Live data from Hacker News

“Learning How to Learn,” the most popular course on Coursera

nytimes.com

151–160 of 196 posts

Re: “Learning How to Learn,” the most popular course on Coursera

#151
post #9

Took the course. A lot of it is cruft and motivation for the underlying core ideas. The techniques suggested are things many people are already familiar with: recall, deliberate practice, interleaving, spaced repetition, Einstellung, Pomodoro, Feynman Method, Cornell notes or similar (to force recall), exercise regularly, sleep well, focus on concepts not facts (chunking), etc. A composite of these dramatically enhan…

>> A lot of it is cruft and motivation for the underlying core ideas. Someone else made the exact same remark about Cal Newport's book "Deep Work" earlier this week in another thread. I find this to be true about a lot of books/courses/content of a "self-help" nature that are designed to be practical and applicable - the real helpful stuff boiled down to its essence could fit in a tri-fold pamphlet (or a 350-line pas…

A more charitable interpretation could be that all that "filler" is used to motivate you, because just reading about this methods won't be enough to actually help most people.

Re: “Learning How to Learn,” the most popular course on Coursera

#152
post #36

Earlier quoted context omitted.

Not OP but I also recently took the course, my notes* are here for anyone interested: https://workflowy.com/s/E9HW.jGUYboLrGj *Apologies if they are somewhat disorganized; they are generally broken down into the four discrete units that match the course. I wrote the notes with no intention that other eyes would see them :) feel free to reach out if you have any questions on them though

Thanks for the notes. I did first week and honestly felt it was too slow and somewhat discontinued. I am going to read your notes :)

You may try what I did: speed up the video, I don't remember the specific increase but for Dr. Oakley I think I did speed up by around 50% but a little less for Terrence Sejnowski. It helps a lot to not get bored because they speak to slow for you.

Re: “Learning How to Learn,” the most popular course on Coursera

#153

I find this interesting for two reasons. I think I recall that not too long ago, the most popular course on coursera was Ng's ML course. It is ironic that people are now more interested in teaching oneself how to learn versus a machine. This change could be attributed to other reasons like change in user demographics, or, market saturation, so that naturally popular courses will change once a large majority moves fro…

Ng's ML course was one of the first ML MOOC courses. Right now you have hundreds if not thousands of excellent ML courses to choose from.

Do you have any specific recommendations, especially ones that are more recent and/or have a high level of relevance and foundational information?

Re: “Learning How to Learn,” the most popular course on Coursera

#154
post #9

Took the course. A lot of it is cruft and motivation for the underlying core ideas. The techniques suggested are things many people are already familiar with: recall, deliberate practice, interleaving, spaced repetition, Einstellung, Pomodoro, Feynman Method, Cornell notes or similar (to force recall), exercise regularly, sleep well, focus on concepts not facts (chunking), etc. A composite of these dramatically enhan…

>> The techniques suggested are things many people are already familiar with [...] And you then list 11 concepts, 5 of which I'd never heard of before I took this course, 3 more which I had not associated with aiding learning and the remaining 3 not being sufficient to help me learn more. Your tone is unhelpful, elitist and snobby. This course is extremely useful to a lot of people - there is a reason it's popular, a…

Nothing wrong with his comment. Yours is rude.

Re: “Learning How to Learn,” the most popular course on Coursera

#155
post #9

Took the course. A lot of it is cruft and motivation for the underlying core ideas. The techniques suggested are things many people are already familiar with: recall, deliberate practice, interleaving, spaced repetition, Einstellung, Pomodoro, Feynman Method, Cornell notes or similar (to force recall), exercise regularly, sleep well, focus on concepts not facts (chunking), etc. A composite of these dramatically enhan…

>> A lot of it is cruft and motivation for the underlying core ideas. Someone else made the exact same remark about Cal Newport's book "Deep Work" earlier this week in another thread. I find this to be true about a lot of books/courses/content of a "self-help" nature that are designed to be practical and applicable - the real helpful stuff boiled down to its essence could fit in a tri-fold pamphlet (or a 350-line pas…

My favourite example of this is Harold Pollack who got tonnes of media coverage for coming up with 9 rules for personal finance which fit on an index card. I listened to his interview on Freakonomics, and he's now selling a 256 page book called "The Index Card: Why Personal Finance Doesn't Have to Be Complicate". 256 pages, about his rules that fit on an index card.

As an aside I always hear podcast adverts for https://www.blinkist.com/ which does audio/text summaries of non-fiction books, I haven't tried it yet but it might be worth checking out.

Re: “Learning How to Learn,” the most popular course on Coursera

#156
post #53

I haven't followed a course on Coursera since the first iteration of Crypto I. I heard that it became really bad, asking you to pay for a lot of courses.

I'd counter this mindset by saying that OF COURSE they ask you to pay. In my experience they've provided extremely high quality educational material, and even while paying its a fantastic deal. I especially liked the algorithms course by Tim Roughgarden. It's one thing to say that their lectures are objectively bad, but quite another to argue that asking for payment in return for a service is itself a bad thing. I pe…

> OF COURSE they ask you to pay

I had no idea this was going to be their business model. When Coursera started I thought it was this great scheme like Wikipedia is, free education for everyone. That's why I'm disappointed, the amount of money might be little for some, but it will be a barrier to others.

Re: “Learning How to Learn,” the most popular course on Coursera

#157
post #117

Earlier quoted context omitted.

Agreed. There's a serious problem of information density in content these days. Most publisher-driven content is almost intentionally sparse. It takes tremendous expertise to weld information with context in a way that doesn't feel hollow and commercial. Two books I've come across recently that do well to circumvent that problem: Sapiens by Yuval Harari and Designing Data-Intensive Applications by Martin Kleppmann. I…

The problem of low information density isn't new. Most of the "classics" of western literature have the same problem. We're just more attuned to it now because we've been trained by internet articles to expect an idea to be introduced, explored and concluded in 1000-2000 words. Beyond that, these days many people won't even bother with that if there isn't a quick hook or a TL;DR that sucks them into reading further.

I think one good test is actually the table of contents.

Some older books which were content-rich had the feature of "descriptive" TOCs. They would not only have a clear headline, but also a sort of teaser description of what that chapter would tell you. So you could use that to get a high level understanding of what the book is about and what it would tell or teach you without even having to skim the actual content.

On the other side of the spectrum, many of those sparse business/popsci/etc books often have extremely cryptic TOCs. That's a good sign to just ignore that book.

Re: “Learning How to Learn,” the most popular course on Coursera

#158
post #122

Earlier quoted context omitted.

I agree in general, but reading 3-page summary will not leave you with deep impression if it is applicable/correct and why. Blog posts are like that and how many you read vs how many you changed your behavior? I tried one of the services that provide summaries for the books but disliked it. But I agree that this books, in general, could have around 100-150 pages, 200 tops.

One of the promises of hypertext was to be able to "dig in" into deeper information if you wanted to. It would be nice to see a single page summary of a book, and then be able to click on hyperlinks to see more details about the interesting bits. Self-help books and textbooks are ideal for this, but I can see that it's a lot of work to structure something like that.

I always found these online versions of Wittgenstein's Tractatus very enlightening, because the entire book is hierarchically structured and thus lends itself to top-down exploration (which doesn't work well in traditional books): http://www.tractatuslogico-philosophicus.com/

Re: “Learning How to Learn,” the most popular course on Coursera

#159

Earlier quoted context omitted.

Ng's ML course was one of the first ML MOOC courses. Right now you have hundreds if not thousands of excellent ML courses to choose from.

Do you have any specific recommendations, especially ones that are more recent and/or have a high level of relevance and foundational information?

fast.ai if you want to get started quickly and don't mind much of underlying algorithms being treated as black boxes. Andrew Ng's new coursera course if you want a better understanding of the underlying theory.

Re: “Learning How to Learn,” the most popular course on Coursera

#160
post #117

Earlier quoted context omitted.

Agreed. There's a serious problem of information density in content these days. Most publisher-driven content is almost intentionally sparse. It takes tremendous expertise to weld information with context in a way that doesn't feel hollow and commercial. Two books I've come across recently that do well to circumvent that problem: Sapiens by Yuval Harari and Designing Data-Intensive Applications by Martin Kleppmann. I…

The problem of low information density isn't new. Most of the "classics" of western literature have the same problem. We're just more attuned to it now because we've been trained by internet articles to expect an idea to be introduced, explored and concluded in 1000-2000 words. Beyond that, these days many people won't even bother with that if there isn't a quick hook or a TL;DR that sucks them into reading further.

> The problem of low information density isn't new. Most of the "classics" of western literature have the same problem.

I would submit that you may not be tuned into the value in them, if you think that. Maybe it's true of some, but most are just an acquired taste of the same sort as, say, the freer-form varieties of jazz, or classical music. It takes some work and some experience before you get what they're doing, before you understand their rhythms, learn to recognize when they're telling you to pay attention (and to what, which may not be obvious at first), when and how they're simply playing and what about the particular play they're engaging in makes it remarkable and worthwhile, and so on.

There's a sensitivity required for much (not all) of the body of classic literature that most people have to work to develop before they really get it, and lots of allusions that have the chicken/egg problem of requiring you already to be familiar with a lot of literature before they start to land with any frequency—again, same as some genres of music, like classical and jazz again, hip hop especially, others too, but usually not as much as those. Just because you can read doesn't mean you get literature, just like being able to see doesn't mean you get fine art, though in either case you might still enjoy at least some of it, and you might still develop strong opinions about the works you don't enjoy.

Movies and TV are similar. You're not born recognizing good film-making, understanding why it's good, how the way the camera moved in that shot actually contributed meaning rather than just being an artifact of the medium, how that shot was a reference to some DW Griffith movie and what purpose that might have, how that shot is weak and works against the tone or action of the scene in a way that makes it bad, and so on. Most people who watch and enjoy film never bother to develop an actual understanding of it, but that doesn't mean the things that go over their heads are bad.

That's just how all art is, really. Sure there's some posturing and BS mixed in with it all, but you in fact do have to put some serious hours and effort into arts to get the most out of them that you can—even just on a critical or enjoyment or character/intellect/empathy-building level, not even getting into making art—and very probably by the time you've done that a bunch of things that seemed like worthless overrated crap when you started won't seem that way anymore.

Post reply on HN