Skip to content

Latest commit

 

History

History
437 lines (242 loc) · 82.7 KB

The-anatomy-of-failure-and-success.md

File metadata and controls

437 lines (242 loc) · 82.7 KB
Raw

The anatomy of failure and success

剖析失败和成功

[TOC=2,5]

Formula for research failure

研究失败的公式

Some intuitions about spaced repetition are pretty common. This gives rise to a major question: why has not spaced repetition been investigated earlier and why didn't it permeate into the practice of learning? Intuitions are not enough, a good experimental design is also vital. This section explains why others were close but failed. How Ebbinghaus or Spitzer might have brought spaced repetition to life 90-130 years earlier. There must be something wrong with the immediacy of gratification in peer review and battle for grants. Why is there so much buzz in the field of drugging kids for school, while diseases that take a heavy death toll in less developed countries get little interest?

一些关于[间隔重复]的直觉(https://supermemo.guru/wiki/Spaced_repetition)很常见。这就产生了一个主要的问题:为什么没有更早地研究间隔重复，为什么它没有渗透到学习实践中?直觉是不够的，一个好的实验设计也是至关重要的。本节解释了为什么其他公司接近但失败了。艾宾浩斯或斯皮策如何在90-130年前将间隔重复带入生活。在[同行评审](https://supermemo.guru/wiki/Peer_review)和为获得资助而进行的斗争中，满足的即时性肯定出了问题。为什么在[给孩子下药上学](https://supermemo.guru/wiki/Confusing_creativity_with_ADHD)领域有这么多的讨论，而在欠发达国家造成重大死亡的疾病却很少引起人们的兴趣?

In this chapter, I try to figure out why spaced repetition was so late to come. Here is my list of delay factors prioritized by their impact:

  • computers make a dramatic difference in learning efficiency in spaced repetition, early formulations would not be viral enough even in the era of the Internet
  • web perpetuates knowledge and crystallizes its essence (e.g. at Wikipedia)
  • intuitions do not ensure good experimental design. Myself, Ebbinghaus, Spitzer, and others, produced designs that would add more noise and complexity to the issue
  • human culture is in a perpetual flow. We massively forget and re-discover old findings. This is as true of individuals as it is of cultures. Science is also subject to fads, fashions and forgetting. Only now, spaced repetition has reached the "impact density" needed to become "common knowledge". See: Discontinuity in the research on spacing effect
  • self-interest and self-learning are best drivers of applicability. Science permeates lives via practical applications. SuperMemo was the first practical application of spaced repetition that could reach thousands and then millions of users
  • there used to be a great deal of confusion between short and long-term memory, between spacing in lists, and spacing in a single day, and even between procedural and declarative learning
  • expanding, contracting or equally spaced schedule can all be made superior given appropriate timing
  • experimenters tend to use heterogeneous material (poems, lists, nonsense syllables, pages of questions, groups of students, etc.)
  • experiments where passive review is used instead of active recall do not benefit from the testing effect
  • experiments where intervals are measured in intervening items are based on different memory mechanisms and should not even be labelled as spaced repetition. To this day, a lot of confusion in research is caused by mixing up the measures of intervals without a clear separation of terminology
  • short-span of research projects makes spaced repetition investigations hardly possible
  • focus on classroom applications has confused the outcomes. Schools are not a good place for learning. Research that optimizes for school environment is of little relevance in free learning
  • ages old distinction between learning and retention confuses the thinking about optimal education. The distinction comes from schooling in which we learn first and then pray to forget little
  • terminology keeps mutating and this makes hard for new generations of researchers to capitalize on prior work. Even our own 1994 paper uses the term repetition spacing. See the list of terminological mutations to gauge the scope of the problem. Web, wisdom of the crowds, and wikis (like this one) are likely to remedy the problem of terminology

在这一章中，我试图找出为什么间隔重复这么晚才出现。以下是我列出的按影响程度排序的延误因素:

学习和保留之间由来已久的区别混淆了关于最佳教育的思考。区别在于学校教育，我们先学习，然后祈祷不要忘记什么

-术语不断变化，这使得新一代的研究人员很难利用以前的工作。甚至我们自己的1994年论文也使用了术语重复间隔。查看术语突变列表来判断问题的范围。网络、群众的智慧和维基(像这个)(https://supermemo.guru/wiki/SuperMemo_Guru)可能会解决术语的问题

In 2006, Will Thalheimer produced an excellent research review of spaced repetition research. It was based on over 100 research articles. However, in the summary Thalheimer cast doubt about the value of progressive spacing. This shows that even dozens of papers won't help if the experimental design is based on incorrect models.

In spaced repetition, it is not enough for review to be expanding. The review needs to be optimally expanding

2006年，威尔·塔尔海默(Will Thalheimer)发表了一篇关于间隔重复的优秀研究综述。它基于100多篇研究论文。然而，在总结中[Thalheimer对渐进间隔的价值提出了质疑](https://supermemo.guru/wiki/Will_Thalheimer: _doubts_about_expanding_排练)。这表明，如果实验设计基于错误的模型，即使几十篇论文也无济于事。

在间隔重复中，仅仅扩大复习是不够的。审查需要优化扩展

Failed experimentation

失败的实验

I will list three cases of experiments that were bound to fail or to confuse:

  • Herman Ebbinghaus (1885) was motivated by the science of memory. His intuitions were excellent, however, in the name of purity, he focused on nonsensical syllables. Inadvertently, he contributed to complexity of memory and interference, which he aimed to avoid. His forgetting curve is ruthless and discouraging. Had Ebbinghaus learned meaningful material of interest, he might have had extended the scope of his experiment. We all know from school that cramming nonsense is a form of mental torture. Had Ebbinghaus opted for practical applications, like myself, we might have had a paper variant of spaced repetition born a neat 100 years before the fact. Naturally, without computers, without the web, his ideas might still fall into the pit of silence for a century, as much as it has actually happened to the concept of the spacing effect
  • Herbert Spitzer (1939) was motivated by improving classroom performance. He opted for the worst case of heterogeneity. While I struggled with pages of mixed difficulty material, and Ebbinghaus struggled with meaningless material, Spitzer compounded those effects by the heterogeneity of minds. Instead of pages of questions, Spitzer had groups of pupils exposed to pages of reading materials. In those noisy conditions, it is hard to see the regularities of exponential forgetting and spaced repetition. Even if Spitzer's results were overwhelmingly positive, I doubt the bureaucracy of the education system would make a good use of the procedure. Education is driven by standardized testing, grades, and certificates. It is hardly ever driven by science and the actual impact of the learning procedure on learning outcomes. It is a factory system. Spitzer's efforts were as doomed as those groundbreaking ideas of Benezet produced a decade earlier (1929)
  • Wozniak (1985). My own misleading experiment on spaced repetition started on Jan 31, 1985, and might have ruined my efforts had it sufficiently confused my mind. Luckily, before the results came in, I designed a better experiment, and was never impacted by the confusion.

我将列举三个注定会失败或混淆的实验案例:

Ebbinghaus experiments (1885)

艾宾浩斯的实验(1885)

The myth 神话

A popular myth says that Ebbinghaus invented spaced repetition back in 1885. That myth was born from our own SuperMemo documentation. When we compiled the history of SuperMemo for the web in 1997, we added a few names with contribution to memory research. As explained here, it was important to keep SuperMemo grounded in science. My own contribution has always been minimized because of the negligible weight of my name. As Ebbinghaus came at the top of the list for chronological reasons, it soon gave birth to the idea that he was the father of spaced repetition. Even our own materials evolved in that direction by mistake of careless editing. Today, the web is swamped with Ebbinghaus and the serrated set of forgetting curves. If you Google for the forgetting curves, you will see this:

一个流行的神话说，早在1885年，艾宾浩斯就发明了间隔重复。这个神话起源于我们自己的SuperMemo文档。当我们在1997年为web编写SuperMemo的历史时，我们为记忆研究添加了几个名字。正如所解释的那样(https://supermemo.guru/wiki/Who_invented_the_name:_spaced_repetition%3F)，保持SuperMemo以科学为基础是很重要的。我自己的贡献总是被最小化，因为我的名字微不足道的重量。由于艾宾浩斯是按时间顺序排列的，所以很快就有人认为他是“间隔重复”之父(https://supermemo.guru/wiki/Spaced_repetition)。甚至我们自己的材料也由于编辑疏忽而向那个方向发展。今天，网络上充斥着艾宾浩斯和锯齿状的遗忘曲线。如果你对[遗忘曲线]进行谷歌(https://supermemo.guru/wiki/Forgetting_curve)，你会看到:

Google search for forgetting curves

Figure: Google search for "forgetting curve" (November 2017). The "serrated set" of forgetting curves dominates the search. Many of the pictures are labelled by wrong attribution with the name of Hermann Ebbinghaus. The more fit origin of the picture is presented in Two components of memory

**图:**谷歌搜索“遗忘曲线”(2017年11月)遗忘曲线的“锯齿集”主导了搜索。许多照片被错误地贴上了赫尔曼·艾宾浩斯的标签。图像的更合适的原点出现在内存的两个组件中

These are the same serrated curves that I depicted in Optimization of learning (1990):

这些就是我在优化学习(1990)中描述的锯齿形曲线:

img

Hypothetical mechanism involved in the process of optimal learning

For more on the myth see our blog: Did Ebbinghaus invent spaced repetition?

最优学习过程中的假设机制

想要了解更多，请看我们的博客:艾宾浩斯发明了间隔重复吗?

The fact 事实

We should rather be amazed with the fact that Ebbinghaus did not come up with spaced repetition. He was very close. He was touching all the right buttons.

我们应该惊讶于艾宾浩斯没有提出间隔重复。他离我很近。他完全按对了按钮。

This is what he wrote about the spacing effect:

这是他写的关于间隔效应:

For the relearning of a 12-syllable series at a definite time, accordingly, 38 repetitions, distributed in a certain way over the three preceding days, had just as favorable an effect as 68 repetitions made on the day just previous. Even if one makes very great concessions to the uncertainty of numbers based on so few researches, the difference is large enough to be significant. It makes the assumption probable that with any considerable number of repetitions a suitable distribution of them over a space of time is decidedly more advantageous than the massing of them at a single time.

因此，在前三天以某种方式重复38次，与前一天重复68次的效果是一样的。即使一个人基于很少的研究而对数字的不确定性做出很大的让步，这种差异也足以是显著的。它使得这样一种假设成为可能，即随着重复次数的增加，在一段时间内对重复次数进行适当的分配，肯定比在一段时间内对重复次数进行集中更有利。

Why did Ebbinghaus not take the next seemingly obvious step to see how the forgetting curve changes upon review? For one, he did not test forgetting, but the saving on re-learning. His tests were a review. With my approach it is easy to want to achieve a set level of retention. It is conceptually less intuitive to ask for a set reduction in the cost of re-learning.

为什么艾宾浩斯没有采取下一个看似明显的步骤来观察遗忘曲线在回顾时的变化?首先，他测试的不是遗忘，而是重新学习的储蓄。他的考试是复习。以我的方法，很容易想要达到一个设定的保留水平。要求减少重新学习的成本在概念上不太直观。

However, a solution to overcoming that conceptual obstacle would be to persist with learning. Creativity thrives on the investment of time and thinking in varying contexts. To me, the answer to the Ebbinghaus puzzle is very simple. Like a good scientist, Ebbinghaus was theory-perfectionist. He chose to memorize nonsense syllables to minimize interference from his prior knowledge. Unfortunately, this had a few bad side effects:

  • his lists were hard to learn and his forgetting curve was very steep. It makes for a very discouraging picture for any student who cares about remembering in the long term
  • his lists gave him no pleasure in learning. This is the exact opposite of what I experienced in 1985. Every item in my paper collection was a little step forward in advancing my knowledge to a new level. Once I plotted my first forgetting curve, I wanted to know what happens after the next review
  • Ebbinghaus has focused on the time or effort needed for re-learning, while I always was primarily interested in retention with little interest in how much time or effort is needed to bring the recall back to 100%

然而，克服这一概念障碍的解决办法是坚持学习。在不同的环境中，创造力的发展取决于时间和思维的投入。对我来说，艾宾浩斯之谜的答案很简单。像一个优秀的科学家一样，艾宾浩斯是一个理论完美主义者。他选择记忆无意义的音节，以减少他先前知识的干扰(https://supermemo.guru/wiki/Interference)。不幸的是，这有一些不好的副作用:

I will then venture a claim that in all his heroic memorization efforts, Hermann Ebbinghaus got fed up with learning nonsense. This would break the most persistent student. My better luck came from the burning need for getting good results in learning. This resulted in self-perpetuation of the effort. This is helpful in overcoming creative obstacles.

Interestingly, in 2015, Jaap Murre reproduced the original Ebbinghaus experiment with meticulousness worth Ebbinghaus himself (even digging into original manuscripts). Murre was prudent enough not to subject himself to the mental strain of nonsense syllables, which actually might have introduced some bias. He took the best subject available: a 22-year-old student who was rewarded with co-authorship of the publication. Incidentally, Murre seems to have been able to capture the circadian effect of learning in that his curve shows a minor bump after the period of 24 hours.

我敢说，在赫尔曼·艾宾浩斯(Hermann Ebbinghaus)所有英勇的记忆努力中，他已经厌倦了学习废话。这会使最顽固的学生崩溃。我的好运气来自于我迫切需要在学习上取得好成绩。这导致了这种努力的自我延续。这有助于克服创造性障碍。

有趣的是，在2015年，Jaap Murre转载默尔很谨慎，不让自己受到无意义音节的精神压力，因为这些音节实际上可能会带来一些偏见。他选择了最合适的题材:一位22岁的学生，他是这本书的合著者。顺便说一句，Murre似乎已经能够捕捉到学习的[昼夜节律]效应，他的曲线在24小时后显示出一个小的波动。

Saving on re-learning 节省重新学习

There is a huge difference between retrievability and "savings in re-learning". Depending on the timing, no retrieval may mean that the fact has "just been forgotten, and easily re-relearned" or "forgotten for good". Saving on re-learning may be non-zero on materials with zero retrievability. In simple terms, lack of access to memories does not imply zero stability. In that sense, Ebbinghaus curve is not even a good expression of exponential forgetting. There will be a different saving on re-learning for difficult material and for easy material.

可检索性和“节省再学习”之间存在巨大差异。根据时间的不同，不检索可能意味着“只是忘记了，很容易重新学习”或“永远忘记了”。对于可检索性为0的材料，重新学习的节省可能是非零(https://supermemo.guru/wiki/Retrievability)。简单来说，缺乏对内存的访问并不意味着不稳定(https://supermemo.guru/wiki/Stability)。从这个意义上说，艾宾浩斯曲线甚至不是指数遗忘的一个好的表达。对于难学的材料和容易学的材料将会有不同的节省。

Forgetting curve (1984) 遗忘曲线

While writing the chapter about forgetting curves, I thought that my own notion of the forgetting curve was erroneous at the beginning. However, while digging into my archives, by accident, I discovered that I had also plotted a forgetting curve for the retention of English vocabulary in 1984, just a few months before designing SuperMemo on paper:

在写关于遗忘曲线的章节的时候，我认为我自己对遗忘曲线的概念一开始是错误的。然而，在翻看我的档案时，我意外地发现，就在我设计SuperMemo的几个月之前，我也在1984年绘制了一条英语词汇记忆的遗忘曲线:

The very first forgetting curve for the retention of English vocabulary plotted back in 1984, just a few months before designing SuperMemo on paper

Figure: My very first forgetting curve for the retention of English vocabulary plotted back in 1984, i.e. a few months before designing SuperMemo on paper. This graph was not part of the experiment. It was simply a cumulative assessment of the results of intermittent learning of English vocabulary. The graph was soon forgotten. It was re-discovered 34 years later. After memorization, 49 pages of ~40 word pairs of English were reviewed at different intervals and the number of recall errors was recorded. After rejecting outliers and averaging, the curve appears to be far less steep that the curve obtained by Ebbinghaus (1885), in which he used nonsense syllables and a different measure of forgetting: saving on re-learning

***图:*我的第一个关于英语词汇记忆的遗忘曲线是在1984年绘制的，也就是说，在设计SuperMemo之前的几个月。这张图不是实验的一部分。它只是对间歇学习英语词汇的结果进行了简单的累积评估。图表很快就被遗忘了。它在34年后被重新发现。记忆后，以不同的时间间隔复习49页~40对英语单词，并记录记忆错误的数量。在剔除了异常值并进行平均后，曲线似乎远没有艾宾浩斯(1885)所得到的曲线那么陡峭。在艾宾浩斯的曲线中，他使用了无意义的音节和另一种衡量遗忘的方法:节省再学习的时间

I forgot about the fact of plotting that curve. I presume, once I had SuperMemo, that curve no longer seemed important or relevant. I was not focused on the "science of memory". I just wanted to get good results in learning. Apparently, I did not think that plotting a forgetting curve was a big deal. I nearly missed that figure in 2018 too as it was labelled in small print: average speed of forgetting for the first memorization.

The result came from 49 pages of 40-word pairs each, i.e. 1960 words (compare: 13 years of school in a month). I did all my learning for learning's sake, not for the experiment. There was no experiment actually. All I needed was to collect the numbers from my actual learning effort and plot the curve. The simplicity of the calculation might also explain why I forgot the exercise so easily.

I wrote about simple intuitions before the 1985 experiment. In the light of this 1984 curve, it might seem that the intuitions may have been derived from that little experiment. What today seems obvious might not be that obvious without the crutch of that little calculation. However, I can guess that by Feb 1985, I have already forgotten my curve because I did not skip Stage A in the spaced repetition experiment. My forgetting curve agreed with the experiment and the first review interval should be 1 day indeed.

The erroneous notion of sigmoidal curves must have been born later. Today, I seem to recall remotely that I originally thought that the sigmoidal nature kicks in only after the first review. I might never know how my thought process worked at that time. All I cared was the efficiency of SuperMemo. In the excitement of the discovery of SuperMemo, I forgot my curve for the entire 34 years. Today, that little graph matters only to illustrate how easily we can discover, forget, stray, re-discover, and then re-discover the original discovery. Memory is fallible. God bless spaced repetition.

我忘了画曲线了。我想，一旦我有了SuperMemo，这条曲线就不再显得重要或相关了。我没有专注于“记忆的科学”。我只是想在学习上取得好成绩。显然，我并不认为绘制遗忘曲线有什么大不了的。在2018年，我也差点错过了这个数字，因为它用小字标注着:第一次记忆的平均遗忘速度。

结果来自49页，每页40个单词对，即1960年的单词(比较:13年的学校在一个月)。我学习一切都是为了学习，而不是为了实验。实际上没有实验。我所需要做的就是从我实际的学习努力中收集数据并绘制曲线。计算的简单也可以解释为什么我这么容易就忘记了这个练习。

在1985年的实验之前，我写过关于简单直觉的文章。根据这条1984年的曲线，直觉似乎来自于那个小实验。如果没有这个小小的计算支撑，今天看起来显而易见的事情可能不会那么明显。然而，我可以猜测到1985年2月，我已经忘记了我的曲线，因为我没有跳过阶段A在间隔重复实验。我的遗忘曲线与实验一致，第一次复习的间隔确实应该是1天。

s形曲线的错误概念一定是后来才产生的。今天，我似乎还依稀记得，我最初以为只有在第一次回顾之后，s型曲线的性质才会出现。我可能永远不会知道我的思维过程是如何在那个时候运作的。我只关心SuperMemo的效率。在发现SuperMemo的兴奋中，我忘记了34年来我的曲线。今天，这个小图表只说明我们可以多么容易地发现、忘记、迷失、重新发现，然后再重新发现原来的发现。记忆是不可靠的。上帝保佑间隔重复。

Forgetting curve (Wozniak 1984)

Figure: The very first forgetting curve for the retention of English vocabulary plotted back in 1984 (a few months before designing SuperMemo on paper). This graph was not part of an experiment. It was simply a cumulative assessment of the results of intermittent learning of English vocabulary. The graph was forgotten and re-discovered 34 years later. After memorization, 49 pages of ~40 word pairs of English were reviewed at different intervals and the number of recall errors was recorded. White circles correspond with recall derived from the average number of errors per page after a given interval. Logarithmic regression in orange provides the best fit. Power regression in red follows closely. This could be expected from heterogeneous material (pages of words). This is also very similar to the results obtained by Ebbinghaus (1885), except the curve is far less steep as befits meaningful material. As expected, exponential regression in white provides the weakest fit

***图:*最早的英语词汇记忆遗忘曲线绘制于1984年(SuperMemo设计的几个月前)。这张图不是实验的一部分。它只是对间歇学习英语词汇的结果进行了简单的累积评估。这张图被遗忘了，34年后又被重新发现。记忆后，以不同的时间间隔复习49页~40对英语单词，并记录记忆错误的数量。白圈对应的是根据给定间隔后每页的平均错误数得出的召回率。橙色的对数回归是最合适的。红色的权力回归紧随其后。这可以从不同的材料(几页的单词)中预期。这与艾宾浩斯(1885)的研究结果非常相似，只是曲线没有那么陡峭，更适合有意义的材料。正如所料，白色的指数回归提供了最弱的拟合

Spitzer experiment (1939)

斯皮策的实验(1939)

In 1939, Herbert F. Spitzer investigated various testing schedules on 3605 sixth-grade pupils in 9 Iowa cities. This was the entire population of 91 schools. The kids read a 6-page text and their knowledge was later tested with 25 specific questions.

Spitzer designed different testing schedules, however, the design clearly did not aim at optimum spacing, and his research summary did not make any recommendation in that respect.

1939年，赫伯特·f·斯皮策(Herbert F. Spitzer)调查了爱荷华州9个城市的3605名六年级学生的各种考试时间表。这是91所学校的全部人口。孩子们读了一篇6页的课文，然后用25个具体问题测试他们的知识。

斯皮策设计了不同的测试时间表，然而，该设计显然没有以最佳间距为目标，他的研究总结也没有在这方面提出任何建议。

Spitzer (1939) Diagram of Procedure

Figure: Diagram of the testing procedure in Spitzer's experiment (1939). Kids were tested at varying intervals to see the effect of timing on results. Two groups received a testing intervention that could be used to draw some rudimentary conclusions on the effect of spaces on learning outcomes

***图:**斯皮策实验所采用的检测程序示意图(1939年)。对孩子们进行了不同时间间隔的测试，以观察计时对结果的影响。两组接受了“测试干预”，可以用来得出空间对学习结果影响的初步结论

Only the first two groups of students used an expanding schedule. These were the groups that came top on tests, however, Spitzer attributed this to the testing effect as these were the only two groups with 3 tests, i.e. the middle test was interpreted as a "testing intervention" (rather than a repetition).

只有前两组学生使用了扩展计划。这些团体位居榜首的测试,不过,斯皮策将此归因于测试效果,因为这些是唯一两组与3测试,即中间测试被解读为“测试干预”(而不是重复)。

Spitzer (1939) Curves of Retention

Figure: Curves of retention in the testing procedure in Spitzer's experiment (1939). The graph shows the benefits of active recall and the impact of review interval. The picture is not too revealing for only two groups of kids received two "repetitions" in the form of tests

***图:*斯皮策实验(1939)中测试程序的保留曲线。该图显示了主动召回的好处和审查间隔的影响。这幅图并不是太明显，因为只有两组孩子接受了两次“重复”的测试

In his recommendations, Spitzer focused on the power of the testing effect, of what I call active recall in reference to my flashcards of 1982. What I like about his paper is the recommendation that testing be used for students to correct their own knowledge, provide feedback, and thus give them a sense of their own progress. This stands in contrast to modern testing that is more often used as a whip to push kids to do more learning.

在他的建议,斯皮策的力量集中在测试效果(https://supermemo.guru/wiki/Testing_effect),我称之为[主动召回](https://supermemo.guru/wiki/Active_recall)在我的抽认卡[1982](https://supermemo.guru/wiki/Birthdate_of_SuperMemo)。我喜欢他的论文的地方在于，他建议学生通过考试来纠正自己的知识，提供反馈，从而让他们对自己的进步有个概念。这与现代的[测试](https://supermemo.guru/wiki/Testing)形成了鲜明的对比，现代的[测试]更多的是用来鞭笞孩子们去做更多的学习。

Spitzer's diagram of intervals reminds me of my own self-administered test that failed to produce convincing results. In Spitzer's case, intervals are clearly clocked by school life rather than by demands of memory.

In his research, Spitzer exhibited ruthless Gatesian approach to testing and learning, typical of early school innovators who often did more damage than good by being too thorough! Educators who do not employ the learn drive in the process, are doomed to fail even if they employ the best tools for boosting memory!

斯皮策的时间间隔图表让我想起了我自己的自我管理测试(https://supermemo.guru/wiki/History_of_spaced_repetition_(print)# wrong_实验者al_design_may_lead_to_wrong_conclusion)，它没有产生令人信服的结果。在斯皮策的例子中，时间间隔显然是由学校生活而不是由记忆需求决定的。

在他的研究中，斯皮策展示了对测试和学习的无情的Gatesian方法，这是典型的早期学校创新者，他们经常因为太过彻底而弊大于利!在这个过程中不使用learn drive的教育者注定要失败，即使他们使用最好的工具来提高记忆力!

I cannot but remark on the issue of ethical considerations. While we slaughter millions of animals in the name of science, we might perhaps fail to notice that this research, with all the best intent at heart, used the labor of 3000+ involuntary participants. What is great for memory research does not need to be great for individual learning. In the era of Google characterized by the plethora of exciting learning material, I found the test texts used by Spitzer pretty offputting. Despite the fact that a great deal of effort was put into the selection of the subject matter and wording, I would consider coercive reading a violation of my freedom. Those materials might be much better than your school average, but schooling will always be about mass production and less about creative freedom.

Reading about a biology of banana plants in Spitzer's materials might be palatable to a gardener or a biologist, which I claim to be. However, why would kids need to read about bananas if their current interest was the Great Depression, the Dust Bowl, Nazi Germany, or baseball?

School is the time when we stop learning and start cramming. It is the time when we lose the love of learning. Adding spaced repetition to this coercive mix could only make things worse

我不得不谈谈道德考虑的问题。当我们以科学的名义屠杀数以百万计的动物时，我们可能没有注意到这项研究，怀着最好的意图，使用了3000多名非自愿参与者的劳动。对记忆有好处的研究不一定对个人学习有好处。在以大量令人兴奋的学习材料为特征的谷歌时代，我发现斯皮策使用的测试文本相当令人不快。尽管在主题和措辞的选择上投入了大量的努力，但我认为强制阅读是对我自由的侵犯。这些材料可能比你们学校的平均水平要好得多，但学校教育总是关于大规模生产，而不是关于创作自由。

在斯皮策的材料中读到关于香蕉植物的生物学，可能会让园丁或生物学家感到愉快，而我自称是这样的。然而，如果孩子们当前的兴趣是大萧条、沙尘暴、纳粹德国或棒球，他们为什么还需要阅读有关香蕉的书籍呢?

学校是我们停止学习，开始填塞的时候(https://supermemo.guru/wiki/Cramming)。这是我们失去对学习的热爱的时候。在这种强制性的混合中加入[间隔重复](https://supermemo.guru/wiki/Spaced_repetition)只会让事情变得更糟

Wozniak experiment (1985)

沃兹尼亚克实验(1985)

Before I managed to compute optimal intervals for reviewing pages with English vocabulary, I designed an experiment which might have put me on wrong tracks. The experiment was to show the value of the progressive increase in review intervals. Instead, it showed that equally spaced review might be superior.

Experiments in science should not be considered "failed" just because they come with unexpected results. However, in this case, unexpected results came from bad design, and might have resulted in confusion that would inhibit further progress. To this day researchers speak of "mixed results" in spaced repetition due to wrong design or even simple terminological confusion.

Fortunately, my intuitions about memory in 1985 were too solid, and it was pretty obvious that with rapid acceleration in the expanding length of intervals, forgetting might overwhelm consolidation. My progressive schedule was simply not optimum. I would have certainly designed a better follow-up. Most of all, I wanted to continue learning with success. I doubt I would have given up without a good solution.

The experiment shows that it is not enough to have a good guess about what optimum intervals are. The intervals actually need to be computed.

在我为了复习英语词汇而计算出最佳间隔之前，我设计了一个实验，这个实验可能会让我误入歧途。这个实验是为了显示复习间隔的递增值。相反，它表明等间隔的复习可能更好。

科学实验不应该仅仅因为结果出人意料就被认为是“失败的”。然而，在这种情况下，意外的结果来自于糟糕的设计，并可能导致混乱，从而阻碍进一步的进展。时至今日，研究人员仍在谈论间隔重复由于错误的设计甚至简单的术语混淆导致的“混合结果”。

幸运的是，我在1985年对记忆的直觉太过坚定，很明显，随着时间间隔的快速增长，遗忘可能会压倒巩固。我的进度表根本不是最优的。我肯定会设计一个更好的后续行动。最重要的是，我想继续成功地学习。我怀疑如果没有一个好的解决方案，我是不会放弃的。

实验表明，仅仅猜测最佳间隔是不够的。实际上需要计算这些间隔。

Archive warning: Why use literal archives?

This text was part of: "Optimization of learning" by Piotr Wozniak (1990)

At the beginning of 1985, I designed two experiments which consequently revolutionized my learning methodology and led to the formulation of the SuperMemo method.

The first experiment can be a good illustration of how a misconceived idea can yield valuable conclusions (the second is described here). It is a common intuition that with successive repetitions, knowledge should gradually become more durable and require less frequent review. Thus repetitions separated by increasing intervals should be more effective than those whose intervals are always the same. This belief proved to be false. Let us see an experiment which demonstrates the fact:

档案警告:为什么使用文字档案?

本文是[Piotr Wozniak]的《优化学习》(https://supermemo.guru/wiki/Piotr_Wozniak)(1990)的一部分。

1985年初，我设计了两个实验，这两个实验彻底改变了我的学习方法，并最终形成了SuperMemo method。

第一个实验可以很好地说明一个错误的想法是如何产生有价值的结论的(第二个实验的描述这里)。一种常见的直觉是，随着连续的重复，知识应该逐渐变得更持久，需要更少的频繁审查。因此，通过增加间隔来分隔重复应该比那些间隔总是相同的更有效。这个信念被证明是错误的。让我们看一个实验来证明这个事实:

Experiment on the influence of various repetition spacing patterns on the effect of repetitions on knowledge retention (Jan 31, 1985 - Aug 2, 1986)

不同重复间隔模式对重复对知识保留效果影响的实验(1985年1月31日- 1986年8月2日)

  1. The memorized knowledge consisted of 195 items divided into three equal groups: A, B and C.

  2. Each of the items had the following form:

  3. All items of a particular group were memorized in one session by repeating them until all were known (group A - Jan 31, B - Feb 2, C - Feb 3).

  4. Two final control dates were established: Dec 6-7, 1985 and Aug 1-2, 1986 on which the level of retention in all groups was measured at the same time (for each item the accuracy of recall was measured in a four grade scale).

  5. Before the control dates, all of the groups A, B and C underwent 6 independent repetitions in the following intervals (expressed in days):

  6. 记忆的知识由195个项目组成，分为三组:A、B和C。

  7. 每一项的形式如下:

  8. 一个小组的所有项目都是通过重复记忆，直到所有的项目都被记住(a组- 1月31日，B组- 2月2日，C组- 2月3日)。

  9. 最后确定了两个对照日期:1985年12月6-7日和1986年8月1-2日，在这两个日期上同时测量所有组的保留水平(对于每个项目，召回的准确性采用四级量表测量)。

  10. 对照组开始前，A、B、C三组均进行6次独立重复，重复时间间隔如下(以天为单位):

Repetition number Group A (equal spacing over a longer time) Group B (spacing based on increasing intervals) Group C (equal spacing in 30 days)
1 18 days 1 day 5 days
2 18 days 5 days 5 days
3 18 days 9 days 5 days
4 18 days 24 days 5 days
5 18 days 44 days 5 days
6 18 days 70 days 5 days

The intent of the experiment was to prove that increasing intervals are the best for memory consolidation (group B) as opposed to intervals that are evenly distributed (group A) or concentrated in time (group C). The results of the experiment are shown in Fig. 3.1:

本实验的目的是为了证明，相对于间隔时间均匀分布(A组)或集中分布(C组)，间隔时间递增的记忆巩固效果最好(B组)，实验结果如图3.1所示:

Condensed vs distributed vs progressive schedule

Figure: (3.1) Experiment on the impact of massed, distributed, and progressive schedules on retention. Due to a poor choice of intervals, this experiment seems to show the superiority of long equal-length intervals over a progressive schedule used in spaced repetition (Piotr Wozniak, 1985-1986)

图:(3.1)试验了批量、分布式和渐进计划对保留率的影响。由于音程选择不当，本实验似乎显示了等长音程比渐进式间隔重复的优越性(Piotr Wozniak, 1985-1986)

The results of the second control are not present on the graph because of the prosaic fact that studying at the University of Technology [started in October 1985] required perfect knowledge of irregular verbs, therefore another measurement was pointless.

As it can be seen in Fig.3.1 above, the experiment yielded unexpected results proving that increasing . The principle of using such intervals in the process of learning will be later be referred to as theoptimum repetition spacing principle

While writing History of spaced repetition, I found the original graphs in my archive. It is pretty obvious that massed practice is not a good approach:

第二次对照的结果没有出现在图表上，因为在理工大学[1985年10月开始]学习需要对不规则动词的完美知识，所以另一个测量是没有意义的。

从图3.1可以看出，实验的结果出乎意料的证明了这一点。在学习过程中使用这种间隔的原则将在后面被称为最佳重复间隔原则

在写《间隔重复的历史》(https://supermemo.guru/wiki/History_of_spaced_repetition)的时候，我在我的档案里找到了原始的图表。很明显，“大量练习”不是一个好方法:

imgGroup A: distributed spacing in 18 day intervals imgGroup B: spacing based on increasing intervals imgGroup C: massed spacing in 5 day intervals

Why spaced repetition idea succeeded in the end?

为什么间隔重复的想法最后成功了?

Spaced repetition is intuitive. After some reflection, it is even obvious. Why didn't we come up with solutions earlier? SuperMemo on paper would be feasible as soon as mankind came up with paper. However, in ancient times, the volume of vital knowledge was small enough for the brain to easily cope with the use of natural learning methods. By working in the field, the farmer quickly gains all necessary expertise that helps him excel for as long as health permits.

Things started changing with the arrival of print in the 15th century. Knowledge started proliferating and it also gained means of efficient perpetuation. Newton at plague years might be an interesting example of an individual who would theoretically benefit from spaced repetition. However, Newton's book collection was limited and the problems to think about and solve were so numerous that it is easy to imagine he would never be bothered with his failing memory. All he needed was to make notes.

Over the next two centuries, the stores of available knowledge kept increasing and human natural appetite for remembering might have been increasing at the same time. Research by Ebbinghaus in the 1880s is an example of a continual interest in the workings of memory. However, even today, most of the scientists are rarely bothered by their forgetting. Making notes and Google can satisfy most of the needs for most of them. When Vannevar Bush conceived memex device in the 1930s, he saw it as a memory augmentation. However, even if memex was to be seen as a distant cousin of incremental reading or neural creativity, all its knowledge, like in Google, lived predominantly outside of the human brain. My quest reminds me that of V. Bush except I want to see all that knowledge make a direct imprint on human creativity (see: neural creativity).

间隔重复是直观的。经过一番思考，它甚至是显而易见的。我们为什么不早点想出解决办法呢?一旦人类发明了纸，“纸上的超级备忘录”(http://super-memory.com/articles/paper.htm)就将成为可能。然而，在古代，至关重要的知识的容量非常小，以至于大脑可以很容易地使用[自然学习方法](https://supermemo.guru/wiki/Free_learning)。通过在田间工作，农民很快获得了所有必要的专业知识，帮助他在健康允许的情况下脱颖而出。

随着15世纪印刷术的出现，情况开始发生变化。知识开始扩散，也获得了有效的延续手段。牛顿在瘟疫年可能是一个有趣的例子，一个人谁将从理论上受益于间隔重复。然而，牛顿的藏书是有限的，要思考和解决的问题是如此之多，很容易想象他不会为他的记忆力衰退而烦恼。他所需要的只是做笔记。

在接下来的两个世纪里，可获得的知识储备不断增加，人类对记忆的自然欲望可能也在同时增加。19世纪80年代艾宾浩斯的研究就是对记忆运作方式持续感兴趣的一个例子。然而，即使在今天，大多数科学家也很少被他们的遗忘所困扰。做笔记和谷歌可以满足大部分人的需要。上世纪30年代，当瓦内瓦尔•布什(Vannevar Bush)构想出memex设备时，他将其视为一种记忆增强。然而，即使memex被视为渐进式阅读或神经创造力的远祖，它的所有知识，就像在谷歌，主要生活在人类大脑之外。我的探索让我想起了v•布什(V. Bush)，但我希望看到所有这些知识都能直接影响人类的创造力(参见:神经创造力)。

Why was I particularly bothered by forgetting? There are many kids out there who like to show off at school. When my brother showed me how to differentiate needles of a spruce and a fir, in my 3rd-grade biology class, I was eager to show that I know more than your average kid. That eagerness was quickly suppressed by schooling, but some traces survived. When we started learning chemistry in Grade 7, I was proud to be able to recall complex names like adenosine triphosphate or deoxyribonucleic acid. I was even looking up difficult chemistry names to memorize and impress. I memorized the details of the anatomy of a jawless fish. Again I tried to impress at school, but no one was interested.

When I got deeper into zoology in Grade 5, some kind of OCD for information started showing up. I wanted to know all animals in the Poznan zoo, their habitat, and even their taxonomy and Latin names. I got fantastic books with photographs of all animals and knowing them seemed like an interesting endeavour.

In 1974, I became interested in boxing, football, and sports in general. I used to visit the waste paper store in our school and search for sports journals. My colleague Robert asked me to collect pictures with naked ladies. I complied. My puberty was late and I was not interested. When the stack of photos spilled at home, I explained it to my mom "not for me!". She nodded compassionately with a smirk. In the meantime, I would use an iron to remove wrinkles from my recycled sports journal collection. I would stack them up neatly and archive by date. The stack filled up several shelves in bookcases in my room. That obsessive neatness would make for a disturbing picture for any parent, however, my mom was tolerant. I was free to follow my passions even if they seemed preposterous. I have no doubt that freedom is one of the vital ingredients of healthy development. Don't many kids show similar symptoms in obsessions with stamp collecting, postcards, and football cards? Pity most parents are not as tolerant as my mom, and child's passions are easily extinguished.

为什么我特别讨厌忘记?有很多孩子喜欢在学校炫耀。当我哥哥在我三年级的生物课上教我如何分辨云杉和冷杉的针时，我很想表现出我比一般孩子懂得多。这种渴望很快就被学校教育压制住了(https://supermemo.guru/wiki/Schools_suppress_the_learn_drive)，但一些痕迹仍然存在。当我们在七年级开始学习化学时，我很自豪能够回忆起复杂的名字，如三磷酸腺苷或脱氧核糖核酸。我甚至还查了一些很难记的化学名字，想要给他们留下深刻印象。我记住了一条没有下巴的鱼的解剖结构的细节。我又一次试图在学校给人留下深刻印象，但没有人感兴趣。

当我在五年级深入学习动物学时，某种强迫症开始出现。我想知道波兹南动物园里所有的动物，它们的栖息地，甚至它们的分类学和拉丁名。我有很棒的书，里面有所有动物的照片，了解它们似乎是一种有趣的尝试。

1974年，我开始对拳击、足球和体育产生兴趣。我过去常去我们学校的废纸店找体育杂志。我的同事罗伯特让我收集裸体女士的照片。我照做了。我的青春期很晚了，我也不感兴趣。当那堆照片散落在家里时，我向妈妈解释道:“不适合我!”她笑着同情地点了点头。与此同时，我会用熨斗熨平我回收的体育杂志上的皱纹。我会把它们整齐地堆放起来，按日期归档。我房间里的书架上堆满了书。那种执著的整洁会让任何家长感到不安，然而，我的母亲是宽容的。我可以自由地追随自己的激情，即使它们看起来很荒谬。我毫不怀疑，自由是健康发展的重要组成部分之一。难道很多孩子在痴迷于集邮、明信片和足球卡片时没有表现出类似的症状吗?遗憾的是，大多数父母都不像我母亲那样宽容，而且孩子的激情很容易熄灭。

Some early drive in the direction of incremental reading showed some time between ages 10 and 12. I started making notes about horses by following links in an encyclopedia (paper book). It was fun to start with one entry, and hope to write a booklet about horses by just following the links. In Grade 1 of high school (aged 12), I did a similar exercise with an encyclopedia of mammals. There was some inner need to codify all information about certain subjects in biology. I decided to write an encyclopedia-sized book about the living world. I used a typewriter at mom's work and my own rich illustrations to write the history of evolution. I never finished. The futility of that exercise should be obvious. It never bothered my mind. The effort was enjoyable and addictive.

I showed a similar informavorous nature when I started learning zoology, chemistry, biology, and later biochemistry during summer vacations 1977-1979 (aged 13-15). This was a definite start of my growing need for remembering. I would produce calligraphically written and meticulously illustrated books with my newly acquired knowledge. However, after a longer while that knowledge tended to evaporate. That futility started bothering me slowly. What's the point of writing books if soon they turn into the same dead material like all other books on the shelf?

Around the year 1981-1982 (aged 19-20), I started writing down my knowledge in the form of questions and answers for active recall tests. I knew that this was the only way to preserve things in memory for longer. Needless to say, after a while, the chaotic review did not provide full satisfaction either. At that stage, I was clearly very hungry for knowledge. I wanted to remember. Simple intuitions led to simple experiments and then to a simple SuperMemo for DOS. I was driven by practical applications that combined research with the pleasure of learning.

一些早期的增量阅读显示在10到12岁之间。我开始在一本百科全书(纸质书)的链接上做关于马的笔记。以一个条目开始是有趣的，并且希望通过跟随链接来写一本关于马的小册子。在高一(12岁)，我做了一个类似的实验，用的是一本哺乳动物百科全书。有一种内在的需要把生物学中有关某些学科的所有信息编成法典。我决定写一本关于生活世界的百科全书大小的书。我在妈妈工作的地方用打字机和我自己的丰富插图来书写进化史。我从来没有完成。这种做法的徒劳无功是显而易见的。我从不介意。这种努力是令人愉快和上瘾的。

当我在1977-1979年的暑假(13-15岁)开始学习动物学、化学、生物学以及后来的生物化学时，我也表现出了类似的信息丰富的天性。这确实是我不断增长的记忆需求的开始。我将用我新学到的知识制作书法书和精心配图的书籍。然而，过了一段时间，这种知识就会消失。这种徒劳开始慢慢地困扰着我。如果书很快就像书架上所有其他的书一样变成了同样的死材料，那么写书还有什么意义呢?

大约在1981-1982年(19-20岁)，我开始在主动回忆测试的问题和答案中写下我的知识。我知道这是唯一能让记忆更长久的方法。不用说，过了一段时间，混乱的评论也没有让人完全满意。在那个阶段，我显然非常渴望知识。我想记住。简单的直觉导致了简单的实验，然后产生了一个简单的超级备忘录DOS。我被结合了研究和[学习的乐趣]的实际应用所驱动(https://supermemo.guru/wiki/Pleasure_of_learning)。

As of that point, my path became deterministic. I loved learning too much to give up SuperMemo in difficult times. My love of learning was easily suppressed at school or when I was involved in running a business. However, once I decided to work from home (1997), the positive feedback loop has set in between learning, love of learning, and the progress of SuperMemo. As this also happens to be the way to earn a living, that loop is for ever. Only bad health or death can break this cycle. That part of the formula for progress is easy to figure out.

What could parents do to facilitate similar passions in kids? I think two ingredients come into play: inspiration and freedom. While inspiration is helpful, freedom is essential. These days, inspiration can be found all around. YouTube plays a huge role where humans don't live up to the mark. All around the world, it is the freedom ingredient that is missing most. Kids are enslaved by schooling. They are also enslaved by authoritarian parenting. I was told many times that some of my crazy behaviors are a side effect of growing up without a father. This might be true. However, to compensate, I had three parents: my mom and two older siblings: inspiring brother and a childless sister who channeled all her love into the little brat of me. I venture a claim that I came to this world with a reasonable endowment, not much above the average. I often lingered at the bottom of the class. Rarely would I come to the lead. I have dozens of examples when I converted meager skills or a bit of talent into excellence in a narrow area of creative pursuits. Freedom and rage to master were always the key ingredients. When I bread dozens of animals or collected dead body parts, my family might ask "why is the kid not going to church instead?". My mom would shield me from those influences and concerns. I could do things my own way.

从那一刻起，我的人生道路就变得确定了。在困难时期，我太喜欢学习了，以至于无法放弃SuperMemo。我对学习的热爱很容易在学校或我参与经营企业时被抑制。然而，一旦我决定在家工作(1997)，在学习、热爱学习和SuperMemo的进步之间就形成了正反馈循环。因为这恰好也是谋生的方式，所以这个循环是永远的。只有糟糕的健康或死亡才能打破这个循环。进步公式的这一部分很容易计算出来。

父母能做些什么来激发孩子们的类似热情呢?我认为有两个因素在起作用:灵感和自由。虽然灵感是有用的，但自由是必不可少的。如今，灵感无处不在。在人类没有达到标准的地方，YouTube扮演了一个巨大的角色。在世界各地，最缺乏的是自由的成分。孩子们受学校教育的奴役。他们也被专制的父母所奴役。我被告知很多次，我的一些疯狂行为是在没有父亲的环境下长大的副作用。这可能是真的。然而，作为补偿，我有三个父母:我的母亲和两个年长的兄弟姐妹:鼓舞人心的哥哥和一个没有孩子的妹妹，她把所有的爱都倾注到我这个小淘气身上。我敢断言，我来到这个世界的时候，禀赋还算合理，并不比一般人高多少。我经常在班上垫底。我很少带头发言。我有几十个这样的例子，当我在一个狭窄的创造性追求领域里，把贫乏的技能或一点才华转化成卓越。自由和对主人的愤怒一直是关键的成分。当我给几十只动物做面包或者收集尸体的时候，我的家人可能会问“为什么孩子不去教堂呢?”我妈妈会让我远离那些影响和担忧。我可以用自己的方式做事。

My development was then primarily boosted by the impact of freedom. Freedom helps the learn drive thrive, and the learn drive has self-amplifying powers. This led me to obsessions that make it possible to boast today of fathering spaced repetition. Passions and obsessions should be cherished. Instead, we tend to exterminate them in the name of robotization of development. We set benchmarks and ask kids to jump through the hoops. As if being led by the hand through college has ever contributed to anyone's independent thinking. This decorticating process needs to stop! The future belongs to free learning.

As for spaced repetition, the appetite for solving the problem of forgetting increased substantially with the arrival of personal computers. In the early 1990s, I heard it with increasing frequency from users of SuperMemo that they thought of a similar solution for themselves but decided to look around for a specific software application first.

我的发展主要是受到自由的影响。自由有助于学习动力茁壮成长，而学习动力具有自我放大的能力。这让我产生了一些痴迷，让我有可能在今天夸耀自己是[间隔重复]的父亲(https://supermemo.guru/wiki/Spaced_repetition)。我们应该珍惜激情和执着。相反，我们倾向于以发展自动化的名义消灭它们。我们设定标准，让孩子们去完成。好像在大学里被人牵着鼻子走对每个人的独立思考都有帮助。这个去皮过程必须停止!未来属于[免费学习](https://supermemo.guru/wiki/Free_learning)。

至于“间隔重复”，随着个人电脑的出现，人们解决遗忘问题的欲望大大增强。上世纪90年代初，我越来越频繁地听到SuperMemo的用户说，他们也想到了类似的解决方案，但决定先寻找一个特定的软件应用程序。

It is the demands of school that cause most frustrations in students today, however, schools are not a good place for creative thinking. Students do flock to spaced repetition mostly because it is served on a silver platter. It is ready to consume. Very few contemplate their own solutions.

In addition to personal freedoms, some of the good fortunes that gave a good start to SuperMemo were: (1) free university in communist Poland, (2) family sponsorship in purchasing an outrageously expensive PC, (3) lenient schools that did not impose much on my free time, etc.

You will not see many 22-24 year olds tinkering with their own science at the cost of the state or their family. There is nothing wrong with young adults living with their parents if they pursue creative goals.

Spaced repetition was born at the confluence of good forces: free learning in a communist system, the transition from communism to the market economy, and the arrival of personal computing. Those global processes were boosted by freedom at home, and a dose of healthy obsession with learning. The main ingredient of that equation is freedom. Freedom is replicable. All it needs is to be granted.

正是学校的需求导致了今天学生的大多数挫折，然而，学校不是一个进行创造性思维的好地方。学生们会蜂拥去参加间隔重复学习，主要是因为它是放在银盘子里的。它已经准备好消费了。很少有人考虑自己的解决方案。

除了个人自由之外，SuperMemo给我带来的一些好运还包括:(1)共产主义波兰的免费大学;(2)家庭资助购买贵得离谱的个人电脑;

你不会看到许多22-24岁的年轻人以国家或家庭为代价来修补他们自己的科学。如果年轻人追求有创意的目标，和父母住在一起并没有什么不对。

间隔重复诞生于各种力量的汇合:共产主义制度下的免费学习、共产主义向市场经济的过渡，以及个人电脑的到来。这些全球进程是由国内的自由和健康的[对学习的痴迷]推动的(https://supermemo.guru/wiki/Addiction_to_learning)。这个等式的主要成分是自由。自由是可以复制的。它所需要的只是被给予。

First decade of SuperMemo: Battling skepticism

SuperMemo的第一个十年:与怀疑论作斗争

Skepticism surrounded the early days of SuperMemo in Poland. This was expressed pretty accurately in the Polish computer journal Enter in 1994 (see full article):

人们对波兰早期的超级备忘录持怀疑态度。这在1994年的波兰计算机杂志Enter上得到了相当准确的表述(参见全文):

SuperMemo might work but it cannot be that good If one is convinced of the validity of what has been said about SuperMemo, will he or she be already convinced that the program is a perfect cure for the ailing memory? Can it really capitalize on the properties of the nervous system and let learning proceed a dozen times faster than in standard circumstances? After all there have been generations of students trying to figure out better methods of learning, and a breakthrough comparable with SuperMemo seems highly unlikely even to quite an open-minded observer. Wozniak discounts the low-probability argument as the viable source of skepticism, and says that he has more than once traced down evidence that SuperMemo-like approaches to learning have already been tried before with lesser or greater degree of success. Moreover, it is worth noticing that SuperMemo might not see the light were it not implemented as a computer program which can easily be transferred between individuals. In other words, it could have fallen into oblivion as the previous attempts to put order in the process of learning. One must remember that the skeletal algorithm of SuperMemo has been formulated in 1985, and only 1987 saw its very slow expansion in selected scientific circles in Poznan. Another turning- point to be kept in minds is that SuperMemo World would not have been founded in 1991 were it not for the inspiring meeting of minds between Wozniak and his colleague from the university, Krzysztof Biedalak, currently SuperMemo World's Vice-President. Both top-students at the university, Wozniak intended to study neuroscience in the US, Biedalak wanted to do the same in the field of artificial intelligence. Only by coincidence, they were both thrown into the world of entrepreneurial science. All this shows that despite the fact that the principles of SuperMemo are extremely simple and might have been invented several dozen times independently in several dozen countries of the planet, SuperMemo is not just a run-of-the-mill. The distinctive merit of SuperMemo World was to put the idea in practice, invest a great deal of man-hours in the development of the software, and focus on marketing the idea to the potential customer. Otherwise, SuperMemo would have forever remained limited to the small circle of its early enthusiasts.

2796/5000
SuperMemo可能有用，但没有那么好

如果一个人相信关于SuperMemo的说法是正确的，那么他或她是否已经相信这个程序是治疗记忆问题的完美方法?它真的能利用神经系统的特性，让学习的速度比标准情况下快十几倍吗?毕竟，一代又一代的学生都在试图找出更好的学习方法，即使对一个思想相当开放的观察者来说，与SuperMemo类似的突破也几乎不可能。沃兹尼亚克(https://supermemo.guru/wiki/Piotr_Wozniak)认为低概率的论证是怀疑论的可行来源，并说他已经不止一次地找到证据，证明类似于超级记忆的学习方法已经在以前的试验中获得了或多或少的成功。此外，值得注意的是，如果SuperMemo不是作为一个可以在个人之间轻松传输的计算机程序来实现的，那么它可能看不到光。换句话说，它可能已经被遗忘了，就像之前试图在学习过程中建立秩序一样。我们必须记住，SuperMemo的骨架算法是在[1985](https://supermemo.guru/wiki/Birth_of_SuperMemo)中形成的，只有在[1987](https://supermemo.guru/wiki/Algorithm_SM-2)中，它在波兹南特定的科学圈子中才得到了非常缓慢的扩展。另一个转弯点保持在思想(SuperMemo世界)(https://supermemo.guru/wiki/SuperMemo_World)就不会(成立于1991年)(https://supermemo.guru/wiki/Employing_forgetting_curves_in_spaced_repetition_(1991))之间如果没有思想的鼓舞人心的会议(沃兹尼亚克)(https://supermemo.guru/wiki/Piotr_Wozniak)和他的同事从大学[Krzysztof Biedalak) (https://supermemo.guru/wiki/Krzysztof_Biedalak),目前SuperMemo全球副总裁。沃兹尼亚克和比达拉克都是这所大学的尖子学生，他们打算去美国学习神经科学，比达拉克也想在人工智能领域做同样的事情。巧合的是，他们都被扔进了创业科学的世界。所有这些都表明，尽管SuperMemo的原理非常简单，而且可能在地球上的几十个国家独立发明了几十次，但SuperMemo并不是一个普通的东西。[SuperMemo World](https://supermemo.guru/wiki/SuperMemo_World)的独特之处在于将想法付诸实践，在软件开发上投入大量人力，并专注于向潜在客户推销想法。否则，SuperMemo将永远局限于其早期爱好者的小圈子。

The future is bright 未来是光明的

I mentioned that the progress of spaced repetition has been bogged down by guesswork, weak practical toolset, and terminological confusion. Today, all those factors fall by the roadside. New web applications come into force, they combine big data, and practical needs of users. They all use the same terminology: forgetting curve, spacing effect, and spaced repetition. The future of learning looks brighter than ever.

我提到过间隔重复的进展已经陷入了猜测，薄弱的实用工具集，和术语混淆。今天，所有这些因素都落在了路边。新的web应用程序开始生效，它们结合了大数据和用户的实际需求。它们都使用相同的术语:遗忘曲线、间隔效应和间隔重复。学习的未来比以往任何时候都更加光明。