Ask HN: What are your best learning methods/hacks/tips?

259 points by justaguyhere ↗ HN
Say you're learning something that is totally new to you and totally out of your comfort zone (Something like learning Chinese when the only language you can speak is English and you're a westerner, learning archery when you have been a couch potato for years etc).

What would be your learning methods? Do you have any tips/hacks etc that works for you? Lets assume you are learning on your own, from books/videos etc, and not learning from a teacher

99 comments

[ 4.6 ms ] story [ 174 ms ] thread
I'd use mind maps to make notes, and spaced recognition to remember, and would try to understand context as much as possible.

With language learning though I'd try to get lessons on italki to speak with native speakers of that language. Grammar drills. Anki/Memrise for spaced repitition.

Anything I am good at, the common theme for why I am good is the interest (or force) to keep at it for years, reading, and tons of practice. There really is no way around practice. Practice is what separates most people who are good at things from those who aren't.

Music, math, programming, writing, learning languages, painting, drawing, physics, finance, etc. Find good books and good teachers/videos/other instruction, and practice relentlessly.

Practice is what separates most people who are good things from those who aren't.

Yes, but mindless practice doesn't help, isn't it? Is there a deliberate method to your practice? How do you keep track of your weak areas, and how do you specifically practice those weak areas?

Prevention of autopilot during practicing is a key aim of practice. When you autopilot, you allow muscle memory/habit to take over, when the entire point of practice is to improve upon your capabilities, not just repeat them over and over with no further thought.
Sometimes, but sometimes muscle memory is what you want. Do you want to learn to play a fast guitar solo? Learn it slow, then play it incrementally faster a hundred times or more over a week.

Do you want to get good at doing squats? Follow a program that makes the motion second nature by squatting multiple times a week and incrementally adding weight.

Muscle memory is really extremely useful for physical motions. Just be careful you are not cementing incorrect motions in your brain. This is why instructors, books, and videos can be very useful.

That is what I get from skiing lessons. You focus on what you are doing so that you are not on autopilot. I sometimes do this with touch typing to make sure I’m not cheating.
Mindless practice will get you substantially further than not doing anything. Even when you have only the vaguest idea what you’re doing doing something beats doing nothing almost universally. Deliberate practice is great but with a tight feedback loop just doing things works wonders.

If you spend 1000 hours doing mindless practice you won’t get as far as someone who spends 500 hours practicing deliberately but you are unlikely to do worse than the person who did 100 hours.

Spot on, and a particularly good point for people overwhelmed with the volume of resources out there to learn things.

Picking a method/book/instructor that seems to be getting good reviews and getting at it is far more important than wasting tons of time jumping from method to method but never actually putting in the time practicing.

More than "practice makes perfect", it really is "practice makes permanent"

At least in martial arts, I have learnt the hard way that if you had practiced a particular form wrong, then it takes that much more time to undo the learning, especially if it is the basics.

Amen.

Growing up, I asked so often for guidance for how to learn to write essays. I got told “just do it” and “just write the damned essay” and “just buckle down” so much I concluded that there wasn’t a way to learn a better writing process. So, I practiced writing via my having an emotional breakdown, procrastinating heavily and doing something at the last minute. As a 29-year-old, I’m trying to unlearn this and figure out how to write a technical blog post in a calm this-is-merely-work sort of way. Its really hard.

It certainly helps in that you learn what doesn't work. Whether you choose to admit that it's not working and try a different tack is up to you.
Indeed. That is why I included "Find good books and good teachers/videos/other instruction". They point out to you what and how you should practice, but practicing is still the key component.
One important piece here is they guide you to where you should focus your attention. I find it so much harder to focus when I don’t know what to focus on.
Besides practicing a lot I like reading books on the topic I’m learning. Usually multiple books and always from start to finish instead of jumping around in the book.

I apply this mostly when learning programming languages or frameworks or tools. It’s amazing how much small details you can pick up when reading an in-depth book on the topic cover to cover. Even for things you’ve been using for years.

I just commented this in another thread, but I highly, highly recommend the free Coursera course Learning How To Learn.

In particular, I found that the following worked especially well for me:

* Emphasis on balance between focus and diffuse mode: taking a walk or exercising after a focused study session.

* Spaced repetition: Anki works wonders. Reviewing notes multiple times is also tremendous.

* Sleep: ignored the importance of this for so long, but it helps so much.

I made it through undergrad without these, but applying these and the other lessons from the course in my post-graduate classes felt like magic. The results (both in understanding gained and grades) were completely different.

All of this. Your brain needs time to convert short term to long term. There are also techniques for memory like mind palaces which can assist learning.
Agreed! In that course, they also mention how important recall is for learning - so anything where you quiz yourself, or explain concepts to another person, etc.

When I'm learning a new subject, I try to make my study time as active as possible. For example, if I'm reading a very dense piece of philosophy, this means taking notes on each section and creating a study guide/outline of the whole paper.

I've recently gone through the course and it is as good if not better than I was told it would be. The better you can cement the strategies taught in the course the farther and faster you will go.
For spaced repetition, if you haven't tried it here is a fun interactive that is well presented to try it out [1].

For spaced repetition tools, besides Anki [2] there is also TinyCards [3].

Ultimately learning new things means building projects in them, but for rote memorization and when getting started spaced repetition is great. It works well because it is similar to how people learn naturally, the more you see something, the more you lock it into memory.

[1] https://ncase.me/remember/

[2] https://apps.ankiweb.net/

[3] https://tinycards.duolingo.com/

How does Tinycards compare to Anki? I've been using Anki for years. In terms of learning, it is the greatest thing since sliced bread.
Similar, both flash card based, and with spaced repetition.

Anki is downloadable and TinyCards is web based and online. TinyCards has a community around it but Anki is probably more individual focused. Both are great tools and you will learn something everytime.

Another comparison highlights differences [1]

> They are both spaced repetition systems, but here are some key differences:

> - They have a different look and feel, so that is a subjective factor. Tinycards is more modern, while Anki is arguably starting to look dated.

> - Tinycards decks are limited to only 150 cards. This is a problem if you like big decks. Anki decks have no practical size limit.

> - Tinycards can connect directly to your Duolingo account and has shared decks (that the original owner still controls), but Anki has a lot of shared decks that instantly become "yours" (free to make changes) once you download them.

> - Anki has more overall functionality, so if you like to add pictures and audio, etc, then you have more control.

[1] https://languagelearning.stackexchange.com/questions/3309/ti...

You lost me when you said "web based and online".
> - Tinycards decks are limited to only 150 cards.

Huge deal breaker for me. I have thousands of cards. If I study for any certification exam, I will have a minimum of a couple hundred cards.

I'm ashamed to say I had to stop watching this because I couldn't handle the woman's voice. It was like reverse-ASMR.
You can may be try the book or watch it at 1.5x playback speed.
How do you do anki with something like lecture notes? Do you just chunk the lecture notes into bite-size pieces and then review them as flash cards?
It definitely depended on the kind of notes, but pretty much!

For something like an algorithms or math class, I would have the problem one side and a summary of how to do it on the other side. This meant I wouldn't focus on the answer itself but more the process. Definitions and terms are also easy, where I would have the term on one side and a summary on the other.

Sometimes I skipped the bite-size and would make cards like "Review Lecture N." This doesn't seem exactly like The Anki Way of breaking down into small parts, but it was good for keeping track of when I should be reading over my notes periodically.

I have a similar experience too. I learned about that course after graduating but I apply the course's principles on my personal learning endeavors. I would like to add Cal Newport's Deep Work is also an excellent companion to this book. He ties in concepts from Peak, Grit and many more resources into concrete strategies that can be applied to effectively master cognitively demanding tasks. I wish I learned about those 2 books earlier.
I have tried a few times to get on the anki/supermemo bandwagon but have always struggled a bit figuring out how to structure the cards. Maybe I just have some hangups with it but would love to see a good workflow or best practices for incorporating it into my study.

I think specifically I would second guess myself for the syntax of the cards I was creating - and if there was another better way to do it. I always wanted to find almost like a guided walkthrough that would take me through the whole process or something and help me build the skills associated with it.

> I think specifically I would second guess myself for the syntax of the cards I was creating - and if there was another better way to do it.

This kind of perfectionism has also been bothering me. ("I need to write the card perfectly, or else it's not even worth doing or even actively harmful.") But there's a couple of points worth remembering to change this belief:

1) If a card is bad, you will notice it when reviewing. It will be difficult to remember (i.e. you will fail the card often compared to other cards); it will be annoying to review (there's a general sense of "ugh" and/or confusion when you see the card); it will be unexpectedly time consuming to review, etc.

2) Bad cards can always be refactored. You can suspend the card (where the card is still in the database, but removed from the learning queue); reword; or split into multiple cards.

Michael Nielsen [0] gives an example of a card which asked for the syntax for creating a symbolic link in Linux. He always messed up the order of the filname/linkname, so he created an additional card that explicitly asked for the order of the filname/linkename in the ln-command.

3) The only way of learning how to make good cards is by just starting making cards, and then noticing which ones don't work.

When a card doesn't stick, it's useful to ask yourself what doesn't work and why. Is the back side surprising when you reveal it? If so, maybe rewrite the card to add more context to the front to make it clearer what you're asking for. Do you always miss one or two pieces of the answer? If so, maybe split the card into multiple cards, each of which asks for one part of the answer. (Or add an additional card to direct your attention specifically towards what you struggle with, ala Nielsen.) Etc.

4) There's diminishing returns on card improvement. Time spent on perfecting an already OK card is time taken away from creating new cards to remember new information. If your goal is to remember as much as possible in a given time, spending time on perfecting already existing cards is trade-off not always worth making. (The quote: "a poem's never finished, only abandoned" comes to mind to highlight this.)

[0] http://augmentingcognition.com/ltm.html

yes - great points! Now that I've gotten into the routine of using Anki, I basically end up taking notes in it. I'll cram a whole lecture in a card. I can always figure out how to split it up properly into multiple cards after a few reviews.

in general the trap of "letting perfect be the enemy of good" is so easy to fall into

Thanks for your detailed reply. These are all good points. I need constant reminders about perfectionism and I like that quote.
If you don't know how to structure cards to be effective, I highly recommend you read through the 20 rules of knowledge formatting https://supermemo.guru/wiki/20_rules_of_knowledge_formulatio...

I used Anki for 7 years without these rules and figured some of it out on my own.

After I used Anki / supermemo to learn these rules, I can learn at a much higher rate with less reviews and higher recall.

Thank you, I will check those out.
How much does the course cost?
All the Coursera courses I've looked at can be audited for free. There are some advanced features of the courses that require enrolling for a fee. If you enroll, and pay the fee, you also get a Certificate of Completion. If you cannot afford the fee, you can apply for financial aid.
Thanks, that explains why there were no prices, just a signup form :)
She also recently published a book Learning How To Learn last August.

It is very easy to ready as it targets tweens, but it covers the same topics as the course.

Make a routine and commit to it, like going to the gym, or brushing your teeth. You just find time to carve out in the day to practice/learn, even if its like twenty minutes. Typically I would use this time to actually DO something - work through a tutorial, write a scraper to scrape NBA data, try some new library, etc. If you have a particularly helpful book on your topic, then use the time to read that (and preferably practice the material in the book).

I always found the Jerry Seinfeld method helpful - he would commit to writing material every single night, and mark his calendar with an X when he was done. Over time, he built up such a streak that it became almost second nature - he had come so far he did not want to break his streak, and break up that marvelous chain of 'X's on his calendar.

Best way to learn is to read the necessary information once or twice, then just do it.

If you're taking an exam, learn it by taking practice exams and looking up what you get wrong.

If you're building an application, build it, mess up, then rebuild it.

If you're learning a language, memorize the basic nouns and verbs, then jump in and start writing and speaking in that language. Look up what you don't know, then jump back in and keep on keeping on.

Use mental maps and word associations for memorizing.

Reading while walking helps me a lot surprisingly.

Learn stuff right before you sleep and you're brain will process the information while you sleep.

Exercise to keep your mind right. It's another way to meditate, but you also get a healthy body.

> If you're taking an exam, learn it by taking practice exams and looking up what you get wrong.

Nothing exposes the fact that you don’t know a damned thing like attempting past or specimen exam papers, or answering the questions in a textbook.

Ha yeah, but after the first run at least you know what you don't know.
Throw yourself into it, such as by speaking Chinese, in this example. I've learnt much more that way than theoretical reading, videos, etc.

I learnt 100x more about running a startup after starting one, and initially doing a bad job, than by reading thousands of blog posts.

What should you do if before throwing yourself into a situation, you think, “I can’t imagine how I would handle this well.” And then during the situation you think “aaaahhhhh! I don’t know what is going on or how to usefully focus my attention or get anything done!” And then after the situation you think “well, I failed pretty badly there”?

I’ve found in that sort of situation, it is tremendously useful to find a good book or other form of guidance so that I have a mental framework to organize information and a way to imagine myself succeeding. You know the OODA loop? I think theoretical knowledge really helps with the ‘Orient’ step.

There may be some people for which my advice is bad.

But otherwise, these are lessons. If you don't focus, and so get nothing done, you'll learn to focus, by saying no to some things, even if everyone tells you they're critical.

Conversely, if you focus too much on one thing, do it excellently, and fail at other things, which I've done, I've learnt to do things to the extent needed, not more, as opposed to building a car with the world's best engine but no steering wheel.

This has just been my experience. I respect that it may not work for you.

If I get a notion to do something, I first research what it is I want to achieve and how others have solved it. Once I decide on an approach (tools, time, etc.) I then read as much as I can before starting. I then prototype, make mistakes, get it right. I then document while implementing. Lather, rinse, repeat.

The Marines teach a concept know as BAMCIS (Bam-sys), which I still use loosely.

BAMCIS is an acronym: Begin planning, Arrange for reconnaissance, Make reconnaissance, Complete the plan, Issue the order, and Supervise.

This, in a nutshell, is how Marines plan and execute, and explains much of how they get stuff done, save lives, be successful with missions.

I don't know if it's really scientific or not but I was convinced (or maybe the right word is duped) by this BBC article:

>An effortless way to improve your memory

>A surprisingly potent technique can boost your short and long-term recall – and it appears to help everyone from students to Alzheimer’s patients.

Link:

http://www.bbc.com/future/story/20180208-an-effortless-way-t...

I found it so compelling that I consciously follow this effortless method.

Read and listen, practice, fix mistakes, repeat.
Applying what you are learning to a project you are working on, you get free "spaced repetition" that way every time you use that bit and/or come back to re-read/edit it.
You need to put the concepts you are learning into practice. If you just read and study something, no matter how long you do it, you will never truly digest the information.

Try things, fail, learn why you failed, do it the right way, succeed, repeat.

Learn by doing.

I taught myself morse code by actually using morse code on ham radio, not by drilling myself silly with code practice tools (which is imperative to learn the alphabet, but after that it's less practical).

There's no number of videos or essays or books etc that will make you magically get better. You must take what you discover from those sources and put muscle (and/or brainpower) into the activity yourself.

Its important not to misread this advice as “don’t seek out books/videos/guidence”. The value of those things is they point out where to direct your focus and thereby help you stay focused.
What I would like to know is how to retain what I learn in the long-term. It's not too difficult to memorise things, either by rote memorisation or using some of the techniques described here. But, how can I remember something banal that I learn today in 10 or 20 years - without using spaced repetition, as I find that is not always maintainable in the long term.

Why is my brain so efficient at erasing things I went through so much trouble at one point, and for a specific purpose, to memorise. It seems to decide for itself that this piece of information is irrelevant at some point and just wipes it away. Like a overly pro-active personal assistant.

If you don't use it, you lose it. Spaced repetition is "using it", even if only artificially. If you want to completely avoid spaced repetition you are going to have to find some other way to consistently use whatever you have learned, which arguably might end up being more strenuous than using a spaced repetition system like Anki because these try to figure out the minimum frequency of repetition you require to retain some piece of knowledge. And they're not bad at it. The intervals between which you see a specific card usually grow to a couple months quickly.

However, relearning something takes very little time compared to learning it from scratch, so unless you need to relearn things very often it might not even be worth worrying about much.

There are a bunch of evolutionary reasons that your brain forgets things that aren't used. Check this article for some related to sleep: https://www.supermemo.com/en/archives1990-2015/articles/slee...

But based on the current models of memory, you seem to be correct, your brain has a couple of criteria that it uses to determine whether or not it should retain memories. Frequency (spacing) is one, novelty could be another.

Even memories learned through mnemonics can ultimately be forgotten if not reviewed.

And I think the brain keeps itself plastic by optimizing "unnecessary" information out. Unfortunately, without help we can't choose which information is removed.

I often make an explicit effort to convert frequent actions from "System 2" to "System 1" (in the language of Daniel Kahneman) as soon as I can. For example, if I'm learning a foreign language, I commit common expressions (not words) to memory as soon as possible. If I'm learning a new software (say Adobe Illustrator or emacs), I commit the common shortcuts to memory as well. It's often one of the first things I do.

Then, when I actually learn the more complicated stuff, I don't ever think about the basic stuff (and I get that out of the way very early in the learning process). I think it frees up my brain to focus on the complexities of any domain a bit sooner than I would otherwise.

I'm not sure this method generalizes to other people, but it seems to work for me across various domains. I think the use of rote memorization is a bit underappreciated (given how useless it seems in high school or university), but I've continually found it useful in my professional life.

[Edit] It's also useful to identify what the "common action" are, and in general that's only from "learning by doing" and realizing what's important/what's not - then saving the important stuff to the "hot path" of processing.

I upvote the rot learning thing. It is hates by most people because it requires efforts but for things including basic words of a language and other writing systems it is what makes the difference between making real progress or not.
I even try to go far enough to move decisions into System 1 if I can. People sometimes treat System 1 like their "monkey brain" to be derided or otherwise fought against, but that doesn't actually work in practice.

So often your System 1 wins and your System 2 is never aware, that relegating System 1 to "impulse" is just hiding the problem from yourself. Training System 1 at least gives you a fighting chance to have accurate-enough first impressions/instincts.

Of course there are more-successful and less-successful ways of training System 1, but that's true of anything.

I think rote learning is only not applicable when you are learning disparate facts with no relation to the whole.

For example, learning all of the HTTP status codes before you know anything about HTTP could seem like a waste of time.

However, when you are writing a server that makes api calls to a third-party service, when you receive a 400, if you know the codes by rote you won't have to research / reason why you are receiving that code.

I think "learning by doing" usually ends with more self-directed, problem solving learning. This type of learning can show your weaknesses and provide you with holes in your knowledge that you can use rote memorizing to fill.

Learn how to learn and deliberate practice.
Depends on the domain. For languages, having learnt or being learning a few (including Japanese & Chinese) my approach is to first learn about the phonology. Then basic vocabulary and grammar, searching words I don’t know in the street (not everything) and having input from authentic material. I prefer learning from books using multiple one at the same time and cherry picking on what pick my interest. Having a few friends in the target language is a huge help to understand some concept or learn about cultural point. Also Wikipedia.

For other things, reading accessible research papers, reference books and searching for unknown concepts when I encounter them, sometimes ask people about a thing or two.

For me personally, I tend to remember things better if I write them down.

If I'm at a lecture, I take notes. If I'm reading a textbook (e.g., studying), I take notes. If I'm in a meeting, I take notes. And once I'm done taking notes, I transcribe those notes into more structured notes, since they were taken hastily, and I probably wouldn't be able to decode them later if I didn't do it right away.

Anyway, that's my process. Hope it helps.

1. Produce learning artefacts. Scribbles, notes, flash cards, mind-maps, drawings, projects, whatever. Don't let that overtake you or break your ingestion flow, and don't worry about that being usable in the future. Act of structuring that information is the key here.

2. Avoid learning what you won't put in practice immediately after. This is also useful to cull effectively infinite space of what you may want to learn next.

3. Learn the basics. Do not worry about this being marketable skill or knowledge - it'll pay dividends later on. "Basics" is not quantum physics, it's whatever's layer below the layer you're comfortable with.

4. Keep using what you learned, even if it's a toy project. Don't worry about making it public or useful to others.

I can't stress how important 2. is to learning. I don't think I learned anything well unless I used what I learned in practice. Reciprocal don't add anything to your learning if you won't use it on a regular basis.
i cant claim fantastic success, but I tend to flip this around. rather than say ' i want to learn to do x', i just start trying to do x and figure it out. obviously thats not very effective.

but when I do start to read the canon, everything makes so much more sense. pleased at having extracted some useful hints I go back to screwing everything up, but a little better this time..and repeat.

that means that i've wasted alot of time doing things the wrong way, but after that*, insight happens in a flash.

Try to make it fun.

If you enjoy the learning process, there is a much higher likelihood that you'll stick with it in the long-term, and a much higher likelihood that you'll dedicate a lot of time to the learning process.

The single most underestimated tactic to enhance your learning is to improve your speed and retention in Reading.