When I'm learning a new skill, there's a technique I often use to quickly gain the basics of the new skill without getting drowned in the plethora of resources that exist. I've found that just 3 resources that cover the skill from 3 separate viewpoints(along with either daily practice or a project) is enough to quickly get all the pieces I need to learn the new skill.

I'm partial to books, so I've called this The 3 Books Technique, but feel free to substitute books for courses, mentors, or videos as needed.

The "What" Book

The "What" book is used as reference material. It should be a thorough resource that gives you a broad overview of your skill. If you run into a novel situation, you should be able to go to this book and get the information you need. It covers the "surface" section of the learning model from nature pictured above.

Positive reviews of this book should contain phrases like "Thorough" and "Got me out of a pinch more than once." Negative reviews of this book should talk about "overwhelming" and "didn't know where to start."

The "How" Book

The "How" Book explains the step-by-step, nuts and bolts of how to put the skill into practice. It often contains processes, tools, and steps. It covers the "deep" part of the learning model covered above.

Positive reviews of this book should talk about "Well structured" and "Clearly thought out." Negative reviews should mention it being "too rote" or "not enough theory."

The "Why" Book

The "WHY" book explains the mindset and intuitions behind the skill. It tries to get into the authors head and lets you understand what to do in novel situations. It should cover the "transfer" part of the learning model above.

Positive reviews of this book should talk about "gaining intuitions" or "really understanding". Negative reviews should contain phrases like "not practical" or "still don't know what steps to take."

The Project or Practice

Once I have these 3 resources, I'll choose a single project or a daily practice that allows me to practice the skills from the "How" book and the mindsets from the "Why" book. If I get stuck, I'll use the "What" book to help me.

Examples

Overcoming Procrastination

"What" Book: The Procrastination Equation by Piers Steel

"How" Book: The Now Habit by Neil Fiore

"Why" Book: The Replacing Guilt blog sequence by Nate Soares

Project or Practice: Five pomodoros every day where I deliberately use the tools from the now habit and the mindsets from replacing guilt. If I find myself stuck, I'll choose from the plethora of techniques in the Procrastination Equation.

Learning Calculus

"What" Book: A First Course in Calculus by Serge Lange

"How" Book: The Khan Academy series on Calculus

"Why" Book: The Essence of Calculus Youtube series by 3blue1brown

Project or Practice: Daily practice of the Khan Academy calculus exercises.

Conclusion

This is a simple technique that I've found very helpful in systematizing my learning process. I would be particularly interested in other skills you've learned and the 3 books you would recommend for those skills.

New to LessWrong?

New Comment
48 comments, sorted by Click to highlight new comments since: Today at 2:31 PM

I really like the "positive reviews should look like X, negative reviews should look like Y" information. I've never seen it before, and I expect it to actually be useful when looking for resources.

I'm cofused by how "deep" and "surface" are being used in your first picture. From how the "What" and "How" books are described (and from the examples you give), I would have called "What" the deep resource, and "How" as the "surface level" resource. How are you thinking of it?

This may be because of my particular learning style. I tend to get most of my deep learning from the actual application of the skill, which is based on the how resource. I use the what resource in a very surface way, just getting particular facts or techniques when I'm stuck. However, I agree that What books tend to cover material in a deeper way

I think I can relate. You made me notice that there are two things I could point to when talking of "deep learning". One is "making a piece of knowledge or information a deeply ingrained, easily accessible piece of me" and the other is, "not having any whole in my conceptual understanding, every piece of info is well connected in my knowledge graph and well motivated, and things make sense in a powerful way".

I have a question in a similar vein -- how would one go about finding resources that don't have immediately available reviews -- videos, blogs, etc?

The old fashioned way I suppose, going by the reputation of the creator and the description they provide. I think readily available reviews are often worth going with a medium that has them though.

Dude, I'm just a guy, but I love this blog. It has changed my head.

Three books! That's great! I gotta tell you...

Here's a practice that I have long used to gain a skill, or just get through a project. I get the three best-looking books on the topic. The classic, the bible, the tome. (I avoid the "for dummies.") Reading them, I determine my favorite AUTHOR, the guy whose outlook and philosophy and depth and commitment to the skill impresses me the most. Then I CALL THAT SUCKER UP ON THE PHONE.

It helps that I'm as personable and likable on the phone as I am in person. I describe my project a little, then I ask where to get the special sauce or the obscure gizmo. That's always an ice breaker. Before long, he's talking; he's asking ME about my project. I keep it fairly short, having the pertinent questions and data at hand, but I always ask if I can call again (never been refused). When I get stuck or something, I call back.

I have gotten an incredible amount of tutoring and mentoring over the years. Enough to completely restore wood-and-canvas watercraft, build a couple of mandolins, rebuild vintage guitar amps, build a bicycle frame from SCRATCH, install a solar system, sew a full set of outdoor gear, and install a slate roof. More, too. The wood-and-canvas canoe guy actually took to calling ME, asking how things were going, checking on my progress, adding tips he forgot to mention. He sent me a gallon of the obsolete historically correct dope. I have straddled a roof ridge, on a cell phone with the world's number one slate guru talking me through an origami copper flashing technique. To this day, no leaks.

So there--I feel like I have contributed something to acknowledge all the great stuff I have read here.

Goddamn that is a great idea! This style totally suits how I learn in that I A) am personable, B) Immensely enjoy phonecalls, and C) Strongly, strongly prefer projects when I can verbally ask questions of someone who has done before.

How long does it generally take you to find the person's number? Have you had more difficulty getting them to pick up now that VOIP spam is so common?

I would offer $20 to anyone who can give examples for the three different books in an area, that they actually read, up to a maximum of $60. (e.g. three books in Linear Algebra, or three books in carpeting, or three books about programming).

Buddhism: What: What Makes You Not a Buddhist by Dzongsar Jamyang Khyentse Rinpoche (There are probably books with more detail and a more broad view, but I love this one for how it contrasts ideas from popular culture with Buddhism to highlight similarities and differences, making it very accessible); How: Meditation Is Not What You Think by Jon Kabat-Zinn; Why: Bring Me the Rhinoceros by John Tarrant (It's not explicit in explaining the why, but presents zen koans that cause you to enter the mindset of letting go of assumptions)

Crisis Intervention Counseling: What: Interviewing for Solutions by Insoo Kim Berg and Peter De Jong; How: Motivational Interviewing by Miller and Rollnick; Why: On Living by Kerry Egan

Teaching: What: Understanding by Design; How: Made to Stick by Dan and Chip Heath (not strictly a "How" book, but an excellent book to develop strategy for presenting curriculum); Why: Pedagogy of the Oppressed by Paulo Freire

I don't want your money - as soon as I finished this article I started thinking about the topics that I study the most and what books I would recommend to people. Also, I am a suicide intervention trainer by trade, so I have an obsession with figuring out how to teach people about empathy. :D

Sadly, there aren't a lot of books dedicated to suicide intervention, so it's harder to make a list for that.

I have read a lot of books on Buddhism, but the first in your list is the only one I have even heard of. I will have to check them out.

Little late to the party here but I was reading this and at least one immediately came to mind. Corporate valuation:

What: Investment Valuation: Tools and Techniques for Determining the Value of any Asset by Aswath Damodaran,

Why: Corporate Finance, 4th Edition, Ivo Welch, but perhaps better in lecture form. Khan Academy's finance series is good as well.

How: HBR Tools: Business Valuation

Thanks! (Might be easier to read the comment if the three books were separated onto separate lines, although ymmv)

Is the point of your comment that you think people very rarely read (completely or almost completely) 3 books in one field?

(if yes, then I agree)

Oh, no. Sorry. I didn't intend to make any point. I just thought that having additional examples would be really valuable for this post, and wanted to incentivize people to post them.

I love your thoughts on this. As someone who has studied educational design and psychology, I'd like to offer a perspective on what you're describing.

In educational psychology, there are three categories of educational goals: knowledge, skills, and attitudes. (Presently, there is a bias against teaching attitudes, so the "KSA" is frequently changed to "knowledge, skills, and abilities," which is incorrect and redundant.)

It seems that your "What" book represents the knowledge of the field, the "How" the skills, and the "Why" the attitude.

I think you would be interested in reading about Bloom's taxonomy of educational objectives. I think it would integrate well with the model you are creating, especially in light of the comments earlier about what "depth" means in this context and whether "What" or "How" is of greater depth. Either one can be shallow or deep as they relate to different kinds of learning.

This sounds like a topic worth digging into. Don't suppose you have a curated list of books (or other resources) to share?

Thank you for the article. It's very interesting.

My suggestion for the topic of Decision Making:

What-book: 2011 Daniel Kahneman - Thinking, Fast and Slow

How-book: 2013 Chip and Dan Heath - Decisive

Why-book: 2007 Phil Rosenzweig - The Halo Effect

Ok. I have read the first two, guess I gotta read the third. I wonder what sort of project one could apply these to. I remember when I read Thinking Fast and Slow I tried to use it to earn more tips, but the application, while seemingly useful, wasn't an excellent fit.

A very interesting technique, I especially like the examples of what reviews will say about each type of book.  

Lets see if I can create a few example lists from my library:

Experiencing Flow 

What Book: Flow by Mihaly Csikszentmihalyi

How Book: Joy on Demand: The Art of Discovering the Happiness Within Book by Chade-Meng Tan

Why Book: Trying Not to Try: The Art and Science of Spontaneity by Edward Slingerland

Having low conflict romantic relationships

What Book: The Usual Error by Kyeli Smith and Pace Smith

How Book: It's Not You, It's the Dishes by Paula Szuchman and Jenny Anderson

Why Book: Loving in Flow: How the Happiest Couples Get and Stay That Way by Susan K. Perry

I curated this for being a handy abstraction for learning. As comments downthread have said, things like distinguishing positive from negative reviews, and the actual examples you gave (procrastination and calculus), were both really helpful. And the post was as short+simple as it could be.

(I'd love to see more examples in the comments.)

this is very interesting, i'm likely to try it, thank you :)

btw, i also found the graphic in the beginning confusing

Anyone successfully approached this with computer programming? It's a subject I think I have reason to believe I would be talented at (I have exhibited strong aptitude for mathematics and languages), but I don't see a good path forward on working at it, so I learn other things. I think the "how" book might be something like codeacademy or khanacademy, but how about the "what" book or the "why" book?

SICP is a "Why" book, one of the few timeless texts on the topic. It's subsumed by studying any healthy functional programming language to a sufficient extent (idiomatic use of control operator libraries, not just syntax), but it's more straightforward to start with reading the book.

Videos of the lectures that go with that book are also freely available online. I remember seeing two different versions.

Installing MIT Scheme might be an obstacle for someone new to computer programming. It might be easier to start by installing DrRacket and the SICP Collections.

Composing Programs might also be worth mentioning. It's a Python-based textbook in the spirit of SICP and covers some of the same ground in a more approachable way, but perhaps in less depth.

Petzold's CODE might be another good "Why" book. Its focus is a little more on computer architecture than programming per se, but that's an important topic for programmers to understand. It's a pretty easy read. I'd call it more of a pop book than a textbook, but it still covers important concepts. If you're looking to learn programming, I'd highly recommend reading it.

<3 that book.

How to Design Programs (HtDP) might be a good "How" book.

For programming, I think starting with a project and using that to decide what books to read my work best. Assuming you want to learn to program rather than learn Computer Science, the books that will be helpful will depend highly on the area in which you're interested.

Do you just generally want to see if you'll be good at programming? Even if so, is there a specific area which you'd be interested in writing a program, e.g. an operating system, a server, a web app, etc.?

I agree with the comment below that SICP is a good "Why" book but did want to note that I personally didn't find SICP nearly as enlightening when I started programming as many others seem to. I've gone back to it since and loved it, but it definitely was not the thing that motivated me to practice programming a lot. Like everything else, it depends on your personality.

The 3 books technique includes a project or practice.

Whoops, will edit my comment to reflect that.

Other than suspecting I may have an aptitude, my interest in computer programming is driven by finding two fields cool and fun sounding: data science and applications of blockchain technology to stuff like verifying carbon sequestration and other changes in reality. Quite a bit of social science I deeply admire has been done using data science, and I have a couple of friends working to improve the world using blockchain technologies whom I also admire. I want to see if I am good at programming to see if I can participate in these endeavors I admire—or at least use my skills as a wordsmith and communicator to aid such endeavors.

If it is relevant, my own background is in social sciences and math.

Thank you so much for all your advice. You guys are awesome. Like An1lam and Mr.-hire point out though, I would actually need a project to work on. A practical, immediate application makes skills learnable much faster. Any advice on that?

Try making a simple video game. Think Tetris, not Skyrim. I personally made variants of Pong, Joust, and Asteroids when I was learning programming. You can learn a lot from making video games because it requires a variety of programming techniques. You need to understand basic algorithms, some mathematics, and you can't be too inefficient about it or you'll notice performance problems. And seeing your little world come to life is very motivating. This kind of thing is not terribly difficult to do in Python or Racket.

Note that many of the textbooks we've been recommending include exercises. Some of the more difficult ones count as small projects.

Introduction to Algorithms (CLRS) might be a good "What" book. [Edit: I'm no longer confident this fits into the "what" category.]

I think CLRS is a pretty questionable book for someone who hasn't programmed. I don't think it's great as a reference for writing algorithms, e.g. I think internet searching will often help you find better resources. And in terms of a straight read-through, it's one of the more theoretical algorithms texts, and a large fraction of its exercises are proofs.

If the OP is interested in an algorithms book but has never done any programming or CS, I'd recommend The Algorithm Design Manual (which I've read much of and done a decent number of exercises) or Jeff Erickson's free algorithms book (which I've read sections of and been impressed by).

I quite like this approach. :) I’ll see if I can apply it to electrical engineering and pure mathematics soon, as those are the subjects I am studying in school. Linear algebra will be my first stop.

3blue1brown has a series on the essence of linear algebra as well. It's pretty great, and could do well as the Why.

I also like Linear Algebra Done Right a lot, but it doesn't fit neatly into this framework. It's a bit too rigorous to be Why, not practical enough to be How, and it's approach differs enough from other books to make it difficult to look things up in.

Linear Algebra Done Right wins the prize for books which deliver on their title, as promised imho. The typical approach with determinants front and centre is a pedagogical dead end. I find Axler's approach tracked more closely to the uses of linear algebra, rather than historic proof techniques. I'd say it covers the how and why sections - although perhaps the 'how' book is the Matrix Cookbook ;). The what will probably have to come from some other discipline: are you working on stats, or differential equations, or something else?

I haven't used this technique, but this feels like largely a sad thing about me not really deliberately trying to thoroughly learn a novel skill. Curious if anyone else has reports on this technique.

I will second Hazard's comments on the way to approach the reviews. Seems so obvious after you said it.

I am wondering what you were considered in terms of skills.

I think the general approach should work really well in some settings. I do wonder what the limits might be. For instance, I could probably use this approach well to learn how to write in a programming language, or how to weld or how to machine parts. I'm not sure it's as helpful for, say, learning a new language.

Perhaps the distinction is between skill and knowledge?

This is an interesting question. I can imagine the technique being useful for acquiring the general skill of language learning, but a language itself I can only really see the "what" and "how" books being useful, not the why.

Nor can I imagine this technique being very helpful to learn how to ride a bike, although I imagine it could be useful to become a competitive bike racer.

The distinction between skill and knowledge seems a good start, but it seems like there's more going on here.

Maybe the "why" book for a language would be something like reading manga for someone wanting to learn Japanese - something that makes the language and culture seem cool and something that you want to learn. :)

Given that learning a language also includes a fair chunk of learning the culture (e.g. knowing which forms of address are appropriate at which times), reading literature from that culture is probably actually useful for accomplishing the "Why" book's goal of explaining the mindset and intuitions behind the skill.

I love this frame for learning something new, so I'll toss out a triad of books.

Subject: Journalism (These works are more complementary than overlapping. The third selection is an example rather than an explanation. It could easily be substituted with The Right Stuff or Frank Sinatra Has a Cold or any work that speaks to you.)

What: The Investigative Reporter's Handbook: A Guide to Documents, Databases, and Techniques

How: Writing for Story, Jon Franklin

Why: All the President's Men

Do you generally read the "What" book all the way through, or only use it as a reference when you get stuck? Could a Q&A forum, e.g. StackExchange, serve as the "What" book, do you think?

I almost never read the what book all the way through. I do think that a Q&A forum could work for this.