6 ms·

Somebody here once recommended when doing any kind of text with exercises to open to a random chapter and attempt the first exercise. If you can't do it researc

by feklar 10y ago

Somebody here once recommended when doing any kind of text with exercises to open to a random chapter and attempt the first exercise. If you can't do it research through the index/book where the information to help you solve it is likely to be found and repeat for all exercises. I've been doing this ever since, finished my library of unread books I had collected over the years but never had the time to read.

copperx 10y ago
What kind of books did you read using this technique? Also, did you end up reading most of the text? I'm a bit surprised that your techinque is more efficient than reading the chapter and then doing the exercises, because most of the chapter's content is on your short term memory and you know exactly where to find the information if you need to review a concept to solve the exercise.
- gremy0 10y ago
  I guess if you do the exercise first you actively know what information you need. So you can skip the content that's irrelevant and the stuff you know well enough already.
  - notyourloops 10y ago
    I think there's more to it. By turning material into a problem or an exercise, you've already turned on an active learning response. There's no way to coast through the text if you have something to grapple with already.
- greenshackle 10y ago
  It's not about efficiency, it's about effectiveness. Sequential reading may be more efficient, but it's infinitely less effective if you never pick up the book to read the next chapter. It's a workaround for the 'I started this useful, interesting book, read 50% of it, then never touched it again, for no good reason' bug. You solve the problem by nerd-sniping yourself with an exercise. If you don't experience this bug perhaps it's not good advice for you. My 'technique', if I want to read a particular book, is to carry it, and no other book, in my backpack. I'll inevitably read it in my down times.
  - Graziano_M 10y ago
    Yeah I have the problem of reading too many books at a time. I am in the last few chapters of an awesome but technically dense book, but I keep putting it off because it's more fun to spend an hour reading "The Two Towers".
- Jtsummers 10y ago
  If I pick up a book to learn a new math topic, the first chapter or two may be all refresher to me. I may have a good foundation in some other chapters already, as well. This method bypasses that. Instead of getting bored or distracted, you can identify what information you're missing in particular. Another benefit of this is that you can see what information you've actually learned/internalized, versus what was just staying on the surface of your memory/mind when you last took a related course.
- feklar 10y ago
  If I remember correctly this method was advised in the comments of an HN article about Concrete Mathematics by Knuth et all. I used this for undergraduate texts, mainly in math that I had collected over the years and never got around to reading, like Calculus (Apostol), Algebra: Chapter 0 (Aluffi), Computer Systems: A Programmer's Perspective (Bryant/O'Halloran), a 1000+ page book I flew through as I found I knew some of the material already and used the exercises to test what I had forgotten so I could re-read that material. I also discovered by doing this, subjects I thought I knew and would've skipped, I didn't actually remember correctly and other subjects I thought I didn't know, I could solve without having to read the chapter as previous exercises were helpful enough to understand the problem or I had picked up the information somewhere else. Aluffi's book also helpfully tells you which exercises require solving previous exercises.
  - j2kun 10y ago
    But Aluffi is so engaging even without this technique!
- Jach 10y ago
  "The reader who has read the book but cannot do the exercises has learned nothing." -- J.J. Sakurai A lot of books from my school had exercises you could not solve if your only knowledge about the subject was the previous chapter(s). So for those who could be bothered (most efficient technique is not doing reading or exercises at all but somehow getting a good grade) there were times one would come up with an insight or read further ahead / another reference and discover a new connection. Either way made for deeper learning.
notyourloops 10y ago
Or even if it's not random: Start with the exercises. I've found great success through a "problems-first" approach, and that's been my default since I was a child. Anytime I diverge from it I find notably worse results. Maybe people have different learning styles but this is the only way I really learn well.
shanusmagnus 10y ago
This seems like a really hard version of the credit assignment problem -- if you open to chapter 20 and get some differential equation that you can't solve, then what's the search strategy, other than read chapter 20? Which would make no sense if you hadn't read chapter 19, etc. Or so it seems -- I haven't actually tried it. Could you elaborate a bit on the process? I like the engagement and the 'information chase' aspect of it...
- feklar 10y ago
  Found the original post that explains the algorithm better: https://news.ycombinator.com/item?id=665250 https://news.ycombinator.com/item?id=665250 It is hard if the book has an incomplete index or poorly named chapters forcing you to skim around.
- csours 10y ago
  Yes, that does sound hard, but you might have to chain all the way back to chapter 1. Often times, chapters are grouped in self-contained blocks.
Spellman 10y ago
This is a classic problem in game manuals as well. They have to act as both tutorial AND reference guide. Working through front the front to back is necessary in order to learn the systems and things have to flow in a logical progression. Which deserves a particular style of writing and formatting. Embedding bits and pieces and forming a narrative as well as logical flow. But when you're in the middle of a game and need to know the exact rules surrounding a specific corner case and how it interacts with all the other relevant rules, suddenly a detailed narrative flow isn't what you need. You need a reference guide, or a mapping of how it ties in with other components. Unfortunately, this is quite difficult to execute for things even as simple as board game systems. For a textbook I imagine it's infinitely harder. A professor I had once talked about how each figure and diagram can take a week to polish to that is conveys the information in a detailed, yet succinct manner.

by feklar 10y ago

copperx 10y ago

What kind of books did you read using this technique? Also, did you end up reading most of the text? I'm a bit surprised that your techinque is more efficient than reading the chapter and then doing the exercises, because most of the chapter's content is on your short term memory and you know exactly where to find the information if you need to review a concept to solve the exercise.

gremy0 10y ago
I guess if you do the exercise first you actively know what information you need. So you can skip the content that's irrelevant and the stuff you know well enough already.
- notyourloops 10y ago
  I think there's more to it. By turning material into a problem or an exercise, you've already turned on an active learning response. There's no way to coast through the text if you have something to grapple with already.
greenshackle 10y ago
It's not about efficiency, it's about effectiveness. Sequential reading may be more efficient, but it's infinitely less effective if you never pick up the book to read the next chapter. It's a workaround for the 'I started this useful, interesting book, read 50% of it, then never touched it again, for no good reason' bug. You solve the problem by nerd-sniping yourself with an exercise. If you don't experience this bug perhaps it's not good advice for you. My 'technique', if I want to read a particular book, is to carry it, and no other book, in my backpack. I'll inevitably read it in my down times.
- Graziano_M 10y ago
  Yeah I have the problem of reading too many books at a time. I am in the last few chapters of an awesome but technically dense book, but I keep putting it off because it's more fun to spend an hour reading "The Two Towers".
Jtsummers 10y ago
If I pick up a book to learn a new math topic, the first chapter or two may be all refresher to me. I may have a good foundation in some other chapters already, as well. This method bypasses that. Instead of getting bored or distracted, you can identify what information you're missing in particular. Another benefit of this is that you can see what information you've actually learned/internalized, versus what was just staying on the surface of your memory/mind when you last took a related course.
feklar 10y ago
If I remember correctly this method was advised in the comments of an HN article about Concrete Mathematics by Knuth et all. I used this for undergraduate texts, mainly in math that I had collected over the years and never got around to reading, like Calculus (Apostol), Algebra: Chapter 0 (Aluffi), Computer Systems: A Programmer's Perspective (Bryant/O'Halloran), a 1000+ page book I flew through as I found I knew some of the material already and used the exercises to test what I had forgotten so I could re-read that material. I also discovered by doing this, subjects I thought I knew and would've skipped, I didn't actually remember correctly and other subjects I thought I didn't know, I could solve without having to read the chapter as previous exercises were helpful enough to understand the problem or I had picked up the information somewhere else. Aluffi's book also helpfully tells you which exercises require solving previous exercises.
- j2kun 10y ago
  But Aluffi is so engaging even without this technique!
Jach 10y ago
"The reader who has read the book but cannot do the exercises has learned nothing." -- J.J. Sakurai A lot of books from my school had exercises you could not solve if your only knowledge about the subject was the previous chapter(s). So for those who could be bothered (most efficient technique is not doing reading or exercises at all but somehow getting a good grade) there were times one would come up with an insight or read further ahead / another reference and discover a new connection. Either way made for deeper learning.

notyourloops 10y ago

Or even if it's not random: Start with the exercises. I've found great success through a "problems-first" approach, and that's been my default since I was a child. Anytime I diverge from it I find notably worse results. Maybe people have different learning styles but this is the only way I really learn well.

shanusmagnus 10y ago

This seems like a really hard version of the credit assignment problem -- if you open to chapter 20 and get some differential equation that you can't solve, then what's the search strategy, other than read chapter 20? Which would make no sense if you hadn't read chapter 19, etc. Or so it seems -- I haven't actually tried it. Could you elaborate a bit on the process? I like the engagement and the 'information chase' aspect of it...

feklar 10y ago
Found the original post that explains the algorithm better: https://news.ycombinator.com/item?id=665250 https://news.ycombinator.com/item?id=665250 It is hard if the book has an incomplete index or poorly named chapters forcing you to skim around.
csours 10y ago
Yes, that does sound hard, but you might have to chain all the way back to chapter 1. Often times, chapters are grouped in self-contained blocks.

Spellman 10y ago

This is a classic problem in game manuals as well. They have to act as both tutorial AND reference guide. Working through front the front to back is necessary in order to learn the systems and things have to flow in a logical progression. Which deserves a particular style of writing and formatting. Embedding bits and pieces and forming a narrative as well as logical flow. But when you're in the middle of a game and need to know the exact rules surrounding a specific corner case and how it interacts with all the other relevant rules, suddenly a detailed narrative flow isn't what you need. You need a reference guide, or a mapping of how it ties in with other components. Unfortunately, this is quite difficult to execute for things even as simple as board game systems. For a textbook I imagine it's infinitely harder. A professor I had once talked about how each figure and diagram can take a week to polish to that is conveys the information in a detailed, yet succinct manner.