READING HEURISTICS
1. TRUST RECOMMENDATIONS — BUT NOT TOO MUCH
2. TAME THE THRILLERS
3. BLEND A BIZARRE BOWL
4. TRUST THE LINDY EFFECT
5. FAVOR BIOGRAPHIES OVER SELF-HELP

Most middle-class Americans at least act as if:
- Exactly four years of higher education is precisely the right level of training for the overwhelming majority of good careers.
- You should spend most of your waking hours most days of the week for the previous twelve+ years preparing for those four years. In your free time, be sure to do the kinds of things guidance counselors think are impressive; we as a society know that these people are the best arbiters of arete.
- Forty hours per week is exactly how long it takes to be reasonably successful in most jobs.
- On the margin, the cost of paying for money management exceeds the cost of adverse selection from not paying for it.
- You will definitely learn important information about someone’s spousal qualifications in years two through five of dating them.
-Human beings need about 50% more square feet per capita than they did a generation or two ago, and you should probably buy rather than rent it.
- Books are very boring, but TV is interesting.

All of these sound kind of dumb when you write them out. Even if they’re arguably true, you’d expect a good argument. You can be a low-risk contrarian by just picking a handful of these, articulating an alternative — either a way to get 80% of the benefit at 20% of the cost, or a way to pay a higher cost to get massively more benefits — and then living it.[1]

Here’s a LessWrong post that describes it (including the insight “I honestly didn’t think Zettelkasten sounded like a good idea before I tried it” which I also felt).

yeah doesn't sound like a good idea to me either. idk

It might sound a bit stupid, but I just realized that a better reading strategy could help me learn faster, almost three times as fast as before.

To enter a research field, we sometimes have to read tens of research papers. We could alternatively read summaries like textbooks and survey papers, which are generally more comprehensive and more friendly for non-experts. But some fields don’t have good summaries out there, for reasons like the fields being too new, too narrow, or too broad.

...

Part 1. Taking good notes and keeping them organized.

Where we store information greatly affects how we access it. If we can always easily find some information — from Google or our own notes — then we can pick it up quickly, even after forgetting it. This observation can make us smarter.

Let’s do the same when reading papers. Now I keep searchable notes as follows:
- For every topic, create a document that contains the notes for all papers on this topic.[1]
- For each paper, take these notes: summaries, quotes, and sufficient bibliographic information for future lookup.[2, pages 95-99]
- When reading a new paper, if it cites a paper that I have already read, review the notes for the cited paper. Update the notes as needed.
This way, we won’t lose what we have read and learned.

Part 2. Skipping technical sections for 93% of the time.

Only 7% of readers of a paper will read its technical sections.[1] Thus, if we want to read like average, it might make sense to skip technical sections for roughly 93% of papers that we read. For example, consider reading each paper like this:
- Read only the big-picture sections — abstract, introduction, and conclusion;
- Scan the technical sections — figures, tables, and the first and the last paragraphs for each section[2, pages 76-77] — to check surprises;
- Take notes;
- Done!
In theory, the only 7% of the papers that we need to read carefully would be those that we really have to know well.

Skim notes could never work on the iPhone, because SKim notes data depend on AppKit, which is not available in iOS. So any app for iOS would just be some comletely separate PDF app, that has nothing to do with Skim in particular.

This page describes human error rates in a variety of contexts.

Most of the error rates are for mechanical errors. A good general figure for mechanical error rates appears to be about 0.5%.

Of course the denominator differs across studies. However only fairly simple actions are used in the denominator.

The Klemmer and Snyder study shows that much lower error rates are possible--in this case for people whose job consisted almost entirely of data entry.

The error rate for more complex logic errors is about 5%, based primarily on data on other pages, especially the program development page.

https://twitter.com/Scholars_Stage/status/1199702832728948737
https://archive.is/cc4zf
I reviewed today my catalogue of 420~ books I have read over the last six years and I am in despair. There are probably 100~ whose contents I can tell you almost nothing about—nothing noteworthy anyway.

Figure 2.

English (IREN = 1.08) shows a higher Information Rate than Vietnamese (IRVI = 1). On the contrary, Japanese exhibits the lowest IRL value of the sample. Moreover, one can observe that several languages may reach very close IRL with different encoding strategies: Spanish is characterized by a fast rate of low-density syllables while Mandarin exhibits a 34% slower syllabic rate with syllables ‘denser’ by a factor of 49%. Finally, their Information Rates differ only by 4%.

Is spoken English more efficient than other languages?: https://linguistics.stackexchange.com/questions/2550/is-spoken-english-more-efficient-than-other-languages
As a translator, I can assure you that English is no more efficient than other languages.
--
[some comments on a different answer:]
Russian, when spoken, is somewhat less efficient than English, and that is for sure. No one who has ever worked as an interpreter can deny it. You can convey somewhat more information in English than in Russian within an hour. The English language is not constrained by the rigid case and gender systems of the Russian language, which somewhat reduce the information density of the Russian language. The rules of the Russian language force the speaker to incorporate sometimes unnecessary details in his speech, which can be problematic for interpreters – user74809 Nov 12 '18 at 12:48
But in writing, though, I do think that Russian is somewhat superior. However, when it comes to common daily speech, I do not think that anyone can claim that English is less efficient than Russian. As a matter of fact, I also find Russian to be somewhat more mentally taxing than English when interpreting. I mean, anyone who has lived in the world of Russian and then moved to the world of English is certain to notice that English is somewhat more efficient in everyday life. It is not a night-and-day difference, but it is certainly noticeable. – user74809 Nov 12 '18 at 13:01
...
By the way, I am not knocking Russian. I love Russian, it is my mother tongue and the only language, in which I sound like a native speaker. I mean, I still have a pretty thick Russian accent. I am not losing it anytime soon, if ever. But like I said, living in both worlds, the Moscow world and the Washington D.C. world, I do notice that English is objectively more efficient, even if I am myself not as efficient in it as most other people. – user74809 Nov 12 '18 at 13:40

Do most languages need more space than English?: https://english.stackexchange.com/questions/2998/do-most-languages-need-more-space-than-english
Speaking as a translator, I can share a few rules of thumb that are popular in our profession:
- Hebrew texts are usually shorter than their English equivalents by approximately 1/3. To a large extent, that can be attributed to cheating, what with no vowels and all.
- Spanish, Portuguese and French (I guess we can just settle on Romance) texts are longer than their English counterparts by about 1/5 to 1/4.
- Scandinavian languages are pretty much on par with English. Swedish is a tiny bit more compact.
- Whether or not Russian (and by extension, Ukrainian and Belorussian) is more compact than English is subject to heated debate, and if you ask five people, you'll be presented with six different opinions. However, everybody seems to agree that the difference is just a couple percent, be it this way or the other.

--

A point of reference from the website I maintain. The files where we store the translations have the following sizes:

English: 200k
Portuguese: 208k
Spanish: 209k
German: 219k
And the translations are out of date. That is, there are strings in the English file that aren't yet in the other files.

For Chinese, the situation is a bit different because the character encoding comes into play. Chinese text will have shorter strings, because most words are one or two characters, but each character takes 3–4 bytes (for UTF-8 encoding), so each word is 3–12 bytes long on average. So visually the text takes less space but in terms of the information exchanged it uses more space. This Language Log post suggests that if you account for the encoding and remove redundancy in the data using compression you find that English is slightly more efficient than Chinese.

Is English more efficient than Chinese after all?: https://languagelog.ldc.upenn.edu/nll/?p=93
[Executive summary: Who knows?]

This follows up on a series of earlier posts about the comparative efficiency — in terms of text size — of different languages ("One world, how many bytes?", 8/5/2005; "Comparing communication efficiency across languages", 4/4/2008; "Mailbag: comparative communication efficiency", 4/5/2008). Hinrich Schütze wrote:

(This story would suggest that in physics students are maybe missing out on learning the styles of thought that produce progress in physics. My guess is that instead they learn them in grad school when they are doing research themselves, by emulating their supervisors, and that the helpfulness of this might partially explain why Nobel prizewinner advisors beget Nobel prizewinner students.)

The story I hear about philosophy—and I actually don’t know how much it is true—is that as bits of philosophy come to have any methodological tools other than ‘think about it’, they break off and become their own sciences. So this would explain philosophy’s lone status in studying old thinkers rather than impersonal methods—philosophy is the lone ur-discipline without impersonal methods but thinking.

This suggests a research project: try summarizing what Aristotle is doing rather than Aristotle’s views. Then write a nice short textbook about it.

Reading speed and comprehension interest me, but I don’t have as much information as I would like.  I would like to see the distribution of reading speeds ( in the general population, and also in college graduates).  I have looked a bit at discussions of this, and there’s something wrong.  Or maybe a lot wrong.  Researchers apparently say that nobody reads 900 words a minute with full comprehension, but I’ve seen it done.  I would also like to know if anyone has statistically validated methods that  increase reading speed.

On related topics, I wonder how many serious readers  there are, here and also in other countries.  Are they as common in Japan or China, with their very different scripts?   Are reading speeds higher or lower there?

How many people have  their houses really, truly stuffed with books?  Here and elsewhere?  Last time I checked we had about 5000 books around the house: I figure that’s serious, verging on the pathological.

To what extent do people remember what they read?  Judging from the general results of  adult knowledge studies, not very much of what they took in school, but maybe voluntary reading is different.

https://westhunt.wordpress.com/2012/06/05/reading/#comment-3187
The researchers claim that the range of high-comprehension reading speed doesn’t go up anywhere near 900 wpm. But my daughter routinely reads at that speed. In high school, I took a reading speed test and scored a bit over 1000 wpm, with perfect comprehension.

I have suggested that the key to high reading speed is the experience of trying to finish a entire science fiction paperback in a drugstore before the proprietor tells you to buy the damn thing or get out. Helps if you can hide behind the bookrack.

https://westhunt.wordpress.com/2019/03/31/early-reading/
There are a few small children, mostly girls, that learn to read very early. You read stories to them and before you know they’re reading by themselves. By very early, I men age 3 or 4.

Does this happen in China ?

hmm:
Beijingers' average daily reading time exceeds an hour: report: http://www.chinadaily.com.cn/a/201712/07/WS5a293e1aa310fcb6fafd44c0.html

Free Speed Reading Test by AceReader: http://www.freereadingtest.com/
time+comprehension

http://www.readingsoft.com/
claims: 1000 wpm with 85% comprehension at top 1%, 200 wpm at 60% for average

https://www.wsj.com/articles/speed-reading-returns-1395874723
http://projects.wsj.com/speedread/

https://news.ycombinator.com/item?id=929753
Take a look at "Reading Rate: A Review of Research and Theory" by Ronald P. Carver
http://www.amazon.com/Reading-Rate-Review-Research-Theory/dp...
The conclusion is, basically, that speed reading courses don't work.
You can teach people to skim at a faster rate than they'd read with maximum comprehension and retention. And you can teach people study skills, such as how to summarize salient points, and take notes.
But all these skills are not at all the same as what speed reading usually promises, which is to drastically increase the rate at which you read with full comprehension and retention. According to Carver's book, it can't be done, at least not drastically past about the rate you'd naturally read at the college level.

http://unenumerated.blogspot.com/2006/01/letter-from-industrial-revolution.html
http://unenumerated.blogspot.com/2006/02/roundhead-revolution.html
http://unenumerated.blogspot.com/2007/09/institutional-changes-precedent-to.html
http://unenumerated.blogspot.com/2009/05/what-caused-birth-of-agriculture-and.html
http://unenumerated.blogspot.com/2013/11/european-asian-divergence-predates.html
http://unenumerated.blogspot.com/2014/10/transportation-divergence-and.html (critiqued by Pseudoerasmus)

https://westhunt.wordpress.com/2017/04/05/interview-2/
opinion of Scott and Hanson: https://westhunt.wordpress.com/2017/04/05/interview-2/#comment-90238
Greg's methodist: https://westhunt.wordpress.com/2017/04/05/interview-2/#comment-90256
https://westhunt.wordpress.com/2017/04/05/interview-2/#comment-90299
You have to consider the relative strengths of Japan and the USA. USA was ~10x stronger, industrially, which is what mattered. Technically superior (radar, Manhattan project). Almost entirely self-sufficient in natural resources. Japan was sure to lose, and too crazy to quit, which meant that they would lose after being smashed flat.
--
There’s a fairly common way of looking at things in which the bad guys are not at fault because they’re bad guys, born that way, and thus can’t help it. Well, we can’t help it either, so the hell with them. I don’t think we had to respect Japan’s innate need to fuck everybody in China to death.

https://westhunt.wordpress.com/2017/03/25/ramble-on/
https://westhunt.wordpress.com/2017/03/24/topics/
https://soundcloud.com/user-519115521/greg-cochran-part-1
2nd part: https://pinboard.in/u:nhaliday/b:9ab84243b967

some additional things:
- political correctness, the Cathedral and the left (personnel continuity but not ideology/value) at start
- joke: KT impact = asteroid mining, every mass extinction = intelligent life destroying itself
- Alawites: not really Muslim, women liberated because "they don't have souls", ended up running shit in Syria because they were only ones that wanted to help the British during colonial era
- solution to Syria: "put the Alawites in NYC"
- Zimbabwe was OK for a while, if South Africa goes sour, just "put the Boers in NYC" (Miller: left would probably say they are "culturally incompatible", lol)
- story about Lincoln and his great-great-great-grandfather
- skepticism of free speech
- free speech, authoritarianism, and defending against the Mongols
- Scott crazy (not in a terrible way), LW crazy (genetics), ex.: polyamory
- TFP or microbio are better investments than stereotypical EA stuff
- just ban AI worldwide (bully other countries to enforce)
- bit of a back-and-forth about macroeconomics
- not sure climate change will be huge issue. world's been much warmer before and still had a lot of mammals, etc.
- he quite likes Pseudoerasmus
- shits on modern conservatism/Bret Stephens a bit

- mentions Japan having industrial base a tenth the size of the US's and no chance of winning WW2 around 11m mark
- describes himself as "fairly religious" around 20m mark
- 27m30s: Eisenhower was smart, read Carlyle, classical history, etc.

but was Nixon smarter?: https://www.gnxp.com/WordPress/2019/03/18/open-thread-03-18-2019/
The Scandals of Meritocracy. Virtue vs. competence. Would you rather have a boss who is evil but competent, or good but incompetent? The reality is you have to balance the two. Richard Nixon was probably smarter that Dwight Eisenhower in raw g, but Eisenhower was probably a better person.

local vs. global errors in technical papers

old:
https://plus.google.com/+TerenceTao27/posts/78aoEHoPhpS

compilation errors in academic papers

old:
[Google Buzz closed down for good recently, so I will be reprinting a small n...
https://plus.google.com/u/0/+TerenceTao27/posts/TGjjJPUdJjk

I’ll organize my thoughts around the following ideas.

- Prioritize reading readable sources
- Build narratives
- Study other mathematician’s taste
- Do one early side project
- Find a clump of other graduate students
- Cast a wide net when looking for an advisor
- Don’t just work on one thing
- Don’t graduate until you have to

thinking like a physicist:

The key element in thinking like a physicist is being willing to push simultaneously to extreme levels of abstraction and specificity. This sounds paradoxical until you see it in action. Then it seems obvious. Abstraction means that you strip away inessential detail. Specificity means that you take very seriously the things that remain.

Abstraction vs. Radical Specificity: https://paulromer.net/abstraction-vs-radical-specificity/

how to read a paper
love this answer (eg, figures!): https://www.quora.com/What-is-the-best-way-to-read-scientific-papers/answer/Inna-Vishik

also: on reading papers