow down this process. Many people believe that [[Glossary:Incremental_reading|incremental reading]] is good for cramming, but not for serious learning, esp. when complex subject matter is concerned.

=== Example Opinion 1: tackling quantum mechanics ===

Here is an example opinion of someone who never tried incremental learning:

<blockquote>
''I believe that [[Glossary:Incremental_reading|incremental reading]] (IR) may slow down or hamper learning complex subjects that require deeper understanding -- as opposed to superficial remembering. From what I'm told, in IR, the student spends but a few seconds on a passage related to one subject, and then jumps to another, completely unrelated subject. Such switching among disparate subjects makes it difficult, if not impossible, to engage in deeper thoughts and thus to discover deeper truths beyond the superficial remembering of passages. When I read, I look for concepts and methods that underlie the necessarily descriptive language used to convey them, and I try to imagine their applicability in my other pursuits. In other words, discovering deeper truths requires deeper thoughts. The latter are difficult to reach if only a few seconds are spent on a subject at a time. As a practical aside, technical papers in hard sciences and engineering are peppered with notational inconsistencies among different authors, different research groups, and different journals. The same symbols may mean different things in different scientific papers. Similarly, different authors may use different symbols to denote the same quantity. Sometimes the differences are gross and easy to spot, and sometimes they are only subtle. When reading one paper at a time, the reader familiarizes him/herself with the terminology and notation, and can follow the line of reasoning while recognizing the subtleties. Having a few dozen (or hundreds!) articles read in parallel, frequently switching among them, creates a nightmare scenario of confusion where the meaning is lost in the jungle of various notations and subtle terminology differences. This makes it impossible to follow the line of reasoning in any one of the papers. Below are a couple of (imperfect) metaphors that may convey the sense of loss when trying to apply IR to learning complex subjects and to discovering deeper truths:''
# ''If you are a wine connoisseur, imagine that you are only allowed to smell the wine, but that you are not allowed to drink it, or even to put it in your mouth. How good would your knowledge of wine be? With practice, you would probably learn to distinguish various wineries, age, etc. You may even be more efficient at it, and be able to do it in record time. Yet, I venture a guess that you would feel something missing, that you would feel somehow cheated, that you would feel that you don't know all that there is to know about wine. And rightfully so. Same with IR. You just &#39;smell&#39; the knowledge, but you never &#39;taste&#39; it. It may be enough if your goal is to learn a foreign language. But some areas require that you dig deeper if you want the real insight, the real truth.''
# ''Physical exercise. A well-rounded routine has both cardiovascular and resistance training elements to it. IR is like having only the high-intensity cardio. Sure, if you stick with cardio, after a while you will be able to deliver a small package across town in record time -- and that may be all you aspire to. However, it will not give you the upper body strength to help move a 400-kg optics bench. Same with IR: you may impress your audience with the breadth of your knowledge, but you will have to leave the heavy intellectual lifting to others.''
</blockquote>

The above reasoning can easily be proven false. We can imagine two extremes:

* all learning is done using traditional textbook processing with all the academic arsenal of teachers, practise, reasoning, interaction, labwork, and more
* all learning is purely incremental with no breaks for free thought, calculations on the margin, walking around, conversation, etc.

It is easy to show that both are flawed. For example, traditional learning is deprived of [[Glossary:Spaced_repetition|spaced repetition]], while pure incremental learning may be seen as deprived of conversation or lab practise. Clearly, the optimum falls somewhere in the mix between the two extremes. '''Incremental learning is a supplementary tool in well-rounded education'''. It is not supposed to monopolize your day and your thinking. It should not be seen as a replacement, but as an enhancement. As such, it is up to you to find your optimum.

[[Glossary:Incremental_reading|Incremental reading]] is all you want it to be. It can be speed-reading, cram-reading, or mass-reading. It all depends on the priority criteria which you choose. For that reasons, it would be best described as a reading management technique. As such it is indispensable independent of the complexity of the studied matter. On one hand, you can speed-read articles faster than in conventional speed-reading and yet leave vital paragraphs for future review. On the other hand, you can meticulously dismantle individual paragraphs and convert them into classical questions-answers material that will stay with you for ever. In addition, you can freely manipulate the volume of the material flowing into the reading/learning process. You can focus on a hundred most important articles or you can opt for thousands. Naturally, in the latter case, your time allocation for individual articles will be minute. For example, if you import 10,000 articles to SuperMemo, you might end up with 50,000 to 100,000 [[Glossary:Extract|extracts]] within a year of 1-hour daily reading. In such circumstances, low priority articles will indeed linger for months in the process. Naturally, this is exactly the purpose of [[Glossary:Incremental_reading|incremental reading]]: focus on what is important without neglecting anything that falls within your area of interest. If your focus changes, you can use search and navigation tools to speed up the review of most important portions of your reading material.

[[Glossary:Incremental_reading|Incremental reading]] is universal in textbook learning. Whatever complex concepts you need to analyze, and whatever computations you need to make on the margins, you can do in parallel with incremental reading of the textbook (assuming you have an electronic version). An old rule says: whatever you need to remember for life (or at least months), process incrementally (to improve memory, boost understanding, and save time in the long term). Whatever you need at the moment, for the sake of understanding the subject, process in bulk (non-incrementally). By combining the two, you can get the most of your learning.

Metaphorically, a pencil is a useful tool that can enhance your life. You will not want to replace your computer with a pencil, or use a pencil while cooking. However, you can still enhance your life by having the pencil handy. Needless to say, we believe, [[Glossary:Incremental_reading|incremental reading]] is far more useful than a pencil.

The presented reasoning is not only wrong, it is also based on a few misconceptions about incremental learning:

* It is not true that, in [[Glossary:Incremental_reading|incremental reading]], you spend only a few seconds on each [[Glossary:Topic|topic]]. The time depends on your needs. It may be a few seconds or it can be an entire day (e.g. before an exam, when doing research, or when being consumed with one's learning passions)
* It is conceivable that an incremental reading novice would suffer from the lack of knowledge of the toolset and get lost in the process thus making it more confusing than it is the case with standard textbook learning. Weighing up pros and cons should always consider an ideal case in which the student truly understands the methods of incremental learning. This often takes months of use. The learning process should begin with easier concepts and easier subjects before more complex subjects can be tackled.
* Some research papers are less suitable for [[Glossary:Incremental_reading|incremental reading]]. Papers that require a vast use of working memory or a large investment in short-term memory '''are not''' suitable for incremental reading. [[Glossary:Incremental_reading|Incremental reading]] is useful each time you study knowledge with lifetime application. Research papers with complex methodology and rich in new specific symbols and notations are not suitable. You should read such papers in your own way and leave just a few notes (&quot;deep thoughts&quot;) in SuperMemo. Use your working memory while it lasts and dump it right after reading. Naturally, for massive learning, you will target texts that produce a minimum discardable burden of short-term memory, i.e. are well-written with concise formulations focused on lifelong value.
* For novices, knowledge acquired with [[Glossary:Incremental_reading|incremental reading]] may not taste as sweet as for a professional. In the latter case, the &quot;taste&quot; may be exquisite due to a well-established contextual knowledge and solid semantic context that relies strongly on long-term memories. A seasoned incremental reader will have a totally opposite feelings about the wine metaphor. Not only will [[Glossary:Incremental_reading|incremental reading]] allow of sensing the intricate fabric of the full spectrum of volatiles. Instead of just a single evening, the taste, aroma and pleasure will last a lifetime!
* The effects of [[Glossary:Incremental_reading|incremental reading]] will always depend on the goals. If the goal is to impress some audience, it will likely be accomplished. However, if the plan is to do some &quot;heavy intellectual lifting&quot;, [[Glossary:Incremental_reading|incremental reading]] will provide a rich toolset to enhance that process and provide you with solid long-term knowledge grounding needed for the job.

=== Example Opinion 2: struggling with basic physics ===

Here is another example opinion of someone who tried incremental learning but found it difficult:

<blockquote>
<p>''I think [[Glossary:Incremental_reading|incremental reading]] is either very difficult or impossible to use when learning some concepts of physics. For example, I have the following text about the Earth and the Sun, how would you handle this with incremental reading?''</p>
<p style="color: #00F;">''The Earth is moving very very slowly away from the Sun. This happens for two reasons. The first is that the Sun is constantly losing mass because of the solar wind. As the mass of the Sun decreases its pull on the Earth decreases and so the Earth moves slightly further away. The second reason is to do with tidal forces. In exactly the same way that the Moon is slowly moving away from the Earth, the Earth is very slowly moving away from the Sun. In the Earth-Moon case the Moon pulls on the Earth creating tides and slowing the Earth’s rotation very slightly, making the day longer. This action has a reaction - the Moon's orbit is speeded up. If something travels faster it must move outwards to remain in an orbit and so the Moon slowly drifts away from us at a rate of 3.8 centimeters per year. The same situation happens with the Sun but the Earth’s influence on the Sun is much smaller than the Moon’s influence on the Earth. The result is the Earth’s tiny tiny drift away from the Sun''</p>
</blockquote>

The example isn't much harder to process than other pieces of knowledge suitable for [[Glossary:Incremental_reading|incremental reading]]. For success in similar cases one needs an encyclopedic text or a degree of editorial effort to dismantle portions of more elaborate prose. The presented example poses two obstacles:

# '''Implicit enumeration'''. The text mentions two reasons why the Earth moves away from the Sun but it does not name them in an explicit sentence. One needs to read the entire passage to find out the second reason. 
# '''Explaining by analogy'''. The effect of tidal forces on the Sun is explained by describing similar forces created by the Moon. You cannot extract the &quot;second reason&quot; without including and understanding the &quot;''Moon example context''&quot;.

Here is how the presented text would be handled with [[Glossary:Incremental_reading|incremental reading]] (note the editorial effort as well as the need to entirely rephrase one of the sentences):

* [http://super-memory.com/articles/users/earth_files/index.htm progress of the incremental reading process]
* [http://super-memory.com/articles/users/earth_files/earth.txt resultant questions and answers]

Some authors make [[Glossary:Incremental_reading|incremental reading]] very difficult by assuming a great deal of knowledge on the part of the reader or, as it is the case here, loading student's working memory with a great deal of data rather than building knowledge gradually (i.e. from the ground up). In the discussed example of quantum mechanics, the basic vocabulary needed to process any meaningful text is pretty extensive and capitalizes on many branches of physics and mathematics. In addition, good understanding of all concepts requires a great deal of mathematical practice that goes beyond linear reading.

== Incrementalism and interruption are not obligatory ==

If you love to go through texts from cover to cover without a stop, and always find it hard to stop reading a book before sleep, you may still benefit from [[Glossary:Incremental_reading|incremental reading]].

There are no incremental readers who did not begin from misgivings about interruption. Paradoxically, '''the stronger your misgivings, the better candidate for a good incremental reader you might be'''!

A popular misconception is that there are impatient people who are predisposed to be incremental readers - let's call them '''sippers''' - and those who love to devour knowledge in large chunks - let's call them '''gulpers'''. The truth is that all creative individuals are of a gulper nature. Incrementalism is both a skill and a habit all gulpers may learn over time.

Nobody loves SuperMemo as of the first day. It may take a few weeks to notice its power. And still, as we do not have sensors of the speed of forgetting, you need a dose of rational mathematical appreciation of what SuperMemo does to your brain. You cannot easily sense the power of knowledge and how fast it is being undermined by forgetting.

[[Glossary:Incremental_reading|Incremental reading]] takes far longer to be appreciated than SuperMemo itself. To employ SuperMemo, you need to learn only two operations ('''[[Add new]]''' and '''[[Learn]]'''). For [[Glossary:Incremental_reading|incremental reading]], you need a toolset that keeps growing and improving over the years of use. Yes! Even after a few years of learning, you will discover new ways you can speed up your own learning with [[Glossary:Incremental_reading|incremental reading]]. It may take a year before you might notice first signs of addiction to [[Glossary:Incremental_reading|incremental reading]]. A benign form of addiction, mind you, with few negative side effects.

The paradoxical tendency of &quot;gulpers&quot; to become good incremental readers comes from their hunger for knowledge. The fact that you cannot stop reading is a powerful expression of this hunger and it is the primary driving force that will help you become an addictive reader. What you might be missing still is the understanding of the power of [[Glossary:Incremental_reading|incremental reading]] and the hunger to &quot;''switch for more''&quot;. [[Glossary:Incremental_reading|Incremental reading]] will help you develop a hunger for maximizing the value of information you are processing at any given moment.

You can begin [[Glossary:Incremental_reading|incremental reading]] today without ever having to stop reading an article that you find fascinating.

<div class="bs-callout bs-callout-warning">
'''In [[Glossary:Incremental_reading|incremental reading]], interrupted reading is a norm, but is NOT obligatory!'''
</div>

You can read all articles from front to back and only use incremental reading tools for prioritizing articles and extracting most important sentences and converting them to [[Glossary:Cloze_deletion|clozes]]. In other words, you do not need incrementalism to achieve solid [[Glossary:Retention|retention]] of knowledge. An ordinary web surfer has only two alternatives when encountering an article that seems worth reading: (A) ''Fascinating, let's read,'' and (B) ''Not fascinating enough. Perhaps I will read some other time''. In contrast, an incremental reader can determine the [[Glossary:Priority|priority]] of the article and always read only the articles from the top of the current [[Glossary:Priority_queue|priority list]] (perhaps with a user-defined degree of randomization). Moreover, at any time, he or she can say: ''Interesting, but not as much as I thought. Let's downgrade the priority and come back later (if ever)''.

A gulper is driven by a natural neural mechanism that underlies all human progress: '''curiosity'''. The same mechanism can be used to magnify incrementalism: curiosity of what article or paragraph comes next. Once you develop a healthy incremental reading process, you will add another natural neural mechanism: '''impatience'''. Impatience is also a buttress of progress. We do not like long stretches of low efficiency. We like instant gratification of success and the bigger the success the better. In [[Glossary:Incremental_reading|incremental reading]], you are constantly driven by curiosity and yet you itch-to-switch as soon as the text you are reading does not bring sufficient value-per-time. The healthier your incremental reading process, the more value per second you can extract. You will develop a sense of the realistic average value stream, and each time you fall below that expectation, you will add up to the incremental nature of reading (even if the fault is yours, not the text author's, e.g. when the gaps in your knowledge produce poor comprehension). By combining curiosity with impatience, you can convert from a ''gulper'' to a ''sipper''. And still you will be able to read top-quality articles top-to-bottom without interruption. '''[[Glossary:Incremental_reading|Incremental reading]] helps you prioritize by content instead of reacting to transient evaluative impressions'''.

You will notice that incremental attitude is a habit you grow as your technical and parsing skills improve. Rarely will you delete lower quality articles, but these will fade in priority and may indefinitely linger in the process. As a result, you will maximize the educational effects of every precious second you spent on learning.

As an incremental reader, you might gradually develop a dislike of old-style books (as opposed to importable e-books and articles). If you choose to read a book, it is as if you said: &quot;''this is the most important reading material in the whole world''&quot;. Then the whole series of paragraphs in the book are considered the most important paragraphs to read in their precise sequence as they appear in the book. You give the author of a book God-like powers to stream information into your brain in a flawless, omniscient, and omnipotent way.

<div class="bs-callout bs-callout-warning">
'''Gulpers and sippers are not biologically different!''' The conversion from one to the other goes via the understanding of incremental learning, mastering SuperMemo toolset, honing the skills, self-control, rationalization of the learning process, and gradually pumping up the average value of knowledge streamed into one's memory.
</div>

== Incremental learning for perfectionists ==

If you are a perfectionist, you may initially have problems with accepting the chaos of incremental learning. You may wonder why you should leave your [[Glossary:Cloze_deletion|cloze deletions]] unfinished before they look perfectly adorned with fonts, stylesheets, and pictures.

If you give incremental learning a more determined try, you will understand that the opposite is true: '''perfectionists should love incremental learning'''! Your perfectionist nature should accept the overriding rule: maximum quality knowledge in minimum time. It is not the beauty of [[Glossary:Cloze_deletion|clozes]] in your [[Glossary:Collection|collection]] that counts, but the beauty of knowledge in your mind. For a skillful student, incremental learning is based on a set of perfectly-formed strict and rigid rules that guarantee the maximum speed of [[Glossary:Acquisition_rate|knowledge acquisition]]. It is true that some of these rules can make you uneasy at first. If you see a sentence that qualifies for a [[Glossary:Cloze_deletion|cloze]], the rule is: execute the [[Glossary:Cloze_deletion|cloze deletion]] and defer worrying about its exact formulation to its first repetition. The mere choice of the cloze keyword will leave sufficient traces in your memory to qualify as a repetition. In such circumstances, perfecting the formulation of the [[Glossary:Cloze_deletion|cloze]] will become art for art's stake. A higher level rule is: '''minimum work for maximum memory effect'''. Therefore, you will improve the formulation of the [[Glossary:Cloze_deletion|cloze]] as soon as you proceed with the first repetition. And again, you will do only as much work as it is necessary to successfully complete a single repetition act. Again you give up details and frills. Ultimately, your [[Glossary:Cloze_deletion|cloze]] will become perfectly formulated, perfectly prioritized, and perfectly placed in your [[Glossary:Knowledge_tree|knowledge tree]]. Alternatively, it will be deleted or left lingering in your &quot;to do&quot; subset. It is the perfect rules of incremental learning and the perfect learning results that should feed your perfectionist needs, not the perfect &quot;look&quot; of your learning material.

Many people tend to hold the world wide web in contempt calling it the &quot;human information garbage dump&quot;. This attitude makes it hard to utilize the web as the &quot;goldmine of human knowledge&quot;. [http://en.wikipedia.org/wiki/Tim_Berners-Lee Tim Berners-Lee] created &quot;perfect rules&quot; ([http://en.wikipedia.org/wiki/HTML HTML], [http://en.wikipedia.org/wiki/HTTP HTTP]) for knowledge dissemination by the populace. We can adapt our own &quot;perfect rules&quot; for mining the web. Incremental learning uses &quot;perfect rules&quot; to convert web data into golden knowledge. As a perfectionist, you should not worry about the chaos of the web or chaos of your [[Glossary:Collection|collection]]. What really matters is the perfect golden end-result: wisdom!

Finally, if you still cannot live with imperfectly formulated [[Glossary:Cloze_deletion|clozes]], nothing prevents formulating them perfectly. The formulations may be more satisfying to your perception, but you will, naturally, learn at a slower rate.

== Incremental reading is not an attention destroyer ==

In incremental learning, it is very important to make the right choice of the learning material. Many texts or videos are unsuitable for incremental processing. SuperMemo user who never tried incremental learning
wrote:

<blockquote>
<p>''It is possible that article structure and quality do matter that much in [[Glossary:Incremental_reading|incremental reading]] because the real learning bottleneck is the human brain (speed of cortex plasticity)? What if you can only absorb X items of knowledge each day -- if you try to do more, you do something bad to your brain? Here's one idea: If you overload your brain with new stuff, it won't have the time to form meaningful connections between the things you know, so your knowledge may be reduced to the ability to answer gameshow-like questions.<br>
Solving any significant problem requires periods of prolonged concentration. I fear that an information addiction (200 new tidbits per day) leads to attention deficit. Your brain is used to getting something shiny and new every 15 seconds (a new tweet, new funny pic, new headline, etc.), so when you tell it to work on one thing for 4 hours, it doesn't listen. I think I was able to concentrate more on one thing in times before Web surfing.''</p>
<p>''So, to put a tabloid spin on it, '''[[Glossary:Incremental_reading|incremental reading]] could be the ultimate attention destroyer!!!'''''</p>
</blockquote>

There is a lot of truth in the above reasoning in reference to attention, memory bottlenecks, &quot;meaningful connections&quot;, etc. However, comparing SuperMemo to [http://twitter.com Twitter] or [http://facebook.com Facebook] as used by Internet junkies is very inaccurate! The reward in [[Glossary:Incremental_reading|incremental reading]] is based on quality learning, not &quot;something shiny or something funny&quot;. Naturally, nothing prevents &quot;shiny/funny&quot; things to be imported to SuperMemo. This is why the nature of the ultimate reward will also depend on one's personality and self-discipline.

When employed along the recommended rules: '''[[Glossary:Incremental_reading|incremental reading]] should dramatically increase the attention''' (as explained in [[Incremental reading#Advantages of incremental reading|Advantages of incremental learning]]).

=== Learning speed bottleneck ===

Cortex plasticity is indeed the bottleneck in the learning process. All speed-reading and speed-learning efforts may go to naught if you do not employ [[Glossary:Spaced_repetition|spaced repetition]], which ultimately determines the speed of establishing long-term memories. Remember that in [[Glossary:Incremental_reading|incremental reading]], the volume of the material may be very high, however, the ultimate number of [[Glossary:Item|items]] entering the learning process, in the ideal case, is relatively small (usu. 10-20 per day, not 200!). It simply takes lots of time to fish for golden knowledge that will bring best value in the long term.

=== Overloading memory and the role of sleep ===

The existence of the memory bottleneck is the direct consequence of the overload concern. You can overload your learning process with excess information, however, you are unlikely to &quot;overload your long-term memory&quot;. The processes of forgetting and garbage collection executed in sleep have evolved precisely to prevent this problem. However much you try to learn excess facts, forgetting will clean up the excess, and memory optimization in sleep will ensure you develop all necessary &quot;meaningful connections&quot;. Naturally, this will happen only if you get all sleep that you need (i.e. avoid using alarm clock, sleeping pills, staying up late, etc.).

For more see: [http://super-memory.com/articles/sleep.htm#Neural_optimization_in_sleep Neural optimization in sleep]

=== Learning vs. problem solving ===

It is true that solving problems requires high concentration. However, in the ideal world, you should devote separate time slots to (1) learning and (2) problem solving. In Covey's terminology, your learning boosts the ''Production Capacity'', while your problem solving time is your ''Production'' time. Naturally, you can marry the two slots when problem solving occurs in conditions of information deficit. [[Glossary:Incremental_reading|Incremental reading]] is an ideal tool for such situations. You can combine the inflow of new information with creative efforts and problem solving while retaining maximum focus on the problem at hand. This is explained in [[Incremental reading#Advantages of incremental reading|Advantages of incremental learning]]: [[Incremental reading#Creativity boost|Creativity]]. You can optimize the degree of monothematic focus by using various tools of [[Glossary:Incremental_reading|incremental reading]], esp. [[Subset learning#Search and review|search&amp;review]] as well as [[Subset learning#Branch review|branch review]].

See also: [[Creativity and problem solving in SuperMemo#Incremental problem solving|Incremental problem solving]].

=== Internet addiction ===

Internet distractions can be focus destroyers when working with SuperMemo. However, this is more a matter of self-discipline than an inherent problem associated with SuperMemo. It is up to you to decide if you wish to stray to [http://facebook.com Facebook] or [http://twitter.com Twitter]. [[Glossary:Incremental_reading|Incremental reading]] may encourage a degree of straying (e.g. to import supplementary material from [http://en.wikipedia.org Wikipedia] and/or dictionaries). The whole concept of the [[Glossary:Priority_queue|priority queue]] was developed precisely to counteract the cost of such straying. In [[Glossary:Incremental_reading|incremental reading]] you stray, import, prioritize and... forget (about the excursion to the web). You may visit 20 pages, but instead of wasting time, you import and prioritize. You are back onto your focused path in minutes. The whole process can be under your rational control and the web may become an ally rather than an enemy. Metaphorically speaking, an Internet junkie is constantly distracted with shiny titbits, while incremental reader focuses on fishing for golden knowledge.

=== Incremental reading boosts attention ===

[[Glossary:Incremental_reading|Incremental reading]] increases attention by letting you focus on a manageable portions of knowledge without feeling overwhelmed, without straying, without getting stuck on harder material, without worrying that you might miss important pieces when speed-reading, etc. Your best way to experience that improvement is to try [[Glossary:Incremental_reading|incremental reading]]. However, you should know that the effects will not be instant. You will need to invest a lot of time in learning the tools, and then even more time in honing your strategies and learning about your own memory and reasoning. Few incremental readers become truly enthusiastic in their first months of learning.

=== Article quality matters in learning ===

In [[Glossary:Incremental_reading|incremental reading]], you will quickly develop skills needed to instantly differentiate between high quality articles and articles that are full of fluff and wasteful prose. You will indeed fish for catchy headlines, meaningful sections, minimum off-topic commentary, etc. Article quality will determine your ability to employ speed-reading, and to quickly prioritize your material. This has nothing to do with instant gratification obtained from social media, instant news, and other net distractions.

== Can you really read thousands of articles at the same time? ==

A visitor to [http://super-memory.com super-memory.com] commented on incremental learning claims:

<blockquote>
<div class="bs-callout bs-callout-comment">
''A friend recommended [[Glossary:Incremental_reading|incremental reading]] to me. However, I was instantly turned off by your claim: read thousands of articles at the same time. This is not only impossible, but also sounds like overhyped marketing. All you do in [[Glossary:Incremental_reading|incremental reading]] is split articles into chunks and read those separately. With this method you lose the big picture. I am afraid you are selling snake oil''.
</div>
</blockquote>

That skepticism is understandable, and yet we stick with the original claim: you can indeed read thousands of articles at the same time. This is because ''reading'' can be understood as a ''process'' or as an
''act''. No one can make a sane claim of multiple reading acts at the same time. In [[Glossary:Incremental_reading|incremental reading]], only one article receives a laser focus at any given moment of time. Few students read more than 50 articles in a single day, and they rarely read them all in their entirety. However, they may easily read or skim a few thousand articles in a month, and keep hundreds of thousands of articles in the incremental learning ''process''. In that sense ''reading thousands of articles at the same time'' is an accurate if somewhat enticing description.

To understand [[Glossary:Incremental_reading|incremental reading]], you need to understand SuperMemo (or [[Glossary:Repetition_spacing|repetition spacing]] in general), as well as the effects of intermittent reading on memory. Reading in chunks without the help of the underlying learning process based on [[Glossary:Spaced_repetition|spaced repetition]] indeed makes little sense. However, once you master and understand the techniques of [[Glossary:Incremental_reading|incremental reading]], you will get the effect opposite to the one expected: not only will you not lose the track of the big picture, you will keep a lasting, durable and coherent memory image of whatever you found important in the processed articles.

'''In a well-managed incremental learning process, the big picture is retained in a state of better coherence than it is the case in traditional learning.'''

== Incremental reading vs. memory interference ==

You may have read about interference in learning. When students learn two things one after another, they perform worse than in cases where they focus on one thing. This might sound like a reason to disqualify [[Glossary:Incremental_reading|incremental reading]] as an effective learning method.

It is true that interference can ruin learning. If you read about a subject without fully understanding it and follow it with another subject that is confusingly similar in nature, you will indeed perform worse in
tests. However, this effect is much less pronounced if the first subject is studied with solid comprehension. [[Glossary:Incremental_reading|Incremental reading]] makes it possible to read only as much as you understand. Then it encourages long-term [[Glossary:Retention|retention]] by producing [[Glossary:Cloze_deletion|cloze deletions]]. Finally, it periodically rediscovers weaknesses in the learning process and fills the gap. When well executed, [[Glossary:Incremental_reading|incremental reading]] produces an opposite effect. It minimizes interference by forcing you to resolve contradiction in your material. It ruthlessly punishes all cases of incomplete understanding. In classroom conditions, you can get a foggy pass at subject A, then worsen the fog by digging into subject B. In [[Glossary:Incremental_reading|incremental reading]], SuperMemo will force you to jump from A to B and back to A until the two form a harmonious body of knowledge with minimum interference and maximum connectivity. Note that the same research on interference produces diametrically different results when the interfering [[Glossary:Topic|topics]] are subject to continual re-reading. Re-reading is frequent in SuperMemo and multiple active repetition of [[Glossary:Cloze_deletion|cloze deletions]] is a norm. The outcome of the experiment may also be obscured by adding a degree of novelty to old reviews which greatly improves attention. Better learning follows in the wake.

== Why Wikipedia is better than Encyclopedia Britannica? ==

=== Wikipedia is better ===

* every month [http://en.wikipedia.org Wikipedia] is richer and mainstream articles show an ever-decreasing error rate
* [http://en.wikipedia.org Wikipedia] is better for [[Glossary:Incremental_reading|incremental reading]] due to its structure determined by crowdsourcing
* some articles cannot be found in [http://britannica.com Britannica] (e.g. search for &quot;SuperMemo&quot; or &quot;incremental reading&quot;)

=== Britannica is hard ===

Due to its style, [http://britannica.com Britannica] is often not suitable for [[Glossary:Incremental_reading|incremental reading]]. It is pleasant to read, it explains things, it digresses, it provides examples, however, it is not fact-packed and context-rich like crowdsourced [http://en.wikipedia.org Wikipedia].

Try working the following text with [[Glossary:Incremental_reading|incremental reading]]. The text was taken from an article about sleep from [http://britannica.com Britannica]:

<blockquote>
''In addition to the behavioral and physiological criteria already mentioned, subjective experience (in the case of the self) and verbal reports of such experience (in the case of others) are used at the human level to define sleep. Upon being alerted, one may feel or say, &quot;I was asleep just then,&quot; and such judgments ordinarily are accepted as evidence for identifying a prearousal state as sleep. Such subjective evidence, however, can be at variance with both behavioral classifications and sleep electrophysiology, raising interesting questions about how to define the true measure of sleep. Is sleep determined by objective or subjective evidence alone, or is it determined by some combination of the two? And what is the best way to measure such evidence? More generally, problems in defining sleep arise when evidence for one or more of the several criteria of sleep is lacking or when the evidence generated by available criteria is inconsistent. Do all animals sleep?''
</blockquote>

This piece begins with an ominous context-buster: &quot;''already mentioned''&quot;. It poses interesting questions. However, it will take a while for an incremental reader to find cloze keywords that would answer the most essential question of long-term learning: ''What do I really want to remember for life from the passage I have just read?''

=== Simple English Wikipedia is dangerous ===

Many users believe that [http://simple.wikipedia.org Simple English Wikipedia] is better for understanding problems than [http://en.wikipedia.org Wikipedia] itself. However, simplified terminology leads to terminological imprecision. Even though the language is simpler, the problems presented may get a superficial or misleading treatment. [http://simple.wikipedia.org Simple English Wikipedia] might be an advantage if you want a short story of the Chinese Empire as opposed to a lengthy Wikipedia dissertation. However, if you try to understand complex economics or quantum physics, relying on simpler texts may backfire. Very often, you should rather struggle with complex terminology by importing supplementary material. Going the easy way may take you astray.

== Incremental reading vs. the books ==

Incremental reading of electronic materials is superior to reading books or reading in the browser (given the same quality of the study material). All basic reading, bookmarks, and highlight methods start falling apart once the volume of books or articles increases beyond a certain level.

Without SuperMemo, you won't easily prioritize, sort, organize, schedule, re-prioritize, search&amp;review, etc. The whole SuperMemo &quot;engine&quot; in the background is the most important component in the process. For a larger volume of material, when reading in the browser, you will probably be just 5-10% as effective as with [[Glossary:Incremental_reading|incremental reading]] (in the long run).

In [[Glossary:Incremental_reading|incremental reading]] you can use the following tools:

* article [[Priority queue|prioritization]]
* article scheduling
* article sorting
* [[Glossary:Read-point|read points]]
* [[Glossary:Cloze_deletion|cloze deletions]] (simpler, faster and more accurate than multiple copy, paste and edit)
* [[Glossary:Extract|extracts]] that leave the trace of prior work, set [[Glossary:Read-point|read-points]], schedule, and prioritize most important pieces of information
* [[Subset learning#Search and review|search&amp;review]]
* article auto-split, auto-scheduling and auto-prioritization
* many more

[[Glossary:Incremental_reading|Incremental reading]] is likely to reduce your tolerance for:

* texts with low information content (high prose, low fact)
* poorly structured texts
* texts with referential ambiguities
* texts with low context reference
* texts with poor information anchoring (needed for speed-reading)

[[Glossary:Incremental_reading|Incremental reading]] will sharpen your skills in:

* text selection
* text prioritization
* semantic/structural processing
* mnemonic processing
* speed-reading

== Incremental learning vs. the news ==

=== You will rather not want to use SuperMemo to process news! ===

Donald Rumsfeld claims to live by the rule: &quot;''First reports are always wrong''&quot;.

News carries knowledge that is usually valid in a short term only. Incremental learning is useful only for long-term memories. The greatest benefits of trying to process news with incremental learning is to realize how transient news is, and to moderate the craving for news by making it &quot;old news&quot; through the incremental approach.

These are the main reasons why news is a poor candidate for your learning material:

* knowledge that comes with the news is incomplete. If you hear &quot;breaking news&quot;, you know this is about news that is &quot;broken&quot;. It is incomplete and inaccurate. If you ever think of retaining it in your memory for long, come back to [http://en.wikipedia.org Wikipedia] in a week, or in a month, or in a few years. The news becomes most useful when it becomes history. Most of the time, however, it all gets forgotten by the next &quot;news cycle&quot;
* knowledge that comes with the news is contradictory. If you happen to incrementally juxtapose news with old news, you will realize the extent to which people change their minds as new information becomes available. That evolution in thinking and opinion is probably the most interesting part of news.
* knowledge that comes with the news is scant. What was news yesterday might still be news today with some tiny change in the angle. News is repetitive and monopolized by single &quot;hot&quot; issue. If you watch news once a week, you will rarely miss anything important for your long-term perspective
* knowledge that comes with the news is trivial. Breaking news might be devoted entirely to a plane crash, or a bomb explosion. Unless you or your close ones are directly affected, you hardly ever become a better person by being intimate with the details of the tragedy. Understanding plane crashes or bomb explosions in general is far more relevant. A National Geographic documentary might be a better source of such knowledge. Unless this is your first ever encounter with crashes and bombs, you rarely learn anything of major significance. Your ability to change the world will better be enhanced with other sources of knowledge. Your optimization in this area will also help build a future with fewer bombs and fewer crashes.

News may have more value for you at younger ages when you still need to learn a lot about the world, or when you are trying to learn a new language and want to understand the news broadcast from another country. In such cases, use news mostly as the source of indications of what supplementary material should enter your learning process for you to comprehend the news better.

Unless you study journalism (and need to understand the process), or political sciences (incl. the impact of news on the public), or you just live and die by being up-to-date, you should limit your craving for news and focus on knowledge with long-term applicability.

=== Example: how breaking news breaks the news ===

Consider this funny episode from CNN that illustrates the obsession with &quot;breaking news&quot;. On a sunny Sunday of September 22, 2013, Fareed Zakaria spoke about Angela's Merkel impact on Germany, and Europe:

<blockquote>
''Merkel has taken important steps to help Europe's struggling economies, spending tens of billions of dollars directly and indirectly on them. On the other hand, Merkel has imposed austerity on much of Europe, which has been excessive and counterproductive. Her argument is that it is the only way to get governments like Greece and Italy to become more competitive. Now here's the irony...''
</blockquote>

Now here is a real irony. The intelligent commentary was interrupted by the breaking news related to Angela Merkel herself. The viewer did not get a chance to learn about the historic role of the German leader. Instead he or she got a 2 min. &quot;deal&quot; to watch Angela Merkel vote in a federal election. Three full minutes of relaxing silence and a chance to admire Angela's dress and body language in this historic moment.

Ah yes, if you cannot stand similar interruptions and want to learn about the historic role of Angela, see Zakaria's transcripts at [http://transcripts.cnn.com/TRANSCRIPTS/1309/22/fzgps.01.html FAREED ZAKARIA GPS: Not the Time for Big Sticks; Interview with Bill Clinton; Will Germany's Merkel Practice What She Preaches?]

== Minimum information principle ==

Minimum information principle says that your [[Glossary:Item|items]] should be as simple as possible and always ask a single simple question. Conglomerating information in [[Glossary:Spaced_repetition|spaced repetition]] results in slower learning. For example, you may be tempted to formulate [[Glossary:Item|items]] like this:

<blockquote>
'''Question''': What was decided at the Council of Trent, beginning in 1545, and how long did the Council go on?<br>
'''Answer''': The basic beliefs of the Catholic Church; 18 years.
</blockquote>

Two separate memories should be separated in SuperMemo due to the fact that they nearly always will require different timing of repetitions. If you can always activate the same mental pathway in thinking about the Council of Trent (&quot;neurons firing together&quot; in the same pattern), your particular [[Glossary:Item|item]] has a good chance of surviving long in the process without a [[Glossary:Lapse|memory lapse]]. However, once you build a large database of similar [[Glossary:Item|items]], and you review your sizeable material under the pressure of time, your review will always tend to strip redundant pieces of information. Overtime, your nice [[Glossary:Item|item]] will be reduced to the bare bones of information that will often fail its primary test: applicability in real life. It may happen, that despite zero [[Glossary:Lapse|memory lapses]], in 2-3 years, someone will ask you about a Council of Trent in a new context and you will be amazed that you won't be able to reasonably answer the question despite having all the necessary pieces of information included in your [[Glossary:Item|item]]. Two memories of different difficulty might be compared to two different planes of different flying characteristics. The difficult piece (e.g. 18 year duration of the Council) might be compared to a slow flying plane. The easy piece (here the reference to the Catholic Church) might be compared to a modern jet. Review of the conglomerated [[Glossary:Item|item]] might be compared to flying both planes at the same speed. In an extreme case, this might be impossible. The compromise speed might be too high for a slow plane, which might disintegrate beyond a certain speed limit, while the faster plane cannot slow down enough without stalling. In our memory, forgetting is equivalent to forgetting, while stalling is caused by the [[Glossary:Spacing_effect|spacing effect]]. By doing complex and repeatable reasoning at each repetition, you might act as if handling both planes using remote control. However, this is always difficult and requires lots of focus and deliberation at repetitions. Your brain has natural defenses against such &quot;enforced repetitive reasoning&quot;. It is designed to be &quot;intellectually lazy&quot; and thus energetically efficient. Practice shows that [[Glossary:Incremental_reading|incremental reading]] produces many more [[Glossary:Item|items]]. However, those [[Glossary:Item|items]] are usually much easier to remember. In the end, you spend less time on reviewing 5-10 [[Glossary:Item|items]] than you would spend on an item that would conglomerate information and suffer repeated [[Glossary:Lapse|memory lapses]] or very short [[Glossary:Interval|intervals]].

In the course of the evolution, the brain developed strategies for abstracting away from the details and retaining only the most essential, useful and frequently used information. Those strategies are great for
survival, but aren't as good in reaching our educational goals. Council of Trent is a typical example of knowledge we wish to have, but that is pretty expensive. This is because, for most people, it does not get reinforced in run-of-the-mill conversations, TV shows, daily applicability, or at water cooler at work. The situation might differ if you, in particular, read a lot on the subject matter. This might help the memory establish itself in an efficient manner. [[Glossary:Incremental_reading|Incremental reading]] makes it possible to root such difficult-to-retain knowledge firmly in the context, and still make sure that individual repetitions focus on a very specific and cheap-to-retain memories.

This is how the same paragraph might be processed with [[Glossary:Incremental_reading|incremental reading]], and paradoxically cause a significant saving in time in the long run:

<blockquote>
'''Question''': The Council of <span class="cloze">[...]</span>, which began in 1545 and lasted for 18 years, made decisions about the basic beliefs of the Catholic Church<br>
'''Answer''': Trent
</blockquote>

<blockquote>
'''Question''': The Council of Trent, which began in <span class="cloze">[...](year)</span> and lasted for 18 years, made decisions about the basic beliefs of the Catholic Church<br>
'''Answer''': 1545
</blockquote>

<blockquote>
'''Question''': The Council of Trent, which began in 1545 and lasted for <span class="cloze">[...]</span> years, made decisions about the basic beliefs of the Catholic Church<br>
'''Answer''': 18
</blockquote>

<blockquote>
'''Question''': The Council of Trent, which began in 1545 and lasted for 18 years, made decisions about the basic beliefs of <span class="cloze">[...]</span><br>
'''Answer''': the Catholic Church
</blockquote>

<blockquote>
'''Question''': The Council of Trent, which began in 1545 and lasted for 18 years, made decisions about <span class="cloze">[...]</span><br>
'''Answer''': (the) beliefs of the Catholic Church
</blockquote>

<blockquote>
'''Question''': <span class="cloze">[...]</span>, which began in 1545 and lasted for 18 years, made decisions about the basic beliefs of the Catholic Church<br>
'''Answer''': The Council of Trent
</blockquote>

In the end, if you are sure this [[Glossary:Item|item]] works for you, check its performance in the course of the next few years. If you pass the [[Glossary:Interval|interval]] of 2 years without a [[Glossary:Lapse|lapse]], you can say that this particular [[Glossary:Item|item]] indeed works for you. In that case, there is no disagreement between you and the [http://www.super-memory.com/articles/20rules.htm 20 rules]. It is just that for most people, this [[Glossary:Item|item]] is pretty likely to generate a [[Glossary:Lapse|lapse]] within 2 years even if reviewed at correct timing. Depending on the item difficulty, the number of repetitions in the first 2 years might be as low as 3 or well above 20. If your [https://supermemo.guru/wiki/Forgetting_index_in_SuperMemo#default_forgetting_index default forgetting index] is 10%, this translates to a span from 70% chance of retaining the [[Glossary:Item|item]] to the totally unacceptable 90% chance of forgetting! This last number is little understood and little realized by the users of SuperMemo, and should always make you think a lot about the rules of efficient formulation of knowledge.

= Examples: Incremental reading in action =

The following examples have been collected from various articles published at [http://super-memory.com super-memory.com]. They illustrate how electronic texts can be converted to [[Glossary:Cloze_deletion|cloze deletions]], and how formulation problems are solved at the level of articles, [[Glossary:Extract|extracts]] and [[Glossary:Cloze_deletion|cloze deletions]].

== Example: Cloze deletion ==

Passively processed ideas in the form of sentences rarely leave a durable trace in your memory even if they are reviewed regularly. Very often, as soon as after 2-3 months, you will notice that at review time, you actually do not seem able to recall that you have ever had a given sentence in your [[Glossary:Collection|collection]]. You will quickly discover that you need '''active recall''' in order to remember. Active recall is a process in which you must answer questions. For example, you may be presented with a picture of [http://en.wikipedia.org/wiki/Charles_darwin Charles Darwin] and be asked to recognize his face. In the long run, you need to replace [[Glossary:Passive_review|passive review]] with [[Glossary:Active_recall|active recall]]. Otherwise, your memory of the fact will not be permanently consolidated.

The fastest way of converting simple sentences into [[Glossary:Active_recall|active recall]] material is to use a [[Glossary:Cloze_deletion|cloze deletion]]. Using [[Glossary:Cloze_deletion|cloze deletion]], you work with simple declarative sentences like:

<blockquote>''WW1 was precipitated by the assassination of Archduke Francis Ferdinand of Austria-Hungary by a Serbian nationalist in 1914''</blockquote>

Those sentences are converted into question-answer pairs that can be used in actively stimulating your memory for best recall:

<blockquote>
'''Question''': WW1 was precipitated by the assassination of Archduke Francis Ferdinand of <span class="cloze">[...](country/empire)</span> by a Serbian nationalist in 1914<br>
'''Answer''': Austria-Hungary
</blockquote>

<blockquote>
'''Question''': WW1 was precipitated by the assassination of Archduke Francis Ferdinand of Austria-Hungary by a Serbian nationalist in <span class="cloze">[...](year)</span><br>
'''Answer''': 1914
</blockquote>

<blockquote>
'''Question''': <span class="cloze">[...](war)</span> was precipitated by the assassination of Archduke Francis Ferdinand of Austria-Hungary by a Serbian nationalist in 1914<br>
'''Answer''': WW1
</blockquote>

== Example: Extracts and deletions ==

=== Example text (submitted by a student) ===

<blockquote>''After the discovery of Pluto, it was quickly determined that Pluto was too small to account for the discrepancies in the orbits of the other planets. The search for Planet X continued but nothing was found. Nor is it likely that it ever will be: the discrepancies vanish if the mass of Neptune determined from the Voyager 2 encounter with Neptune is used''</blockquote>

=== Example processing ===

There are many ways in which the processing of texts can be done in [[Glossary:Incremental_reading|incremental reading]]. The example below is just one of the ways.

==== Extract 1 and 4 clozes: Pluto and orbit discrepancies ====

<blockquote>
<span class="clozed">Pluto</span> is too <span class="clozed">small</span> to account for the <span class="clozed">discrepancies in the orbits</span> of the <span class="clozed">other planets</span>
</blockquote>

<ol>
<li>
'''Question''': <span class="cloze">[...](planet)</span> is too small to account for the discrepancies in the orbits of the other planets<br>
'''Answer''': Pluto
</li>
<li>
'''Question''': <span class="clozed">Pluto</span> is too <span class="cloze">[...]</span> to account for the discrepancies in the orbits of the other planets<br>
'''Answer''': small
</li>
<li>
'''Question''': <span class="clozed">Pluto</span> is too <span class="clozed">small</span> to account for the <span class="cloze">[...]</span> of the other planets<br>
'''Answer''': discrepancies in the orbits
</li>
<li>
'''Question''': <span class="clozed">Pluto</span> is too <span class="clozed">small</span> to account for the <span class="clozed">discrepancies in the orbits</span> of <span class="cloze">[...]</span><br>
'''Answer''': planets
</li>
</ol>

==== Extract 2 and 1 cloze: Planet X ====

<blockquote>
<span class="extract">Pluto was too small to account for the discrepancies in the orbits of the other planets.</span> The search for Planet X continued but <span class="clozed">nothing was</span> found
</blockquote>

<ol>
<li>
'''Question''': <span class="extract">Pluto was too small</span>. The search for Planet X continued and <span class="cloze">[...]</span> was found<br>
'''Answer''': nothing was
</li>
</ol>

==== Extract 3 and 5 clozes: Voyager and Neptune ====

<blockquote>
<span class="extract">Pluto was too small to account for the discrepancies in the orbits of the other planets.</span> The discrepancies <span class="clozed">vanish</span> if the <span class="clozed">mass</span> of <span class="clozed">Neptune</span> determined from the <span class="clozed">Voyager 2</span> encounter with <span class="clozed">Neptune</span> is used
</blockquote>

<ol>
<li>
'''Question''': <span class="extract">Pluto was too small.</span> The discrepancies <span class="cloze">[...]</span> if the mass of Neptune determined from the Voyager 2 encounter is used<br>
'''Answer''': vanish
</li>
<li>
'''Question''': <span class="extract">Pluto was too small.</span> The discrepancies <span class="clozed">vanish</span> if the new <span class="cloze">[...]</span> of Neptune is used<br>
'''Answer''': mass
</li>
<li>
'''Question''': <span class="extract">Pluto was too small.</span> The discrepancies <span class="clozed">vanish</span> if the <span class="clozed">mass</span> of <span class="cloze">[...]</span> determined by the Voyager 2 is used<br>
'''Answer''': Neptune
</li>
<li>
'''Question''': <span class="extract">Pluto was too small.</span> The discrepancies <span class="clozed">vanish</span> if the <span class="clozed">mass</span> of <span class="clozed">Neptune</span> determined from the <span class="cloze">[...]</span> encounter with Neptune is used<br>
'''Answer''': Voyager 2
</li>
<li>
'''Question''': <span class="extract">Pluto was too small.</span> The discrepancies <span class="clozed">vanish</span> if the <span class="clozed">mass</span> determined from the <span class="clozed">Voyager 2</span> encounter with <span class="cloze">[...]</span> is used<br>
'''Answer''': Neptune
</li>
</ol>

== Example: Processing an article ==

=== Reading the article ===

Let us have a look at an example of a very short, self-containing article, posted in April 1999 on the [http://cnn.com CNN] website. This short article can be read in minutes and can serve as a positive incentive towards adjustments in your diet.

<blockquote>
Antioxidants may slow aging process, study says April 5, 1999 Web posted at: 9:39 p.m. EDT (0139 GMT) From Correspondent Linda Ciampa -- Research at Tufts University indicates that a healthy diet fortified with certain fruits and vegetables may slow down and even reverse the aging process. Foods rich in antioxidants -- such as blueberries, strawberries, spinach and broccoli -- have what doctors call high ORAC (Oxygen Radical Absorption Capacity) levels. Middle-aged rats who were fed a high ORAC diet in the USDA-sponsored study experienced less memory loss than those given a normal diet. Some of the older, slower rats became as lively as their younger peers after taking antioxidants. &quot;We prevented both some brain and some behavioral changes that one normally sees in these rats when they hit 15 months of age,&quot; said USDA researcher Jim Joseph. Antioxidants are effective in destroying free radicals -- cell-damaging compounds that can help cause cancer and heart disease and speed the aging process. &quot;It's pretty well accepted that aging is due to the production of free radicals. So anything we can do nutritionally to provide additional antioxidants is likely to protect us in the process of aging,&quot; Joseph said. That fact already has prompted many to eat a diet rich in antioxidants. &quot;I look at it as sort of a savings account. I'm benefiting today from eating right, but I'm also going to have it in the future,&quot; said 30-year-old Cori Alcock. &quot;As I age and grow older, I'll have good health as well.&quot;
<hr>
<span class="reference">#Title: Antioxidants may slow aging process</span><br>
<span class="reference">#Author: Linda Ciampa</span><br>
<span class="reference">#Date: April 5, 1999</span><br>
<span class="reference">#Source: CNN</span>
</blockquote>

=== Extracting the essence from the article ===

In the course of reading, you should select the most important sections of the article. The article introduces some facts related to healthy diet and adds a lot of redundant explanations. For your review, you are only likely to need the core message which usually makes up a fraction of the entire text. Please have a look again at the same text with four most critical sections emphasized (numbering is not needed and is used only for your convenience for further reference):

<blockquote>
Research indicates that a healthy (1) <span class="extract">diet fortified with certain fruits and vegetables may slow down and even reverse the aging process</span>. (2) <span class="extract">Foods rich in antioxidants -- such as blueberries, strawberries, spinach and broccoli -- have high ORAC (Oxygen Radical Absorption Capacity) levels</span>. Middle-aged rats who were fed a high ORAC diet in the USDA-sponsored study experienced less memory loss than those given a normal diet. Some of the older, slower rats became as lively as their younger peers after taking antioxidants. &quot;We prevented both some brain and some behavioral changes that one normally sees in these rats when they hit 15 months of age,&quot; said USDA researcher Jim Joseph. (3) <span class="extract">Antioxidants are effective in destroying free radicals -- cell-damaging
compounds that can help cause cancer and heart disease and speed the aging process</span>. (4) &quot;<span class="extract">It's well accepted that aging is due to the production of free radicals</span>. So anything we can do nutritionally to provide additional antioxidants is likely to protect us in the process of aging,&quot; Joseph said. That fact already has prompted many to eat a diet rich in antioxidants. &quot;I look at it as sort of a savings account. I'm benefiting today from eating right, but I'm also going to have it in the future,&quot; said 30-year-old Cori Alcock
</blockquote>

=== Improving the wording of extracts ===

Once you extract important fragments from an article, you may need to reformulate individual fragments to make sure they are fully context independent, free of redundant information, easy to read, and formulated in such a way that the beginning of the fragment serves as the introduction to the latter phrases and not vice versa. Please have a look at the [[#Extracting the essence from the article|example from the healthy diet article]]. We selected four important fragments and these fragments (presented on the left in the table) were reformulated to become fully-independent pieces of information (on the right). Please note that two fragments have generated more than one reworded fragment and that one fragment was deleted as it appeared to be redundant upon closer analysis.

{|class="wikitable"
!The original fragment pasted without change from the CNN article
!Modified fragment: shorter and easier to read (sometimes split into more than one part)
|-
! colspan="2" | [[Glossary:Extract|Extract]] 1
|-
|<ol><li>diet fortified with certain fruits and vegetables may slow down and even reverse the aging process</li></ol>
|<ol type="a"><li>Diet of fruits and vegetables may reverse aging</li></ol>
|-
! colspan="2" | [[Glossary:Extract|Extract]] 2
|-
| rowspan="2" | <ol start="2"><li>Foods rich in antioxidants -- such as blueberries, strawberries, spinach and broccoli -- have high ORAC (Oxygen Radical Absorption Capacity) levels</li></ol>
| <ol type="a" start="2"><li>Examples of foods rich in antioxidants: blueberries, strawberries, spinach and broccoli</li></ol>
|-
| <ol type="a" start="3"><li>Foods rich in antioxidants have high levels of Oxygen Radical Absorption Capacity (ORAC)</li></ol>
|-
! colspan="2" | [[Glossary:Extract|Extract]] 3
|-
| rowspan="2" | <ol start="3"><li>Antioxidants are effective in destroying free radicals -- cell-damaging compounds that can help cause cancer and heart disease and speed the aging process</li></ol>
| <ol type="a" start="4"><li>Antioxidants destroy free radicals</li></ol>
|-
| <ol type="a" start="5"><li>Free radicals are cell-damaging compounds that cause cancer, heart disease and aging</li></ol>
|-
! colspan="2" | [[Glossary:Extract|Extract]] 4
|-
|<ol start="4"><li>It's well accepted that aging is due to the production of free radicals</li></ol>
|After a closer scrutiny, the fragment on the left seems to be redundant when compared with the one listed above. We can delete it from the set
|}

=== Generating cloze deletions ===

We can now convert [[Glossary:Extract|extracts]] generated earlier into [[Glossary:Active_recall|active recall]] [[Glossary:Item|items]] based on [[Glossary:Cloze_deletion|cloze deletion]]:

{| class="wikitable"
!Original [[Glossary:Extract|extract]]
![[Glossary:Cloze_deletion|Cloze deletions]] generated from the [[Glossary:Extract|extract]]
|-
! colspan="2" | [[Glossary:Extract|Extract]] a
|-
| rowspan="3" | <ol type="a"><li>Diet of fruits and vegetables may reverse aging</li></ol>
| <ol><li>'''Question''': Diet of <span class="cloze">[...]</span> and vegetables may reverse aging<br>'''Answer''': fruits</li></ol>
|-
| <ol start="2"><li>'''Question''': Diet of fruits and <span class="cloze">[...]</span> may reverse aging<br>'''Answer''': vegetables</li></ol>
|-
| <ol start="3"><li>'''Question:''' Diet of fruits and vegetables may reverse <span class="cloze">[...]</span><br>'''Answer''': aging</li></ol>
|-
! colspan="2" | [[Glossary:Extract|Extract]] b
|-
| rowspan="3" | <ol type="a" start="2"><li>Examples of foods rich in antioxidants: blueberries, strawberries, spinach and broccoli</li></ol>
| <ol start="4"><li>'''Question''': Examples of <span class="cloze">[...]</span> rich in antioxidants: blueberries, strawberries, spinach and broccoli<br>'''Answer''': foods</li></ol>
|-
| <ol start="5"><li>'''Question''': Examples of foods rich in <span class="cloze">[...]</span>: blueberries, strawberries, spinach and broccoli<br>'''Answer''': antioxidants</li></ol>
|-
| <ol start="6"><li>'''Question''': Examples of foods rich in antioxidants: <span class="cloze">[...]</span><br>'''Answer''': blueberries, strawberries, spinach and broccoli</li></ol>
|-
! colspan="2" | [[Glossary:Extract|Extract]] c
|-
| rowspan="7" | <ol type="a" start="3"><li>Foods rich in antioxidants have high levels of Oxygen Radical Absorption Capacity (ORAC)</li></ol>
| <ol start="7"><li>'''Question''': <span class="cloze">[...]</span> rich in antioxidants have high levels of Oxygen Radical Absorption Capacity (ORAC)<br>'''Answer''': Foods</li></ol>
|-
| <ol start="8"><li>'''Question''': Foods <span class="cloze">(rich/poor)</span> in antioxidants have high levels of Oxygen Radical Absorption Capacity (ORAC)<br>'''Answer''': rich</li></ol>
|-
| <ol start="9"><li>'''Question''': Foods rich in <span class="cloze">[...]</span> have high levels of Oxygen Radical Absorption Capacity (ORAC)<br>'''Answer''': antioxidants</li></ol>
|-
| <ol start="10"><li>'''Question''': Foods rich in antioxidants have <span class="cloze">(high/low)</span> levels of Oxygen Radical Absorption Capacity (ORAC)<br>'''Answer''': high</li></ol>
|-
| <ol start="11"><li>'''Question''': Foods rich in antioxidants have high <span class="cloze">[...]</span> of Oxygen Radical Absorption Capacity (ORAC)<br>'''Answer''': levels</li></ol>
|-
| <ol start="12"><li>'''Question''': Foods rich in antioxidants have high levels of <span class="cloze">[...]</span> (ORAC)<br>'''Answer''': Oxygen Radical Absorption Capacity</li></ol>
|-
| <ol start="13"><li>'''Question''': Foods rich in antioxidants have high levels of Oxygen Radical Absorption Capacity (<span class="cloze">[...](abbreviation)</span>)<br>'''Answer''': ORAC</li></ol>
|-
! colspan="2" | [[Glossary:Extract|Extract]] d
|-
| rowspan="3" | <ol type="a" start="4"><li>Antioxidants destroy free radicals</li></ol>
| <ol start="14"><li>'''Question''': <span class="cloze">[...](food component)</span> destroy free radicals<br>'''Answer''': Antioxidants</li></ol>
|-
| <ol start="15"><li>'''Question''': Antioxidants <span class="cloze">(destroy/create)</span> free radicals<br>'''Answer''': destroy</li></ol>
|-
| <ol start="16"><li>'''Question''': Antioxidants destroy <span class="cloze">[...](harmful compounds)</span><br>'''Answer''': free radicals</li></ol>
|-
! colspan="2" | [[Glossary:Extract|Extract]] e
|-
| rowspan="4" | <ol type="a" start="5"><li>Free radicals are cell-damaging compounds that cause cancer, heart disease and aging</li></ol>
| <ol start="17"><li><span class="cloze">[...]</span> are cell-damaging compounds that cause cancer, heart disease and aging<br>'''Answer''': Free radicals</li></ol>
|-
| <ol start="18"><li>Free radicals are <span class="cloze">[...]</span>-damaging compounds that cause cancer, heart disease and aging<br>'''Answer''': cell</li></ol>
|-
| <ol start="19"><li>Free radicals are cell-<span class="cloze">(building/damaging)</span> compounds that cause cancer, heart disease and aging<br>'''Answer''': damaging</li></ol>
|-
| <ol start="20"><li>Free radicals are cell-damaging compounds that cause <span class="cloze">[...](health problems)</span><br>'''Answer''': cancer, heart disease and aging</li></ol>
|}

Although we have generated 20 [[Glossary:Cloze_deletion|cloze deletions]] from the original 5 [[Glossary:Extract|extracts]], it is important to stress that reviewing this much of the learning material will ultimately cost you less time and the memory effect will be better! Note that [[Glossary:Cloze_deletion|cloze deletions]] meticulously test your knowledge of all important semantic aspects of the learned article.

=== Converting cloze deletions to plain questions ===

After you extract fragments and formulate [[Glossary:Active_recall|active recall]] questions, you should continue to constantly reevaluate the importance of individual pieces of information, their wording, delete less important pieces and move them for later review, etc. Examples of reformulated [[Glossary:Cloze_deletion|cloze deletions]] can be found below. Note that [[Glossary:Cloze_deletion|Clozes]] 4-6 and [[Glossary:Cloze_deletion|Cloze]] 20 have been split further to eliminate set enumeration (it is easier to independently associate cancer or aging with free radicals than to list all health problems caused by them)

{| class="wikitable"
!Original [[Glossary:Cloze_deletion|cloze deletion]]
!Reformulated [[Glossary:Active_recall|active recall]] [[Glossary:Item|item]]
|-
! colspan="2" | [[Glossary:Cloze_deletion|Clozes]] 4-6
|-
|<ol start="4"><li>'''Question''': Examples of <span class="cloze">[...]</span> rich in antioxidants: blueberries, strawberries, spinach and broccoli<br>'''Answer''': foods</li><li>'''Question''': Examples of foods rich in <span class="cloze">[...]</span>: blueberries, strawberries, spinach and broccoli<br>'''Answer''': antioxidants</li><li>'''Question''': Examples of foods rich in antioxidants: <span class="cloze">[...]</span><br>'''Answer''': blueberries, strawberries, spinach and broccoli</li></ol>
|<ul><li>'''Question''': Are blueberries rich in antioxidants?<br>'''Answer''': yes</li><li>'''Question''': Are strawberries rich in antioxidants?<br>'''Answer''': yes</li><li>'''Question''': Is spinach rich in antioxidants?<br>'''Answer''': yes</li><li>'''Question''': Is broccoli rich in antioxidants?<br>'''Answer''': yes</li></ul>
|-
! colspan="2" | [[Glossary:Cloze_deletion|Cloze]] 12
|-
|<ol start="12"><li>'''Question''': Foods rich in antioxidants have high levels of <span class="cloze">[...]</span> (ORAC)<br>'''Answer''': Oxygen Radical Absorption Capacity</li></ol>
|<ul><li>'''Question''': What does ORAC stand for?<br>'''Answer''': Oxygen Radical Absorption Capacity</li></ul>
|-
! colspan="2" | [[Glossary:Cloze_deletion|Cloze]] 17
|-
|<ol start="17"><li>'''Question''': <span class="cloze">[...]</span> are cell-damaging compounds that cause cancer, heart disease and aging<br>'''Answer''': Free radicals</li></ol>
|<ul><li>'''Question''': What is the name of cell-damaging compounds that cause cancer? (cleaned off with antioxidants)<br>'''Answer''': Free radicals</li><li>'''Question''': What are free radicals?<br>'''Answer''': cell-damaging compounds (causing cancer, heart disease and aging)</li></ul>
|-
! colspan="2" | [[Glossary:Cloze_deletion|Cloze]] 20
|-
|<ol start="20"><li>'''Question''': Free radicals are cell-damaging compounds that cause <span class="cloze">[...](health problems)</span><br>'''Answer''': cancer, heart disease and aging</li></ol>
|<ul><li>'''Question''': Do free radicals cause cancer?<br>'''Answer''': yes</li><li>'''Question''': Do free radicals contribute to heart disease?<br>'''Answer''': yes</li><li>'''Question''': Do free radicals cause aging?<br>'''Answer''': yes</li></ul>
|}

== Example: Rewording texts ==

Wordy articles may require some rewording of sentences before [[Glossary:Cloze_deletion|cloze deletions]] can be generated.

For example, the following texts puzzled a user as it appeared to be hard to process incrementally:
<blockquote>
<p>''In 1892 the Russian botanist Dimitri Iwanowski showed that the sap from tobacco plants infected with mosaic disease, even after being passed through a porcelain filter known to retain all bacteria, contained an agent that could infect other tobacco plants.''</p>
<p>''In 1900 a similarly filterable agent was reported for foot-and-mouth disease of cattle.''</p>
</blockquote>

Before you begin learning, you can save lots of time by looking for articles that are properly structured and written in a concise language that will help you save lots of time. For example, [http://en.wikipedia.org Wikipedia] is an excellent source. As it is edited by many people in an incremental manner, it is highly context-independent. In comparison, [http://britannica.com Britannica] is wordy, full of pronouns, definite clauses, and various context references.

Where [http://britannica.com Britannica] might say (fictitious example): &quot;''Over the next five years, he struggled to obtain a patent for his invention''&quot;, [http://en.wikipedia.org Wikipedia] might say explicitly &quot;''In the years 1883-1889, Edison struggled to obtain a patent for a filament of carbon of high resistance''&quot;. This context-independent style can save you hours of parsing and re-editing.

In the presented example, the first sentence is causing trouble because the author tried to tell you far more than you might wish to process in one go.

One strategy is to start with monster [[Glossary:Cloze_deletion|clozes]], and simplify them incrementally while learning. However, you could save lots of time with another strategy, in which you split the sentences into more manageable portions. Unfortunately, in this case, some editing will be necessary in the beginning. You will also need to carefully parse the meaning of the passage. You could, for example, separate ''who'' and ''what'' components of the sentence

<blockquote>
<p>'''who''': In 1892 the Russian botanist Dimitri Iwanowski showed that the sap from tobacco plants infected with mosaic disease contained an infectious agent smaller than bacteria.</p>
<p>'''what''': In 1892, Iwanowski showed that the sap from tobacco infected with mosaic disease even after being passed through a porcelain filter known to retain all bacteria, contained an agent that could infect other tobacco plants.</p>
</blockquote>

From those two mini-[[Glossary:Topic|topics]], you can generate several [[Glossary:Cloze_deletion|clozes]] that will cover the essence of the passage:

<blockquote>
'''Question''': In <span class="cloze">[...](year)</span> the Russian botanist Dimitri Iwanowski showed that the sap from tobacco plants infected with mosaic disease contained an infectious agent smaller than bacteria<br>
'''Answer''': 1892
</blockquote>

<blockquote>
'''Question''': In 1892 the <span class="cloze">[...](nationality)</span> botanist Dimitri Iwanowski showed that the sap from tobacco plants infected with mosaic disease contained an infectious agent smaller than bacteria<br>
'''Answer''': Russian
</blockquote>

<blockquote>
'''Question''': In 1892 the Russian <span class="cloze">[...](specialty)</span> Dimitri Iwanowski showed that the sap from tobacco plants infected with mosaic disease contained an infectious agent smaller than
bacteria<br>
'''Answer''': botanist
</blockquote>

<blockquote>
'''Question''': In 1892 the Russian botanist <span class="cloze">[...](name)</span> showed that the sap from tobacco plants infected with mosaic disease contained an infectious agent smaller than bacteria<br>
'''Answer''': Dimitri Iwanowski
</blockquote>

<blockquote>
'''Question''': In 1892 the Russian botanist Dimitri Iwanowski showed that <span class="cloze">[...](what?)</span> from tobacco plants infected with mosaic disease contained an infectious agent smaller than bacteria<br>
'''Answer''': sap</blockquote>

<blockquote>
'''Question''': In 1892 the Russian botanist Dimitri Iwanowski showed that the sap from <span class="cloze">[...](type)</span> plants infected with mosaic disease contained an infectious agent smaller than bacteria<br>
'''Answer''': tobacco
</blockquote>

<blockquote>
'''Question''': In 1892 the Russian botanist Dimitri Iwanowski showed that the sap from tobacco plants infected with <span class="cloze">[...](disease)</span> contained an infectious agent smaller than bacteria<br>
'''Answer''': mosaic disease
</blockquote>

<blockquote>
'''Question''': In 1892 the Russian botanist Dimitri Iwanowski showed that the sap from tobacco plants infected with mosaic disease contained <span class="cloze">[...]</span> smaller than bacteria<br>
'''Answer''': an infectious agent
</blockquote>

<blockquote>
'''Question''': In 1892 the Russian botanist Dimitri Iwanowski showed that the sap from tobacco plants infected with mosaic disease contained an infectious agent smaller than <span class="cloze">[...]</span><br>
'''Answer''': bacteria
</blockquote>

<blockquote>
'''Question''': In 1892 the Russian botanist Dimitri Iwanowski showed that the sap from tobacco plants infected with mosaic disease contained an infectious agent <span class="cloze">[...]</span> than bacteria<br>
'''Answer''': smaller
</blockquote>

<blockquote>
'''Question''': In 1892, Iwanowski showed that the sap from tobacco infected with mosaic disease even after being passed through a <span class="cloze">[...](type)</span> filter, contained an agent that could infect other tobacco plants<br>
'''Answer''': porcelain
</blockquote>

<blockquote>
'''Question''': In 1892, Iwanowski showed that the sap from tobacco infected with mosaic disease even after being passed through a porcelain filter known to <span class="cloze">[...](property)</span>, contained an agent that could infect other tobacco plants<br>
'''Answer''': retain all bacteria
</blockquote>

<blockquote>
'''Question''': In 1892, Iwanowski showed that the sap from tobacco infected with mosaic disease even after being passed through a porcelain filter, contained an agent that <span class="cloze">[...](property)</span><br>
'''Answer''': could infect other tobacco plants
</blockquote>

The above questions are only a rough beginning. Only during learning will you be able to identify holes in these [[Glossary:Item|items]]. You will see where they cause trouble, why they may be hard to remember or what questions are imprecise or confusing. You will fix those deficiencies incrementally while learning.

== Example: Conglomerating information ==

A frequent sin committed by new users of SuperMemo is to create &quot;monster items&quot; that include many bits of information. Those [[Glossary:Item|items]] should tackled by treating all bits individually.

=== Example of a monster item ===

<blockquote>
'''Question''': a rod-and-tube element temperature sensor consists of:<br>
'''Answer''': a high expansion metal tube containing a low expansion rod. The rod&amp; tube are attached on one end. The tube changes length with changes in temperature, causing the free end of the rod to move
</blockquote>

=== Suggestions ===

This is a typical case of combining a number of [[Glossary:Item|items]] in one with a detriment to the ability to recall the combined [[Glossary:Item|item]]. The suggestion here is to split the [[Glossary:Item|item]] into a number of simpler [[Glossary:Item|items]] that reproduce the same information in student's memory:

<blockquote>
'''Question''': What are the two parts of a rod-and-tube temperature sensor?<br>
'''Answer''': rod and tube
</blockquote>

<blockquote>
'''Question''': What is the expandability of the '''tube''' in rod-and-tube sensor?<br>
'''Answer''': high
</blockquote>

<blockquote>
'''Question''': What is the expandability of the '''rod''' in rod-and-tube sensor?<br>
'''Answer''': low
</blockquote>

<blockquote>
'''Question''': How is temperature indicated in the rod-and-tube sensor?<br>
'''Answer''': tube moves relative to the rod
</blockquote>

<blockquote>
'''Question''': Where are rod and tube connected?<br>
'''Answer''': On one end
</blockquote>

<blockquote>
etc. etc.
</blockquote>

== Example: Evolution of knowledge ==

Changes to individual pieces of knowledge will take place in steps upon successive reviews. Here are exemplary steps that show a complete evolution of a single article into a finished [[Glossary:Item|item]] based on [[Glossary:Active_recall|active recall]]:

<ol>
<li>Imagine that you find an article on the net, e.g. ''The criticism of global capitalism'', and you decide to read it and remember it for ever</li>
<li>You import the article to SuperMemo</li>
<li>You read the article (e.g. once its turn comes up in [[Glossary:Incremental_reading|incremental reading]])</li>
<li>While reading, you extract most important paragraphs. One of these, let us say, refers to ''[http://en.wikipedia.org/wiki/Kuznets_curve Kuznets hypothesis]''</li>
<li>The extracted paragraphs will live separate lives in SuperMemo and will be scheduled for separate review, i.e. independent of the review of the parent article. The extracted paragraphs in the parent article will be marked as <span class="extract">processed</span>. Once all paragraphs in the parent article are processed, you will terminate the review of the parent article and keep on reviewing only its components (e.g. selected paragraphs)</li>
<li>Upon the first review, usually after a few days, you read the extracted paragraph again and analyze it as to how it should be processed further. You may decide to [[postpone]] it, remove it from the learning process, shorten it or extract the most important sentences that you want to remember</li>
<li>If you decide to extract a single statement in reference to ''Kuznets hypothesis'' it will again be marked as <span class="extract">processed</span> in the original [[Glossary:Extract|extract]] and will assume a separate review cycle in SuperMemo</li>
<li>Upon the first review of the extracted sentence, you make further decisions as to its further life in SuperMemo. Let us say, this is the wording of the Kuznets sentence:
<blockquote>Acc to Kuznets hypothesis, growth (from the low income levels associated with predominantly agrarian societies) would first lead to an increase, and then to a decrease in income inequality</blockquote>
</li>
<li>In order to capture the essence, you would probably decide to shorten the above sentence to the following form:
<blockquote>Acc to Kuznets hypothesis, growth would first lead to an increase, and then to a decrease in income inequality</blockquote>
</li>
<li>At the same time, other parts of the same parent article might establish a memory trace that would say that ''Kuznets hypothesis'' has been based on relatively weak empirical data. Moreover, recent research clearly indicates that the hypothesis is false (growth actually seems to equally benefit both the poor and the rich). You could then enhance the [[Glossary:Extract|extract]] with words ''controversial'' or even ''recently falsified''. For example:
<blockquote>Recently falsified Kuznets hypothesis claimed that growth would first lead to an increase, and then to a decrease in income inequality</blockquote>
</li>
<li>Upon the next review of the same sentence, you may decide to convert it into a number of [[Glossary:Cloze_deletion|cloze deletions]]. This conversion will be incremental, i.e. you may decide to first create a [[Glossary:Cloze_deletion|cloze deletion]] asking about the name of the controversial hypothesis and only later ask about its actual meaning (the meaning is relatively easier to remember and shall survive longer in your memory without [[Glossary:Active_recall|active recall]]). Your [[Glossary:Cloze_deletion|cloze deletion]] could then look like this:
<blockquote>
'''Question:''' Recently falsified <span class="cloze">[...](name)</span> hypothesis claimed that growth would first lead to an increase, and then to a decrease in income inequality<br>
'''Answer''': Kuznets
</blockquote>
This [[Glossary:Cloze_deletion|cloze deletion]] would again assume a separate life from the original sentence in which the keyword ''Kuznets'' will again be marked as <span class="clozed">processed</span>. This is the original Kuznets sentence with one keyword marked as processed:
<blockquote>Recently falsified <span class="clozed">Kuznets</span> hypothesis claimed that growth would first lead to an increase, and then to a decrease in income inequality</blockquote>
</li>
<li>The same sentence will generate a few separate [[Glossary:Cloze_deletion|cloze deletions]] that will be processed independently. Upon the first review of the [[Glossary:Cloze_deletion|cloze deletion]] created in the previous point, you may decide to simplify it in accordance with the [http://super-memory.com/articles/20rules.htm rules of formulating knowledge in learning]:
<blockquote>
'''Question''': Recently falsified <span class="cloze">[...](name)</span> hypothesis claimed that growth would first lead to an increase in income inequality<br>
'''Answer''': Kuznets
</blockquote>
</li>
<li>Upon the next review, you can, but you do not have to, convert the [[Glossary:Cloze_deletion|cloze deletion]] into a standard question-answer [[Glossary:Item|item]]:
<blockquote>
'''Question''': What is the name of the hypothesis that falsely claims that income inequality initially increases with growth?<br>
'''Answer''': Kuznets hypothesis
</blockquote>
</li>
<li>The above question-answer pair is probably as simple as it can only be. Certainly, it is simple enough to be relatively easy to remember. This [[Glossary:Item|item]] will be repeated in [[Glossary:Interval|intervals]] determined by SuperMemo. You can decide how well you want to remember it. By default, it will be remembered with 95% probability of recall and require 5-15 repetitions in lifetime. The establishment of durable memory traces in your memory, completes the life cycle of this particular piece of knowledge. The only thing that remains is the memory-sustaining review in [[Glossary:Interval|intervals]] ranging from months to years (as determined by SuperMemo)</li>
<li>Once you convert all important keywords from the ''Kuznets hypothesis'' into separate [[Glossary:Cloze_deletion|cloze deletions]], you will remove the parenting paragraph from the review process. You will no longer [[Glossary:Passive_review|passively review]] the original declarative hypothesis. You will continue repeating individual [[Glossary:Cloze_deletion|clozes]] and that will ensure your perfect recall of the hypothesis for as long as you deem necessary</li>
</ol>

== Example: Building comprehension incrementally ==

It is not unusual to generate a [[Glossary:Cloze_deletion|cloze]] that will keep causing problems. It is not the subject in question, but the complexity of the sentence that seems to send the brain into a panic mode. You may keep re-reading bad [[Glossary:Cloze_deletion|clozes]] and get the impression that at each re-reading the understanding decreases (instead of growing).

Here is an example of a bad [[Glossary:Cloze_deletion|cloze]] taken from a real learning process. It caused 5 [[Glossary:Lapse|lapses]] in succession, and needed special treatment to make it palatable and useful.

<blockquote>
'''Question''': Companies or other groups are issued emission permits and are required to hold an equivalent number of <span class="cloze">[...]</span> (or ''credits'') which represent the right to emit a specific
amount<br>
'''Answer''': allowances
</blockquote>

With some experience, you will quickly notice that the problem resides in the poor quality of the chosen text. Your first red flag should come with &quot;''companies or other groups''&quot;. Unless you are an insider to the subject, you will instantly wonder what &quot;''other groups''&quot; means. If you look at the essence of your question, you will notice that &quot;''other groups''&quot; does not add to the core of the question and can easily be skipped. After some analysis, you can make an effort to formulate a straightforward question, however, it is easier and more cost-effective to take an incremental approach (as long as you keep understanding the question, which may not be the case with you).

As soon as you notice that &quot;''other groups''&quot; are out of place, you can simplify the [[Glossary:Cloze_deletion|cloze]]:
<blockquote>
'''Question''': Companies are issued emission permits and are required to hold an equivalent number of <span class="cloze">[...]</span> (or ''credits'') which represent the right to emit a specific amount<br>
'''Answer''': allowances
</blockquote>

You may also notice that ''allowances'' and ''credits'' are synonymous (acc. to that particular question), and you may not need to tax your memory with both terms. You can therefore make the question easier:

<blockquote>
'''Question''': Companies are issued emission permits and are required to hold an equivalent number of <span class="cloze">[...]</span> which represent the right to emit a specific amount<br>
'''Answer''': allowances/credits
</blockquote>

During the next repetition you may wish to simplify the question further. You can now start feeling confident that the question will stick to your memory. If that makes you feel better, you will have made a big step towards better recall:</p>

<blockquote>
'''Question''': Companies hold a number of <span class="cloze">[...]</span> which represent the right to emit a specific amount<br>
'''Answer''': allowances/credits (emission permits)
</blockquote>

Some time in the future, you might take yet another incremental step:

<blockquote>
'''Question''': <span class="cloze">[...]</span> represent the right to emit a specific amount (by companies)<br>
'''Answer''': allowances/credits (emission permits)
</blockquote>

Finally, you may realize that you question is actually a definition of a term (''credit/allowance''). As such, it does not even need the [[Glossary:Cloze_deletion|cloze]]. You may also notice that the definition misses an important context point. The original question spoke of carbon emissions (which is probably indicated by the [http://en.wikipedia.org Wikipedia] references that provide the context of the question).

<blockquote>
'''Question''': (What is the name of) the right to emit a specific amount of CO<sub>2</sub> (by companies)?<br>
'''Answer''': (carbon) permits/allowances/credits
</blockquote>

With the above definition, learning should finally be easy and fun. Remember that it is a frequently used standard to separate synonyms or equivalent answers with a slash (&quot;/&quot;). This means that any answer will do ''permits'', ''allowances'', or ''credits''. You do not need to list them all and you do not need to remember about semi-obvious ''carbon'' prefix.

Interestingly, it can easily be found that the text comes from May 7, 2010 version of [http://en.wikipedia.org Wikipedia]. It was red-flagged with &quot;clarification needed&quot;, which instantly tells you that you are not the only one who got an issue with the paragraph. In such cases blame the author and search for better texts. In this particular example, the [http://en.wikipedia.org Wikipedia] text was improved just 2 weeks later, and the current version of [http://en.wikipedia.org Wikipedia] holds a better version (Aug 2013).

== Example: Unsuitable texts ==

Not all texts are suitable or easy to process with [[Glossary:Incremental_reading|incremental reading]]. You will not want to process a literary novel with [[Glossary:Incremental_reading|incremental reading]]. You may still prefer to read it on paper in a bathtub. Examples of texts that are difficult to process are: flowery materials, materials rich in explanations and metaphors, programming code, case studies, mathematical derivations, experimental research documentation, etc. [[Glossary:Incremental_reading|Incremental reading]] is easiest for encyclopedic materials. Materials that are not suitable will often include a valuable message; however, you will often be better off by phrasing it on your own and processing your summary with [[Glossary:Incremental_reading|incremental reading]]. For example, you would not want to memorize the Linux source code. However, you could find some specific facts or regularities in the code, describe them shortly and then learn the description incrementally (perhaps with snippet code illustrations).

=== Example 1: Texts that are too general ===

Here is an example paragraph that caused learning problem to a user of SuperMemo:

<blockquote>
''A good enterprise architect should enable the right balance between the needs of the organization for an integrated IT strategy, permitting the closest possible synergy across the extended enterprise, and allowing individual business units to innovate safely in their pursuit of competitive advantage.''
</blockquote>

The user wrote:

<blockquote>
''I have no problem understanding this phrase (synergy across business units must be balanced against freedom to innovate within business units), and I constructed two [[Glossary:Cloze_deletion|cloze deletions]] from it for either side of the balance, and when presented with either question I can fill in the blanks. Yet, I was in a discussion recently defending a synergy position, not realizing that it might jeopardize innovation, not even realizing I had this SuperMemo question pointing it out to me. In other words, even though I can answer the question in SuperMemo, it is not something that stuck in my memory, i.e. the synergy is not associated with freedom-to-innovate and vice versa (maybe it will after this mail). So, I suspect that I am only able to answer the question based on recognition of the question or some such, and not on recognition of the association. I figure that a better way to associate the two would be to ask something like &quot;what must be balanced against each other&quot; but this question would be so general in nature that it would create serious interference with other questions that deal with other aspects of enterprise architecture that need to be balanced against each other. Or I would have to make it more specific, again risking to give away the answer in the question, which would also not cause the association to form.''
</blockquote>

Can you identify a problem here?

In similar cases, you need to pause to ponder what kind of questions you want to be able to answer having read the passage. If the questions are too general or too obvious, you just need to trust your own intelligence and creativity to be able to answer them on the basis of your experience and more specific questions in the given area of knowledge.

In this case, keywords such as &quot;synergy&quot; or &quot;innovate&quot; might provide a hazy way to capture the meaning of the passage. However, very general texts are not suitable for treatment with [[Glossary:Cloze_deletion|cloze deletions]]. You may waste unnecessary time on re-reading the entire passage at question time, or waste time on simplifying the passage to capture the essence. In school jargon, you might call similar passages &quot;waffle&quot;. They may carry an important message, they may help the flow of text, they might be explanatory, but they do not yield material suitable for memorization. In the extreme case you can juxtapose Wikipedia-like &quot;IT = Information Technology&quot; with obvious &quot;waffle&quot; that cannot be clozed: ''We should be nice for other people''.

In all sorts of exams, you will always need to tackle lots of &quot;waffle&quot;. You will also meet teachers who demand fluent &quot;waffle&quot; performance. However, this is not the type of knowledge that will make you a better expert or a better person. If you meet &quot;waffle&quot;, pause to think if there are questions that truly flow from the text, or if the text is too general to be handled with SuperMemo. In your case, you might do better by perhaps adding some meatier passages on enterprise synergy or constraints on innovation or... Actually, you are the best person to find supplementary material that will help you better understand the underlying issues.

If waffle bothers you, try to find a [http://en.wikipedia.org Wikipedia] equivalent. Due to the nature of crowdsourcing, [http://en.wikipedia.org Wikipedia] lends itself perfectly to incremental processing. Once you get the hang and feel the benefit, you will quickly learn to spot text and passages that are less suitable and provide less benefit when processed with [[Glossary:Incremental_reading|incremental reading]].

=== Example 2: Unsuitable text from super-memory.com ===

Here is another example that comes from [http://super-memory.com super-memory.com]. The text itself is not bad, but it resorts to metaphors that should serve as explanations, not as learning material suitable for generating [[Glossary:Cloze_deletion|cloze deletions]].

<blockquote>
'''''Intelligence as processing power''': the raw nimbleness and agility of the human mind. When you see a smart student quickly learn new things, think logically, solve puzzles and show uncanny wit, you may say: This guy is really intelligent! See how fast his brain reacts! The student has a fast processor installed and his RAM has a lightning access time. He may though still need a couple of years to &quot;build&quot; good software through years of study. IQ tests attempt to measure this sort of intelligence in abstraction of knowledge. The difficulty of improving processing power by training comes for similar reasons as the fact that programming cannot speed up the processor''
</blockquote>

The above text is metaphorical. It reiterates the same message a few times using different words in an attempt to find a metaphor that will strike a cord with the reader. Consequently, it is enough you extract only the core message. For example:

<blockquote>
''Intelligence as <span class="clozed">processing power</span>: <span class="clozed">IQ</span> tests attempt to measure this sort of intelligence in abstraction of <span class="clozed">knowledge</span>''
</blockquote>

You could also add:

<blockquote>
''Intelligence as <span class="clozed">processing power</span>: The difficulty of improving processing power by training is similar to the fact that <span class="clozed">programming</span> cannot speed up <span class="clozed">the processor</span>''
</blockquote>

Once you learn the above 6 [[Glossary:Cloze_deletion|cloze deletions]] (marked by 6 clozed keywords), you will most likely be able to recall that it should be very difficult to train for an improved score in an ideally designed IQ test.

= What do people say about incremental reading? =

Incremental learning has its followers and its hardline critics. The following two sections present both sides of the story.

== Incremental reading in the eyes of its users ==

Here are a few excerpts from blogs that speak of [[Glossary:Incremental_reading|incremental reading]].

=== Dealing with information overflow ===

At [http://bobo23.net bobo23.net] you learn the following:

<blockquote>
<div class="bs-callout bs-callout-comment">
<p>''You want to read your favorite blogs, you get e-mail newsletters every day, you have websites you check regularly, newsgroups, mailing lists, forums, interesting Wikipedia articles – a lot of digital input you want to keep up with. But unless you make reading on the computer your full time job – you can’t. So how to select the really important stuff out of it?''</p>
<p>''[...]''</p>
<p>''This is where the concept of Incremental Reading comes into place. Sometimes you only want to read about a specific topic, sometimes you just want to read a bit of a complicated article or just read about anything randomly to build new connections / enhance creativity. You can all do that with incremental reading and do not have to worry to miss something. Sooner or later (you can influence that) it will appear in your incremental reading process.''</p>
<p>''[...]''</p>
<p>''You collect all the information you want to process and store them in one place. Then you review all the articles (or any other kind of information) randomly or by category. You can highlight important parts, set a reading point (bookmark), extract fragments and generate Question-Answer items for later repetitions.''</p>
</div>
</blockquote>

To read the entire entry, visit [http://bobo23.net bobo23.net]: [http://www.bobo23.net/2011/02/04/incremental-reading/ Incremental Reading - Dealing with Information Overflow]

=== Len Budney on incremental reading ===

Len Budney wrote the following:
<blockquote>
<div class="bs-callout bs-callout-comment">
<p>''Incremental reading is another amazing innovation of Supermemo. I haven't seen or heard about such an idea anywhere else!''</p>
<p>''[...]''</p>
<p>''You are gradually digesting the article into smaller and smaller pieces.''</p>
<p>''When those pieces have become truly bite-size, you can convert them to real flash-cards. With a single click you can create &quot;fill in the blank&quot; type flash-cards, or you can create a normal question-and-answer flash card. When the nugget of information is converted to a flash card, you can &quot;Forget&quot; that piece of text: you've digested it all the way down into facts that you won't be forgetting!''</p>
<p>'''''I use the incremental reading feature for reading long, complicated documents, and it's world's better than any other method I've tried'''. (And with a PhD in Mathematics, I've certainly studied my share of documents!) It forces you to read the important bits very carefully--but still saves you time by helping you to ignore the less important bits. I can't recommend it highly enough. In fact, I'd give my left arm to get a Palm application which supports incremental reading!''</p>
</div>
</blockquote>

Visit Len Budney's blog to read the article: [http://jeenyus.net/supermemo/incremental.html Incremental reading]

=== How to supercharge your language learning ===

In his blog tmwbuckley writes:

<blockquote>
<div class="bs-callout bs-callout-comment">
''Incremental reading is a fairly misunderstood feature of SuperMemo. Incremental reading entails importing your reading material into a flashcard. This can be a huge Wikipedia article for instance, or an entire eBook. Whilst reading, if you encounter a word you’re unfamiliar with, you can highlight the sentence or paragraph it sits in and make it into an entirely new flashcard automatically, ready for you to review later on. By highlighting, pressing Alt+X to extract the segment – you will see the section become blue, meaning this part has been extracted to another flashcard. [...] Incremental refers to reading the content in small chunks over time, until you’ve read all the material and extracted the information you want from it. You can mark where you’ve reached in the reading material up to using the ‘set the read point’ tool on the Read toolbar, so next time you see the article/chapter/whatever, SuperMemo will display it visibly instead of you scrolling down. [...] Once you’ve finished reading the flashcard, and extracted all of the juicy goodness from it, you can delete it using Shift-Ctrl-Enter. This command is different to pure delete, because it will retain all of the flashcards you created from it, whereas delete will cause you to lose these. Which would be a pain in the derriere. SuperMemo, using it’s super-duper algorithm, will schedule the review of these items for you. [...] Incremental reading represents a powerful chimera of extensive and intensive reading. And this can be used with anything – news stories, blog posts, Wikipedia articles.''
</div>
</blockquote>

For the entire article, see: [http://unmilliondemots.wordpress.com/2012/09/26/incremental-reading/ Incremental Reading – How to supercharge your language learning]

=== Taking note: Incremental Reading ===

In a blog devoted to note taking, MK is skeptical when writing about [[Glossary:Incremental_reading|incremental reading]].

MK introduces the process with these words:
<blockquote>
<div class="bs-callout bs-callout-comment">
''The student extracts the most important fragments of individual articles for further review. Extracted fragments are then converted into questions and answers. These in turn become subject to systematic review and repetition that maximizes the long-term recall of the processed texts. The review process is handled by the proven repetition spacing algorithm known as the SuperMemo method.''
</div>
</blockquote>

MK then voices the reservations:
<blockquote>
<div class="bs-callout bs-callout-comment">
''The basic idea is that you &quot;break down&quot; a larger document into smaller chunks in order to learn them by heart. Nothing wrong with that, but this method will '''not''' allow you to read thousands of articles '''at the same time'''. Nor will it necessarily lead you to understand the big picture.''
</div>
</blockquote>

Here is the blog entry from MK: [http://takingnotenow.blogspot.com/2010/07/incremental-reading.html Incremental reading]

==== Big picture ====

The issue of the big picture has been raised over and over by observers who are not intimate to the incremental learning process. See: [[#Big picture in incremental learning|Big picture in incremental reading]]

==== Reading thousands of articles at the same time ====

As for the phrase &quot;''reading thousands of articles at the same time''&quot;, we will continue using that catchy phrase. It should be pretty obvious that we do not mean the same &quot;''moment of time''&quot; but the same &quot;''period of time''&quot;. We do not envisage a user with a thousand of monitors, nor a user photoreading thousands of articles flashing in front of his eyes in a second. We, naturally, mean having thousands of articles in the learning process, which, in heavily overloaded [[Glossary:Collection|collection]], may mean that an article will never actually gets read due to scoring too low in [[Glossary:Priority|priority]]. For more see: [[#Can you really read thousands of articles at the same time?|Incremental reading marketed dishonestly]]

=== The way of the ronin: About incremental reading ===

''The way of the ronin'' blog says:
<blockquote>
<div class="bs-callout bs-callout-comment">
<p>''What have I been doing instead of updating this blog? Mainly modifying and improving the system which I use for studying and starting a new project based in the '''incremental reading''' technique.''</p>
<p>''[...]''</p>
<p>''What I want to speak about it’s my user’s experience after using Anki for a whole year in a regular way and SuperMemo for half year (still doing it) also in a regular way.''</p>
<p>''[...]''</p>
<p>''SuperMemo has a very complex interface which most of the times makes potential users think twice before buying it.''</p>
<p>''[...]''</p>
<p>''After one week or two of using it I got used to it’s interface. In fact, I realized It was quite intuitive.''</p>
<p>''[...]''</p>
<p>'''''What about schools then? Don’t they use the same methods over and over again for years and decades? If all this was true wouldn’t it mean that they should adapt their system to each of their students -which of course they don’t-?'''''</p>
</div>
</blockquote>

Read more in [http://thewayoftheronin.wordpress.com/?utm_content=buffer614da&amp;utm%0D_source=buffer&amp;utm_medium=twitter&amp;utm_campaign=Buffer Anki vs. SuperMemo]

=== How to incrementally read anything ===

LittleFish in his blog writes: 

<blockquote>
<div class="bs-callout bs-callout-comment">
''When I first started using Supermemo, it appeared that Incremental Reading was a valuable feature, but I never really used it. Once I began experimenting with Incremental Reading (Successfully) and adopting the Incremental Reading mindset when I look at learning material, I am now fully convinced of its superiority for nearly all of my intellectual needs. It is light years ahead of its time, and like the core concepts that power Supermemo, eventually it will likely be utilized on almost every level of the educational system (However many years that takes).''

''[...]''
</div>
</blockquote>

Visit the blog: [http://supermemoadventures.blogspot.com/2010/04/how-to-incrementally-read-anything.html Supermemo Adventures: How to Incrementally Read Anything]

=== Luis Gustavo Neves da Silva on memorization ===

Luis Gustavo writes about cramming and wisdom:

<blockquote>
<div class="bs-callout bs-callout-comment">
<p>'''When I learned how to use SuperMemo, one of my first reactions was to show it to everybody around. I was very enthusiastic about the program and its underlying principles. However, I was very disappointed to find out how many people would say: &quot;This is a cramming tool&quot;, &quot;This is just memorizing&quot;, &quot;This has nothing to do with reasoning or intelligence&quot;.''</p>
<p>''[...]''</p>
<p>''I believe that in our culture we have even developed a sort of fear of memory as a human weakness that would allow us to be manipulated, and even dehumanized. Just take a look at our pop culture. How many movie protagonists' memories are being erased, changed or manipulated.''</p>
<p>''[...]''</p>
<p>''Teachers seems to see no limit in piling up new material in front of the kids and if someone's memory fails ... its all his or her own fault or negligence.''</p>
<p>''[...]''</p>
<p>''When asked to explain the reason for believing that memorizing is harmful, the supporters of &quot;reasoning&quot; would come up with a confusing and conflicting mix of words such as memorizing, data, information, knowledge and even wisdom. Looking up these words in a dictionary, you will find a lot of circular references, each word being related to or defined by the use of the others.''</p>
</div>
</blockquote>

Read more: [http://www.super-memory.com/articles/users/memorizing.htm It's more than just memorizing]

=== Incremental reading diagram ===

[http://www.banaban.net banaban.net] writes:

<blockquote>
<div class="bs-callout bs-callout-comment">
''I have produced this [http://www.banaban.net/supermemo/incremental-reading.html simple diagram of the incremental reading process]. The explanation on the SuperMemo web was too long and too scientific for me, I needed something simple &amp; handy. Hopefully it will also help you to jumpstart into incremental reading with SuperMemo. It only includes the core functionality, for all the bells and whistles, more explanation''
</div>
</blockquote>

Note that the diagram uses old icons, old terminology and old keywords.

== Criticism of incremental reading ==

=== Incremental reading is useless ===

Critics of [[Glossary:Incremental_reading|incremental reading]] are not too numerous or vocal. The credit probably goes to the fact that this technology is relatively new and not too popular beyond SuperMemo insider circles. Nearly all criticism comes from those who never used [[Glossary:Incremental_reading|incremental reading]] themselves. As such, it hardly ever addresses the substance of the technology. Some of the criticism seem to address the technology in general (e.g. the Internet, the web, computer software, etc).

For example, Emil K. wrote the following criticism of incremental reading:

<blockquote>
<div class="bs-callout bs-callout-comment">
# ''Information on the internet is often unreliable''
# ''It will take a lot of time to find a 100% reliable information on the web''
# ''The slow internet connection augments the amount of squandered time dramatically''
# ''Software maintenance wastes more time''
# ''It can be used only when there is a computer available - mostly at home.''

''I believe that my method is better than what you propose:''

# ''Read a book that has been proven excellent by time and by ratings of others who have spent THEIR time exploring the book.''
# ''Instead of some software and internet, use a pen and a notebook to scan the book for important information and copy down for later examination/memorization.''

''Also, copy down the source and the date so that when creativity comes, the sources are reliable - it will be a disaster to waste plethora of time on a research only to find out in the end that the sources did not have any statistical significance. Magazines like Nature and Science already publish the worthiest of all research articles. Books by Feynman, Borh, Dirac, and Einstein are already the best in the field. Thus, my letter is a cogent polemic that refutes SuperMemo.''
</div>
</blockquote>

=== Observer's Opinion: Why I would not use incremental learning? ===

Here is an exemplary opinion of an observer who is focused on problem solving and short-term goals. For this observer, lifelong learning is understood as "''I learn what I need right now''". Naturally, this approach is missing an important factor: ''learning in anticipation''.

<blockquote>
<div class="bs-callout bs-callout-comment">
'''Massive learning'''

''I do not care about massive learning. I solve problems. I learn as much as I need to solve the problem. After the problem is solved, I go to another. I do not care about long-term knowledge. I am technical. Knowledge in my field goes out-of-date in 2-3 years.''

'''Lifetime memories'''

''I do not care about lifetime memories. I learn for short-term projects. My memory is good. I am a good problem solver.''

'''High retention'''

''As I learn for short-term, my retention is always high (within the scope of my needs).''

'''Comprehension'''

''I understand my texts well enough. If something is unclear, I look for better texts.''

'''Uniform progress'''

''I do not care about learning &quot;nothing about everything&quot;. I learn stuff that I need here and now. Why waste time and memory space on memorizing British monarchs. That's good for kids at school, not for a real problem solver.''

'''Creativity'''

''I am creative enough. Thank you. I do not need to mix up domains of knowledge in hope of some elusive Nobel-winning discovery. I solve problems. I keep all my knowledge in my head. I put the pieces together when I need them. I do not have problems with creativity.''

'''Consistency'''

''I instantly recognize contradictions when reading about the materials of interest. I work for short-term projects, keep all my knowledge in memory, and instantly know if something fits or does not fit the model. This is a non-issue!''

'''Stresslessness'''

''I love my job. I do not consider it stressful. Just the opposite. I think struggling with complexities of SuperMemo might give me a heart attack!''

'''Attention'''

''I am passionate about my job. I do not force myself to learn boring stuff. My attention is excellent.''

'''Consolidation'''

''I do not work on projects that drag for years. I do not need to consolidate. I learn what I need, solve a problem, and move on.''

'''Prioritization'''

''I prioritize by what I need at the moment. I need A, I read A. I need B, I read B. In problem solving, rarely do you need to read dozens of articles at the same frame of time. You read what you need and fix things.''

'''Speed (of reading)'''

''I read fast enough. I have no idea how interrupted reading could be faster than normal reading or skimming.''

'''Speed (of formulating items)'''

''I do not need items.''

'''Meticulousness'''

''I am very meticulous. I do not need help in that department.''

'''Training'''

''The training in skills that one is supposed to develop with incremental learning is irrelevant. The skills are either of little value for me, or I get them in the course of my work anyway.''

* '''''Recognizing suitable texts''' - all texts are suitable if they lead me to a goal. I Google, I read, I solve. Period.''
* '''''Formulating knowledge''' - why waste time on formulating anything? I read and work. I need no intermediary time waster.''
* '''''Mnemonic skills''' - all those mind maps sound like a monumental waste of time.''
* '''''Speed-reading skills''' - my speed of reading is fast enough for my needs.''
* '''''Semantic skills''' - I develop these in the course of normal reading too. I agree. The more you read, the better you get at understanding what you read.''
* '''''Prioritization skills''' - I do not prioritize my reading. I read and execute.''
* '''''Editing and SuperMemo skills''' - SuperMemo is one thing I really do not need.''

'''Fun'''

''I am not sure how adding complexity to your life with the concept of incremental learning can be more fun than just doing great things, doing them well, and having people praise your for the good you have accomplished.''
</div>
</blockquote>

=== SuperMemo user's opinion: Why I do not use incremental reading? ===

Here is an exemplary opinion of a user of SuperMemo who understands incremental reading, and still has many reservations:

<blockquote>
<div class="bs-callout bs-callout-comment">
''I have been using SuperMemo for 19 years now (with occasional breaks). I understand the concept of incremental reading pretty well. I read about it a lot. Many times I thought of trying it. However, I am still not convinced that I should embark on this journey. Please see my story. I will gladly hear your opinion. Am I mistaken or am I different, you think? Or is the incremental reading picture a bit too rosy?''

'''Learned read-only mentality'''

''I started making repetitions years back with a goal to improving my English. To do that, I relied on learning with a ready-made Advanced English collection containing 40,000+ words and phrases. The learning material was comprehensive (far exceeding my needs), and the default learning process efficient enough to quickly propel me into good command of the language. Encouraged by the results I kept exploring the program but rarely took advantage of the ever-expanding gamut of SuperMemo's new advanced options. It seemed as if I hadn't really needed them to get the results I wanted (at least in the English language department). All I did was make outstanding repetitions, memorize my daily quota of new words and phrases, and, occasionally, add a piece of vocabulary or grammar which wasn’t originally in the collection.''

'''Incompatibility of my profession with incremental approach'''

''As for my other learning needs, they start with my professional life which happens to be computer programming. And it seems like an area of knowledge which is not particularly well-suited for declarative learning. Furthermore, Dr Wozniak himself admits learning materials for programming are not suitable for processing with incremental reading. In other words, I seem to be unable to realize the promise of incremental reading in my life’s vocation thus leaving me less determined to explore it for other domains.''

'''Stress factor'''

''I believe that I have achieved a pretty good mastery of SuperMemo over the years. It does not mean though that SuperMemo still isn’t able to throw a curve ball every now and then. Take the '''Spread priorities''' feature for instance. When I first checked it, I was greeted with ''0!!!'' against the yellow background in one of the fields: '''Step'''. I tried to change the value but no matter what I did I kept coming back to square one. As it turned out, it was SuperMemo’s way of telling me there aren’t enough priority positions for the subset I selected. SuperMemo may lack in the implementation department, even though it tries to address this, many times, at great length, in the documentation. In other words, with enough persistence you can find your way around the program. This journey of discovery may be stressful but nowhere near as much as the whole idea of incremental reading if you give it a careful thought. In fact, the more I read about it (at an attempt to dispel my doubts), the more stressful it becomes. Particularly, the entire prioritization stage seems overwhelming (to the point of nausea). The advantages you so diligently describe may all be true (although some are far from obvious and/or instant). Most of those advantage will hold only as long as you set the priorities right. But the entire process seems like a wild goose chase to me (not only because the priority queue is relative in its design). It would be interesting to know how much the student actually spends on priorities in the course of a daily learning session. For example, take a priority scale. In its relative variation (i.e. when expressed as percentage), it ranges from 1 to 100 which is simply too wide (the creators of the ABC analysis recognized that long time ago). I remember reading somewhere SuperMemo novices find it hard to grade yourself on the 6-piece scale. This now seems like a breeze as compared with giving a priority to a piece of knowledge. I understand the intention for more granular control but it is simply mind-boggling. To make matters worse, there are a number of processes (both manual/conscious and (semi-)automatic/unconscious) that erode your painstakingly set priority hierarchy. I do not mind when it actually changes but what if I accidentally stumble against an item whose priority I set as important, it still remains important, but somehow ended up in the later stages of the priority queue (and, from all what I have read, it is more than a remote possibility). When I read the article, I am often reminded of the following quote by Margaret Thatcher: &quot;''I do not know anyone who has got to the top without hard work. That is the recipe. It will not always get you to the top, but should get you pretty near.''&quot; It holds a great promise but it is far from guarantee. I believe, I recognize really well how complex incremental reading actually is, and having that clearly in mind, I conclude your advantages section offers even less in the form of guarantee. You can invest a whole lot of time, and in the end SuperMemo will only make you realize how dismal you are at managing it all (because, make no mistake, ''you'' need to carefully manage it). The toolset may be rich but is it natural? Furthermore, it all depends on the correct input. Your advantages include [[Incremental reading#Uniform progress|uniform progress]]. Yeah, right. Developing tunnel vision when you import and process a large number of articles related to a single topic is as much if not (considering our human deficiencies) more likely.''
</div>
</blockquote>

==== Comment: You need to try incremental approach to get a good feel ====

In the section [[#Incremental learning is not for everyone|Incremental learning is not for everyone]], you can read about personality factors that may conspire to make you fail with incremental learning. It is hard to make a prediction on the basis of your text. The fact that you studied [[Glossary:Incremental_reading|incremental reading]] is a very good predictor of success. On the other hand, the fact that the mere concept of [[Glossary:Priority_queue|priority queue]] can generate stress, tells you a lot about your personality. You might just be one of those who excel at theory and fail in practical situations for a number of reasons. You should definitely give incremental learning a try and see how it works. If frustrations keep multiplying beyond the first two months of use, your chances may drop. It is vital to start from simplest concepts: import, extract and cloze. All the remaining tools should wait for their turn until you start enjoying the process. For the same reason, you should start from learning about things you love (not about algebra or economics). Your hobby, your pet, your favorite sport, health, your favorite music band, your favorite actor or actress, etc. Use [http://en.wikipedia.org Wikipedia] as its texts are well-suited for [[Glossary:Incremental_reading|incremental reading]]. Only after 2-3 months, when you start sensing the new quality of knowledge emerge in your mind, will you be able to anchor emotionally in the process and find sufficient motivation to overcome successive obstacles.

Short comments on the rest of your text:

* '''tunnel vision''': experience shows that incremental learning favors broad learning as opposed to developing a tunnel vision. After all, the more you know, the more you know you do not know. You are more likely to have your day monopolized by a single topic when you do traditional reading or browsing. In incremental learning you face the whole spectrum of your knowledge (with a natural bias towards your interests). Unpredictability of incremental learning sequences, also in different moods or different contexts (e.g. change of interest), increases the quality of the creative processes. This is where curiosity is born. You are more likely to become interested with a seemingly irrelevant detail when you have a peaceful time allocation typical of incremental learning. This is how you keep branching to new areas. This branching is less likely when you work in conditions of stress, time constraints, and information overload, typical to free browsing on the net. In free browsing, you can open 20 browser tabs and struggle with the right course of action: read now, make favorite, save to disk, skim, give up, etc. In incremental learning, you just import all tabs, and return to your main line of thought. Naturally, it helps greatly if you prioritize the tabs. This is a way of consolidating new interests as opposed to giving them just a temporary brush. SuperMemo's capacity to prevent tunnel vision is mentioned amongst its advantages. See: '''[[Incremental reading#Advantages of incremental reading|Advantages of incremental learning]]: Uniform progress'''
* '''wild goose chase''': the way shifting [[Glossary:Priority|priorities]] work in SuperMemo is exactly how priorities behave in real life situations: your yesterday's interests are ruthlessly replaced by today's interests, and so on. This is a wild goose chase indeed. But the fault is not with SuperMemo, but with human nature. Please recall that [[Glossary:Incremental_reading|incremental reading]] started from [[Glossary:Reading_list|reading lists]]. [[Glossary:Priority_bias|Priority bias]] could result in having a thousand or ten thousand top priority articles. This situation was not manageable and not rational. With relative [[Glossary:Priority|priorities]] you can observe the process, and work to improve your strategies, and your personal honesty about priorities. This will not stop the wild goose chase, but you have a range of tools to correct your mistakes (e.g. when discovering a neglected portion of the material). However long you think about it, you are not likely to find a better remedy to your knowledge greed than a simple [[Glossary:Priority_queue|priority queue]]! You have signaled the problem, but you neglect the fact that in real life, the problem is incomparably greater. In SuperMemo, if you discover neglected material, you can up-prioritize it. In traditional learning, you forget the material and the problem itself. It is better to chase the goose that to have it swept under the carpet. 
* '''stress''': if the prioritization process is stressful, you can initially ignore the [[Glossary:Priority_queue|priority queue]]. You will needed it only when your [[Glossary:Collection|collection]] grows beyond a manageable size. You can then gradually introduce various prioritization levels. For example, you can have a shy start with just two levels: '''''Important: 22%''''' and '''''Unimportant: 88%'''''. You will soon discover that you need more precision, esp. that you will never escape the feeling &quot;''this article must land in top 3 articles''&quot;, which at some point may mean you need decimal places or referencing the [[Glossary:Priority_queue|priority queue]] by position.
* '''user friendliness''': the problem of '''Step''' in '''[[Subset operations#spread_priorities|Spread priorities]]''' is typical of SuperMemo, which claims to be unfriendly for novices to better satisfy the pros. This option might display a message: &quot;''You are trying to squeeze more [[Glossary:Element|elements]] into the range than the size of the range''&quot;. However, this would make pros apoplectic, as it is very easy to input wrong data. You do not want to waste time on clicking away the message. A pro will just want to correct the input. This problem cannot be remedied by '''[[File menu|File]] : [[File menu#Level|Level]]''' because '''[[Subset operations#spread_priorities|Spread priorities]]''' is available only at advanced levels. This means that the user should already be inured to the fact that the first use of any function in SuperMemo may be difficult. A more sophisticated solution, e.g. a wizard or messages incorporated in the dialog, would probably not pass <code>Value/Time</code> criterion on the ever-bloating SuperMemo tasklist. You are right, SuperMemo is not nice for novices. If you have simpler solution ideas, please mail them. They are always read carefully.
* '''programming''': it is not true that [[Glossary:Incremental_reading|incremental reading]] is unsuitable for programmers (for example see: [http://super-memory.com/articles/programming.htm SuperMemo as a new tool for programmers]). This misunderstanding might come from statements such as &quot;''computer code is not suitable as text to be processed with incremental reading''&quot;, even though it is conceivable that [[Glossary:Incremental_reading|incremental reading]] might be useful in code analysis (e.g. when a large number of well-encapsulated short procedures of different priority needs to be reviewed).

= Priority queue =

{{:Priority queue}}

= Visual learning =

{{:Visual learning}}

= Incremental video =

{{:Incremental video}}

= Incremental audio =

== Incremental audio: Introduction ==

Incremental audio is analogous to [[Glossary:Incremental_video|incremental video]]. You can use incremental video to process audio information from [http://youtube.com YouTube], or you can use a dedicated extractor bar in the sound [[Glossary:Component|component]] to import and process sound files (e.g. MP3, WMA, WAV, etc.).

== Incremental audio with sound files ==

Working with sound files is the same as working with video files. The only difference is that you will use sound [[Glossary:Component|components]] instead of video components. Those will be created for you automatically at import with '''[[File menu|File]] : [[File menu#Import|Import]] : Files and folders''' (available from the [[main menu]]). For details see: [[#Incremental video with video files: Outline|Incremental video with video files: Outline]].

== Incremental audio with YouTube ==

See: [[#Incremental video with YouTube: Outline|Incremental video with YouTube: Outline]] to find out how to use [http://youtube.com YouTube] material in incremental audio process.

= Incremental mail processing =

{{:Incremental mail processing}}

= Incremental learning in creativity =

{{:Creativity and problem solving in SuperMemo}}

= Neural creativity =

{{:Neural creativity}}

= Incremental learning myths =

SuperMemo always had to struggle with myths slowing down its popularization. Preventing the reappearance of myths appears to be a never-ending battle. The knowledge about SuperMemo has grown to a substantial volume. Not all users can afford reading dozens of articles. Many users are bound to arrive to the same wrong conclusions independently of others. Some of these myths are rooted in general myths of memory. Others seem to spring from the common sense thinking about learning. Here are some most damaging myths related to spaced repetition, SuperMemo, and incremental learning.

=== Myth: Many people are successful without using SuperMemo, hence its importance is secondary ===

'''Myth: Many people are successful without using SuperMemo, hence its importance is secondary'''.<br>
'''Fact''': Neither [http://en.wikipedia.org/wiki/Charles_Darwin Darwin] nor [http://en.wikipedia.org/wiki/Issac_Newton Newton] had access to computers, yet computer illiteracy may make today's scientist entirely impotent. Similarly, with a growing importance of knowledge, neglecting the competitive advantage of a wider and stable knowledge will increasingly limit your chances of successful career in science, engineering, medicine, politics, etc. You can live without SuperMemo, but it can definitely raise your learning to a new level.

=== Myth: Natural mechanisms for selecting important memories are good enough. We do not need a crutch ===

'''Myth: Natural mechanisms for selecting important memories are good enough. We do not need a crutch'''. The evolution produced an effective forgetting mechanism that frees our memory from space-consuming and perhaps irrelevant garbage. This mechanism proved efficient enough to build the amazing human civilization. Consequently, many believe that there cannot be much room for improvement.<br>
'''Fact''': The forgetting mechanism was built in abstraction from our wishes and decisions. It only spares memories that are used frequently enough. Nowadays, we are smart enough to decide on our own which knowledge is vital and which is not. A single peek into a dictionary may often take more time than the lifetime time cost of refreshing the same word in SuperMemo. This is just the least spectacular example. Human history is rich in monumental errors coming from ignorance. NASA's confusion of imperial and metric units cost a lost Mars probe. Confusion of comma with a dot in Fortran, cost a Venus probe. Errors in English communication caused many aerial and maritime catastrophes. A piece of knowledge in surgeon's mind may be worth the life of his patient. Forgetting is too precarious to leave mission-critical knowledge to your brain's own devices. SuperMemo puts you in the driver's seat. You can decide what your remember and what you forget. For more see: [[#Myth: Memorization is not needed|Memorization is not needed]] and [[#Myth: We are good at remembering important things|Memory has an excellent ability to retain important information]]

=== Myth: We cannot improve memory by training ===

'''Myth: We cannot improve memory by training'''. Infinite memory is a popular optimist's myth. A pessimist's myth is that we cannot improve our memory via training. Even [http://en.wikipedia.org/wiki/William_James William James] in his genius book ''[http://psychclassics.yorku.ca/James/Principles/ The Principles of Psychology (1890)]'' wrote with certainty that memory does not change unless for the worse (e.g. as a result of aging or disease).<br>
'''Fact''': If considered at a very low synaptic level, memory is indeed quite resistant to improvement. Not only does it seem to change little in the course of healthy life. It is also very similar in its properties across the human population. At the very basic level, synapses of a low-IQ individual are as trainable as that of a genius. They are also not much different from those of a mollusk ''Aplysia'' or a fly ''Drosophila''. However, there is more to memory and learning than just a single synapse. The main difference between poor students and geniuses is in their skill to represent information for learning. A genius quickly dismembers information and forms simple models that make thinking easy. Simple models of reality help understand it, process it, and remember it. What [http://en.wikipedia.org/wiki/William_James William James] failed to mention is that a week-long course in [http://en.wikipedia.org/wiki/Mnemonic_techniques mnemonic techniques] dramatically increases learning skills for many people. Their molecular or synaptic memory may not improve. What improves is their skill to handle knowledge. Consequently, they can remember more and for longer. Learning is a self-accelerating and self-amplifying process. As such it often leads to miraculous results.

=== Myth: SuperMemo repetitions take too much time to make it worthwhile ===

'''Myth: SuperMemo repetitions take too much time to make it worthwhile'''. Many users struggle with an increasing load of repetitions and may conclude that the effort is not worth the outcome.<br>
'''Fact''': Just 3 well-selected [[Glossary:Item|items]] memorized per day may produce a better effect than a hundred crammed facts. This means that even a minute per day will make a world of difference, as long as you pay a close attention to what you learn. Not all knowledge is worth the effort of 99% [[Glossary:Retention|retention]]. High [[Glossary:Retention|retention]] should be reserved only for mission-critical facts and rules. Last but not least: [http://super-memory.com/articles/20rules.htm knowledge formulating skills] may cut the learning time for beginners by a wide margin. For more see: [[#Myth: High retention results in slow learning|High retention results in slow learning]]

=== Myth: As you add more material to SuperMemo, your repetition loads mount beyond being manageable ===

'''Myth: As you add more material to SuperMemo, your repetition loads mount beyond being manageable'''. No [[Glossary:Item|item]] added to SuperMemo is considered &quot;memorized for good&quot;. For that reasons, all [[Glossary:Item|items]] are subject to review sooner or later. There must therefore be an inevitable increase in the cost of repetitions.<br>
'''Fact''': It is true that a large number of [[Glossary:Outstanding_element|outstanding repetitions]] is the primary excuse for SuperMemo drop-outs. However, computer simulations as well as real-life measurements show that, with the constant daily learning time, the acquisition of new knowledge does not visibly slow down in time except for the very first couple of months. In other words, from a long-term perspective, the acquisition of new knowledge is [http://super-memory.com/articles/theory.htm#Figure+1 nearly linear]. Older [[Glossary:Item|items]] are repeated less and less frequently leaving room for new material. The exponential nature of this &quot;fading&quot; explains why we can continue with a heavy inflow of new material for decades.

=== Myth: People differ in the speed of learning, but they all forget at the same speed ===

'''Myth: People differ in the speed of learning, but they all forget at the same speed'''.<br>
'''Fact''': Although there are mutations that might affect the forgetting rate, at the synaptic level, the rate of forgetting is indeed basically the same (independent of how smart you are). However, the same thing that makes people learn faster, helps them forget slower. The key to learning and slow forgetting is representation (i.e. the way knowledge is formulated). If you learn with [[SuperMemo]], you will know that [[Glossary:Item|items]] can be very difficult or very easy. The difficult ones are forgotten much faster and require shorter [[Glossary:Interval|intervals]] between repetitions. The key to making [[Glossary:Item|items]] easy is to [http://super-memory.com/articles/20rules.htm formulate them well]. Moreover, good students will show better performance on the exactly same material. This is because the ultimate test on the formulation of knowledge is not in how it is structured in your learning material, but in the way it is stored in your mind. With massive learning effort, you will gradually improve the way you absorb and represent knowledge in your mind. The fastest student is the one who can instinctively visualize and store knowledge in his mind using imagery that provide minimum-information and maximum-connectivity.

=== Myth: Hypertext can substitute for memory ===

'''Myth: Hypertext can substitute for memory'''. An amazingly large proportion of the population holds memorization in contempt. Terms &quot;''rote memorization''&quot;, &quot;''recitatory rehearsal''&quot;, &quot;''mindless repetition''&quot; are used to label any form of memorization or repetition as unintelligent. Seeing the &quot;''big picture''&quot;, &quot;''reasoning''&quot; and leaving the job of remembering to external hypertext sources are supposed to be viable substitutes.<br>
'''Fact''': Knowledge stored in human memory is associative in nature. In other words, we are able to suddenly combine two known ideas to produce a new quality: an invention. Hypertext references are a poor substitute for associative memory. Two facts stored in human memory can instantly be put together and bring a new idea to life. The same facts stored on the Internet will remain useless until they are pieced together inside a creative mind. A mind rich in knowledge, can produce rich associations upon encountering new information. An empty mind is as useful as a toddler given the power of the Internet in search of a solution. Biological neural networks work in such a way that knowledge is retained in memory only if it is refreshed/reviewed. Learning and repetition are therefore still vital for the progress of mankind. This humorous text explains the importance of memory: [http://super-memory.com/articles/users/memorizing.htm It is not just memorizing]

=== Myth: We do not need SuperMemo, all we need is to build an index to knowledge sources ===

'''Myth: We do not need SuperMemo, all we need is to build an index to knowledge sources'''. With multiple on-line sources of knowledge, some people are tempted to believe that all we are supposed to learn is a sort of index to these external sources of knowledge.<br>
'''Fact''': Even &quot;index to knowledge&quot; is subject to forgetting and needs to be maintained via repetition or review. All creative geniuses need knowledge to form new concepts. The extent of this knowledge will vary, but the creative output does depend on the volume of knowledge, its associative nature, and its abstractness (i.e. its relevance in building models).

==== Index to knowledge ====

A less extreme version of &quot;[[#Myth: Memorization is not needed|memorization is not needed]]&quot; myth, is that &quot;we only need to remember an index to knowledge&quot;. It is true that abstract knowledge and general concepts are most useful. However, this type of knowledge is also subject to forgetting. Moreover, not all knowledge is inferential. To reason about [http://en.wikipedia.org/wiki/Antarctica Antarctica], we often need to know that it is cold. We might derive the temperature conclusion from the position of [http://en.wikipedia.org/wiki/Antarctica Antarctica] on the map, but we need the knowledge of the map, the knowledge of the climate, and the knowledge of the earth's position in reference to the sun. Any kid will admit that it is simpler to just remember that [http://en.wikipedia.org/wiki/Antarctica Antarctica] is white and cold. In this case, rules are more useful, but facts are easier to remember.

A ''mere mortal'' will usually be aware that doughnuts are foods that should better be avoided by people who fear a heart disease. The fact &quot;''Doughnuts ain't good for heart''&quot; contributes to an average man's knowledge about health. This fact probably does not need SuperMemo. After all, most of us will refresh the knowledge about doughnuts and the heart each time we see a tasty doughnut. The ''Doughnut Fact'' contributes also to our ''index to knowledge''. It is enough to jump to Google and type ''+doughnut +heart'' to make a good use of this particular entry in our index. The search will help us recover more knowledge about the ''relationship'' between doughnuts and the heart.

If we want to enhance our ability to think and conclude about doughnuts and the heart, we might try to remember the following:

# '''Fact 1''': ''Doughnuts are high in trans fatty acids''
# '''Rule 1''': ''Trans fatty acids in foods tend to lower [http://en.wikipedia.org/wiki/High-density_lipoprotein HDLs] (high-density lipoproteins)''
# '''Rule 2''': ''Lower blood [http://en.wikipedia.org/wiki/High-density_lipoprotein HDLs] are a major risk factor in cardiovascular disorders (e.g. arteriosclerosis)''

There is an advantage to knowing the above facts and rules: upon finding out that ''French fries are high in trans fatty acids'', we will be able to use '''Rule 1''' and '''Rule 2''' to derive a new fact: ''French fries ain't good for the heart''. The awareness of the above rules will increase our reasoning ability. In the terminology of knowledge engineering, we will be able to derive new facts and rules from the existing set of facts and rules. In plain language, we will know more than we have actually learned. We will be able to conclude more. We will become ''more intelligent'' (if intelligence was defined as [http://super-memory.com/articles/genius.htm the inferential ability of the human mind]).

Yet there is a downside to remembering the ''trans fat rules''. They are not as plain as the ''Doughnut Fact'', and they may not effectively be refreshed upon a sight of a doughnut. Consequently, we may simply forget the link between ''doughnuts'', ''trans fat'', and ''the heart''. This is where SuperMemo comes to play a role. It will help you refresh the ''trans fat rules''. It will minimize the number of reviews in lifetime. In other words, it will help you keep the ''trans fat rules'' in your forgetful memory. Thus SuperMemo makes sure that your &quot;''index to knowledge''&quot; remains intact in your mind.

=== Myth: In incremental reading, you spend mere seconds reading a topic ===

'''Myth: In incremental reading, you spend mere seconds reading a topic'''.<br>
'''Fact''': The time devoted to a [[Glossary:Topic|topic]] depends on your needs. It may be a few seconds (e.g. for a low priority boring subject), or it can be an entire day (e.g. before an exam, or when doing research, or just when following your passions).

=== Myth: Memorization is not needed ===

'''Myth: Memorization is not needed'''.<br>
'''Fact''': If all students followed the suggestion that memorization is not needed, we would live in a different world. Here is a humorous take on how this world might look like.

* If memorization was not needed, we could travel around the world without learning languages. After all, finding out words in a dictionary takes mere seconds
* If memorization was not needed, students of medicine would not need to cram details of human anatomy and physiology. Instead they would learn to use state-of-the-art expert systems with all answers built in. If you asked: ''What's up doc?'' You will hear: ''I have no idea but wait ... I will check it out on my computer''
* If memorization was not needed, all exams such as SAT, GRE, TEOFL, FCE, GCSE, USMLE, etc. would be a great waste of human time and resources. Students should rather come to exams with their link to the web and figure out answers ''ad hoc''. Or they would just read out relevant paragraphs from a textbook
* If memorization was not needed, nobody should poke fun at George Bush Jr. for his lack of knowledge of the heads of state of Chechnya, Taiwan, or Pakistan. After all, he can find those names in seconds on his smartphone. If using external storage was permissible, nobody should blame George W. for mixing up Slovakia with Slovenia, or refer to Kosovars as Kosovarians, East Timorese as East Timorians, or Greeks as Grecians (NB: the word [http://www.dictionary.com/search?q=grecian ''Grecians'' is considered correct too])
* If memorization was not needed, you could be a rocket scientist at NASA tomorrow! After all, rocket scientists follow well-known rules written in well-known manuals organized in a very well-known manner. If you need to plot the trajectory for Galileo to beam images of Europa back to Earth... no problem... take the ''Advanced Calculus'' textbook, figure it out, and send your billion dollar mission on course

The advantage of keeping knowledge in your head as compared to keeping it in external sources can metaphorically be compared to the advantage of going from primary through secondary to university education as opposed to getting a week-long course on digging info from external sources. Nearly all parents seem to prefer to choose the former for their kids.

=== Myth: High retention results in slow learning ===

'''Myth: High retention results in slow learning'''.<br>
'''Fact''': You need to understand a clear distinction between the two extremes of learning:

* '''high-retention-low-volume learning''' (as in early versions of SuperMemo) - in which you make sure you remember 95 or more percent of the studied material
* '''low-retention-high-volume learning''' (as in traditional forms of learning) - in which you quickly process large chunks of the material while having to struggle with massive forgetting

Reading books belongs to the low-retention category, while memorizing 10-20 [[Glossary:Item|items]] per day with SuperMemo belongs to the high-retention category. The optimum reading strategy will find the golden mean between these two. You should not give up traditional reading. Neither should you expect to put all your study material into SuperMemo. You should choose a middle-ground strategy. For example, if you consistently spend 90% of your time on reading and 10% of your time on adding '''most important''' findings to SuperMemo, your reading speed will actually decline only by some 10%, while the [[Glossary:Retention|retention]] of the most important pieces will be as high as programmed in SuperMemo (i.e. usually 95%).

[[#Incremental reading|Incremental reading]] provides you with a precise tool for finding the optimum balance between speed and [[Glossary:Retention|retention]]. You will ensure high-retention of the most important pieces of text, while a large proportion of time will be spent reading at speeds comparable or higher than those typical of traditional book reading.

It is worth noting that the learning speed limit in high-retention learning is imposed by your memory. If &quot;memorizing&quot; one-book-per-year sounds like a major disappointment, the roots of this lay in human memory. Our current knowledge of psychophysiology and pharmacology does not provide any means that could allow of breaking beyond those limits. We are left with the choice between high-speed and high-retention. Incremental reading gives you a full hands-on control over finding the optimum balance.

=== Myth: We are good at remembering important things ===

'''Myth: We are good at remembering important things'''.<br>
'''Fact''': The evolution of the brain proceeds too slowly to have helped us adapt its structures to abstract thinking. What was excellent for survival 200,000 years ago does not suffice to process modern abstract knowledge. Simple computational tasks such as multiplication or division proceed in a shamefully inefficient way in the human mind. After all, early humans did not need to multiply (explicitly and a lot). At the same time, for long, computers found it hard to compete with the visual cortex in pattern recognition and processing. Recognizing the enemy or prey was critical not only for ''Homo sapiens'' but also for birds, reptiles, fish or even insects. The only measure of the importance of knowledge our brain synapses have at hand is the pattern of repetition, levels of circulating hormones at the time of exposure, and a limited impact of conscious attentive labeling of information as important at the moment of encoding. Forgetting is needed to optimize knowledge storage; hence we have to forget less important things. Repeated use of the memorized knowledge is a good but far from perfect measure of importance (see: [[#Myth: We remember useful things because we use them|We remember useful things because we use them]])

Modern life has changed the hierarchy of value and importance of knowledge. The link between importance and repeated use has been severed. A flashy lingerie billboard we see every morning is not likely to be more important than dozens of volatile facts pertaining to our professional life. Regrettably, there is no circuit in our brain that would let us consciously etch important memories: ''This is important! I must not forget it!'' All we can do is to use the trick of reverberation or mnemonic techniques which... still will usually not last long unless we apply [[Glossary:Spaced_repetition|spaced repetition]] as in SuperMemo.

If memory had an excellent ability to retain important information:

# you would not tremble before an exam and confusingly run through the notes to be sure that at the zero hour you won't suffer from a blackout. At the same time, you could easily recall details of a [http://en.wikipedia.org/wiki/Schwarzenegger Schwarzenegger] movie seen last evening or even weeks before the exam. Clearly, Arnie beats the ups and downs of the [http://en.wikipedia.org/wiki/Ottoman Ottoman Empire]. And if you think the Ottoman Empire had much greater an impact on humanity than the island shootout in ''Commando'', you are still likely to remember the muscle and the machine gun far better than the timeline of the sultans
# you would never have problems with recalling the date of your mother-in-law's birthday. This is a piece of data that is critical to your marital harmony!
# you should instantly forget the Olympic champions in football in Atlanta 1996 or Munich 1972. After all this might be a classic case of unimportant knowledge. Yet few Africans would forget how Nigeria beat Argentina 4:3 in 1996. Similarly, few Poles would forget the most memorable moment in the history of Polish football: Olympic championship in Munich.

The truth is that we excellently remember only things that are both easy to remember and repeated frequently enough. The brain does not have an internal measure of importance (other than limited volitional control or control via repeated exposure)! Your memory storage ruthlessly deletes your career-critical knowledge with the same ease as it ravages the traces of last year's golf scores or contributing names listed at the end of a boring soap opera. SuperMemo will ensure you remember your mission critical data. What is in, stays in your memory. What is out, is free to go.

=== Myth: We remember useful things because we use them ===

'''Myth: We remember useful things because we use them'''.<br>
'''Fact''': Not only are we limited in our ability to remember things that we consider important. We cannot even rely on the fact that frequently used memories are remembered better. There are two main reason why frequency of use is not sufficient:

* '''[[Glossary:Spacing_effect|spacing effect]]''' (frequently used memories may be very volatile): if you use things often, e.g. a credit card's pin code, you may surprised by a sudden unexpected [[Glossary:Lapse|lapse of memory]]! The facts that are used too often or are too obvious, do not challenge the memory system well enough. They are remembered superficially. Any break from use, stress, change of context, or innocent interference from other memories may cause a sudden inability to recall a fact from memory. Even SuperMemo may not help in those cases. Those easy [[Glossary:Item|items]] often grow their [[Glossary:Interval|intervals]] fast and cannot prevent the spacing effect.
* '''tunnel vision''': when we rely on frequently used memories, we are slow at expanding our horizons. With [[Glossary:Spaced_repetition|spaced repetition]], you can quickly grow stable and coherent knowledge into new areas. This expansion is not easy or even possible when relying on frequent use. A conscious intervention is necessary. In an extreme case, you might ask a cat why it does not master mathematics. In a cat's tunnel vision, simple goals lead to a narrow set of factual knowledge that is needed and remembered. A cat will never fly to Mars (on its own).

Opponents of SuperMemo often say: &quot;''Whatever knowledge I need, I use often. If I use it often, I do not forget''&quot;. This is a false conviction. [[Glossary:Spacing_effect|Spacing effect]] and forgetting are unpredictable. The whole world of useful things escapes your brain on a daily basis. You may say that if need to fly to Mars, you can just read about math tools that you find useful. However, the tool that is useless today might be useful in 5 years, or tomorrow, when you tackle a different problem. You need the whole toolset ready in your mind. The associative power of the extra knowledge just makes you a better problem solver and a stronger thinker.

Tunnel vision effect may be counteracted with extensive reading, however, there is no better way of reading than [[Glossary:Incremental_reading|incremental reading]].

= History of incremental learning =

Incremental learning might be as important to SuperMemo as the original [https://supermemo.guru/wiki/General_principles_of_spaced_repetition repetition spacing idea]. Incremental learning eliminates a number of bottlenecks that limit various stages of ''[[#Knowledge acquisition: Areas of optimization|knowledge acquisition]]''.

The name ''[[#Incremental reading|incremental reading]]'' first appeared in SuperMemo 2000. However, the concept is not new. It originated from combining our natural reading habits with the demands of ''[[SuperMemo|spaced repetition (SuperMemo)]]''. We rarely pick up a book and read it cover-to-cover in one go. At school we often dig through a number textbooks used for different courses. At home we stop reading a book to read a newspaper and then stop reading the newspaper to watch TV. A combination of needs and interests determines how far we go with the reading of an individual text. SuperMemo drives this concept to an extreme by letting you read just one sentence from one chapter from one book and then go on to reading [[Glossary:Extract|extracts]] from a thousand books and/or articles. SuperMemo's contribution here is only the management of this multi-source reading process. As for the creative aspect of [[#Incremental reading|incremental reading]], [https://en.wikipedia.org/wiki/Niels_Bohr Niels Bohr] is known to have used the power of
intermittent reading and intermittent thinking to maximize his creative output. He would keep dozens of shelves with outlines of ideas. He would return to individual shelves from time to time, esp. if he was inspired by a conversation, thinking, experiment, or reading. He would then keep reading a single shelf, think and ponder, add new notes, etc. Many of those shelves ended up as scientific publications. In that sense, [https://en.wikipedia.org/wiki/Niels_Bohr Niels Bohr] employed rudimentary [[#Incremental reading|incremental reading]] in his creative work.

The approach used in incremental reading is widely employed by many creative individuals. Even if it is far less formal that [[#Incremental reading|incremental reading]] or even [https://en.wikipedia.org/wiki/Niels_Bohr Bohr]'s approach. [https://en.wikipedia.org/wiki/Michael_Gazzaniga Dr Michael Gazzaniga] puts it this way: &quot;''I think the creative process is directly related to the amount of time one spends mulling something over. I come back and revisit ideas, data, thoughts, all the time. I think this keeps key semantic networks active and then &quot;bingo&quot; an inconsistency or consistency suddenly presents itself to consciousness and the beginnings of a new idea appear''&quot;.

Here is a brief history of incremental reading:

* '''Before SuperMemo (1980)''': The author of SuperMemo and co-author of this incrementally written text, [https://supermemo.guru/wiki/Piotr_Wozniak Piotr Wozniak], in his student years, used his own &quot;cross-the-notebook&quot; method of learning. He would go through his paper notes and cross out those he was sure he would remember for the exam. He would read his notes over and over again until all the notes have been crossed out. Each pass was faster as there was less and less to read. That method was nearly a guarantee of excellent exam results. All his lecture notes were covered with crossed out shapes that surrounded pictures and portions of text. This &quot;granular treatment&quot; with &quot;processing attributes&quot; can be thought of as an early inspiration for future [[#Incremental reading|incremental reading]]
* '''SuperMemo algorithm (1984-1985)''' - The first ever attempt to measure [[Glossary:Optimum_interval|optimum intervals]] in learning resulted in the formulation of the SuperMemo method in 1985
* '''SuperMemo 1.0 (1987)''' - The birth of SuperMemo is also the birth of ''computational spaced repetition'', i.e. the use of computers in computing [[Glossary:Optimum_interval|optimum intervals]] in learning
* '''[https://super-memory.com/english/sms.htm SuperMemo 2 - SuperMemo 98 (1987-1998)]''' - In the years 1987-1998, users of SuperMemo had only two alternatives in the area of collecting learning material for learning with SuperMemo:
*# type it in and formulate it manually, or
*#  obtain ready-made learning material from a colleague, SuperMemo Library, etc.
: The only way SuperMemo supported learning from electronic sources was via '''Copy''' and '''Paste'''.
* '''[https://super-memory.com/articles/soft/sm99.htm SuperMemo 99 (1999)]''' made the first step towards efficient reading of electronic articles by introducing [https://super-memory.com/articles/read.htm reading lists] and the first primitive reading tools: '''Extract''' and '''Cloze'''. Reading lists were prioritized lists of articles to read. Extracts made it possible to split larger articles into smaller portions. Clozes made it possible to convert short sentences into question-answer format by means of ''[[Glossary:Cloze_deletion|cloze deletions]]'' (i.e. fill-in-the-blank questions)
* '''[https://super-memory.com/articles/soft/sm10.htm SuperMemo 2000]''' greatly increased the efficiency of reading by introducing the concept of ''[[#Incremental reading|incremental reading]]''. Incremental reading makes it possible to simultaneously read dozens of articles. Each article is read in small increments fully controlled and prioritized by the user and/or the default learning process. Components of [[#Incremental reading|incremental reading]] introduced in SuperMemo 2000: new [[Glossary:A-Factor|A-Factor]]-based [[Glossary:Topic|topic]] repetition scheme (i.e. the learning algorithm), [[Glossary:Read-point|read points]], formatting in [[Glossary:Extract|extracts]] and in [[Glossary:Cloze_deletion|clozes]] (SuperMemo 99 would ignore formatting), text highlights, source article link, [[Read toolbar|reading toolbar]], [[#Subset review|subset learning]], subset [[postpone]], and support for longer articles (SuperMemo 99 was limited to 64K articles)
* '''[https://super-memo.com/supermemo/supermemo2002.html SuperMemo 2002]''' brought [[#Incremental reading|incremental reading]] to a new level. In SuperMemo 2002, [[#Incremental reading|incremental reading]] became the primary learning mode for middle-level and advanced students. SuperMemo 2002 introduced [[HTML component|HTML-based]] [[#Incremental reading|incremental reading]]. For the first time, the user would see little difference between the material in his web browser and in SuperMemo. Other new features introduced by SuperMemo 2002: [[Web import|automatic learning material import]] from Internet Explorer, [[Glossary:Mid-interval_repetition|mid-interval repetitions]] that make it possible to review portions of material without damage to the learning process, e.g. before an exam ([https://super-memory.com/english/algsm11.htm Algorithm SM-11]), [[#Search and review|search-based learning]] (i.e. subset learning in which the subset is defined by advanced [[Find elements|search tools]]), dynamically modified [[Glossary:A-Factor|A-Factors]] that fine-tune the [[Glossary:Priority|priorities]] without user intervention, [[Postpone|postpone wizard]] that makes [[Glossary:Reading_list|reading lists]] obsolete, separate [[Glossary:Topic|topic]]/[[Glossary:Item|item]] statistics and new [[#Incremental reading|incremental reading]] progress statistics, [[References|reference labeling]], and more.
* '''[https://super-memo.com/supermemo/supermemo2004.html SuperMemo 2004]''' was developed solely with the view to perfecting the tools of [[#Incremental reading|incremental reading]]. The data collected from months of actual [[#Incremental reading|incremental reading]] was instrumental in enhancing the algorithm and the tools. Fine tuning of the modification of [[Glossary:Topic|topic]] [[Glossary:A-Factor|A-Factors]] enhanced the optimization of new material review in a heavily overloaded process. New tools included: rich statistics for monitoring and optimizing the learning process, tools for handling excessive delays in review, browsing sources of [[Glossary:Extract|extracts]] and [[Glossary:Cloze_deletion|clozes]], one-key [[References|reference labeling]], proliferating remote images, easy integration of remote images, and more.
* '''[https://super-memo.com/supermemo/supermemo2006.html SuperMemo 2006]''' made a major step in rationalizing the material [[Glossary:Overload|overload]] in [[#Incremental reading|incremental learning]] by introducing the ''[[#Priority queue|priority queue]]''. SuperMemo 2006 made it easier to import articles from the Internet (esp. from [https://en.wikipedia.org Wikipedia]). Material [[Glossary:Overload|overload]] could be handled with [[#Auto-sort and auto-postpone|auto-postpone and auto-sort]] tools. SuperMemo 2006 simplified importing, arranging, compressing, converting, zooming, and trimming pictures. SuperMemo 2006 could pick any folder on the user's disk and convert all the file archives into material that can be processed incrementally (e.g. article archives, picture archives, family albums, movie clips, documentation files, or assorted archives). SuperMemo 2006 also made it simple to do one-key searches and import of auxiliary learning material on the web with customizable tools (e.g. Google, encyclopedias, dictionaries, picture archives, etc.).
* '''[https://super-memo.com/supermemo/supermemo2008.html SuperMemo 2008]''' extended incremental learning into the areas of pictures, sounds and video with [[#Visual learning|visual learning]] and YouTube-based [[#Incremental video|incremental video]]. Imports from the web, esp. [https://en.wikipedia.org Wikipedia] and YouTube became simpler with set [[Glossary:Priority|priorities]], [[references]], [[Glossary:Category|categories]], etc. SuperMemo 2008 also simplified and automated the process of creating [[References|source material references]].
* '''[http://super-memo.com/supermemo/supermemo15.html SuperMemo 15 (2011)]''' improved [[Web import|web imports]] (e.g. eliminating article duplicates), splitting articles (e.g. automatic splitting of [https://en.wikipedia.org Wikipedia] articles into separate [[Glossary:Topic|topics]]), [[References|handling references]], and more. As SuperMemo 15 provided a full Unicode support, [[#Incremental reading|incremental reading]] in languages other than English became easier.
* '''[https://super-memo.com/supermemo/supermemo16.html SuperMemo 16 (2013)]''' consolidated all incremental learning technologies with the last two missing links: [[#Incremental audio|incremental audio]] and [[#Incremental video|incremental video]] with video files. It added more tools for [https://en.wikipedia.org Wikipedia] imports (e.g. downloading full resolution pictures intead of thumbs, etc.), and [[#Visual learning|visual learning]] (e.g. global paste). It also made it possible to separate [[Glossary:Extract|extracts]] from source articles in child [[Browser|browsers]]. Dozens of improvements to incremental learning process convert quantity into new quality. For example, SuperMemo 16 added import from Windows (Live) Mail, or a trivial, but extremely useful [[#Incremental reading|incremental reading]] function: '''[[Component menu#delete_before_cursor|Delete before cursor]]'''. For a detailed list see: [https://super-memory.com/archive/help16/new16.htm What's new in SuperMemo 16?]
* '''[https://super-memo.com/supermemo/supermemo17.html SuperMemo 17 (2016)]''' adds neural processing to [[incremental learning]]. [[Glossary:Neural review|Neural review]] is a form of [[subset review]] where semantic connections established by the [[Glossary:Knowledge_tree|knowledge tree]] and user-defined [[Glossary:Concept_map|concept maps]] can be followed in unpredictable manner (using a [[Glossary:Spreading activation|spreading activation algorithm]]). [[Neural creativity]] is a particularly powerful way of employing an incremental learning process in creativity and problem-solving. SuperMemo 17 employs a [https://supermemo.guru/wiki/SuperMemo_Algorithm new spaced repetition algorithm] which can handle most extreme departures from the optimum spacing of repetitions without compromising future optimum scheduling. For a detailed list see: [https://super-memory.com/archive/help17/new.htm What's new in SuperMemo 17?]
* '''[https://help18.supermemo.org/wiki/What's_new_in_SuperMemo_18? SuperMemo 18 (2020)]''': minor improvements to the [[Algorithm SM-18|algorithm]]. Dozens of minor enhancements (incl. speed, the handling of [[Glossary:Concept|concept]]s, etc.)
* '''[[What's new in SuperMemo 19?|SuperMemo 19 (2023)]]''': extending the range of [[web import|supported web browsers]] makes it easy to massively transfer knowledge from the web to [[incremental reading]]. [[Glossary:Extract|Extract]]s can be used on external sites. Working with [[Glossary:Concept|concept]]s is simplified with the benefit to the [[Glossary:Collection|collection]] structure and [[neural creativity]]. Page updates are easier (e.g. on Wikipedia imports)

= Supplementary materials =
* [https://youtu.be/DoQoeK53bP8 Using incremental reading (video)]
* [https://youtu.be/e3BHwRVQlCs Using incremental reading (commentary from Pleasurable Learning)]
* [https://supermemo.guru/wiki/Advantages_of_incremental_reading Advantages of incremental reading]
* [https://supermemo.guru/wiki/How_to_read_a_book_in_an_hour How to read a book in an hour?]
* [https://supermemo.guru/wiki/How_can_I_read_faster%3F How can I read faster?]