Hacker News new | past | comments | ask | show | jobs | submit login
The Most Popular Online Course Teaches You to Learn (2015) (bits.blogs.nytimes.com)
284 points by bootload on Sept 18, 2016 | hide | past | favorite | 64 comments

Can't believe after scrolling through that there's no link to the course in the article:


For the impatient, the author of the course has a nice handout with 10 tips for good and bad studying: http://www.barbaraoakley.com/pdf/10rulesofstudying.pdf

Hmm I disagree with point number 7 under 10 Rules of Bad Studying

> 7. Neglecting to read the textbook before you start working problems.

I think it's a perfectly valid approach. I found that trying to solve the problems first helped me form questions which I would then seek to answer by reading the text. Of course then I would go back to the problems and try to solve them again. Rinse and repeat until I could solve them with confidence.

It's okay to disagree with any of those bullets. I disagree with the entire hand out, but that's not relevant to the people it works for in whole or in part.

I'm a total autodidact. I've been doing it long enough that I know what works for me, but I know that many people don't have the base knowledge to do it, so if they need a resource on how to become a better self-learner (or be successful at coursera), at least it is a start.

Testing/recall and spaced repetition well validated. Focus seems like a gimme. What are your specific problems?

Curious to know, what works for you?

Not the parent but the most powerful technique for me is definitely doing recalling, and for this you definitely need to have patience.

Totally agree. Besides, many textbooks are incredibly bad. Add in a low attention span like mine, and reading chapters start to end becomes an incredibly frustrating experience where I'm happy just to get through, let alone learn anything. That's similar to my issues with lectures, actually.

I agree. I always looked at the math questions first. Then I worked backward to figure them out.

> 9. Eat your frogs first. Do the hardest thing earliest in the day, when you are fresh.

I agree with that in general, but I do also find that it's much less intimidating to start out on something easy (so you're more likely to start) and also helps you to ease into a state of focus. In college I would usually start with a reading (while taking reading notes) since it was something I knew I could always do without too much fuss. Afterward, it was much easier to start in on thornier stuff.

I think that once one has accepted that he/she must start at that moment no matter what, and it will a long time till finishing, then starting with the hard stuff makes all the sense. You might be feeling drained by the end, when you were to take on those hard problems/subjects. Or the easier stuff you can attack at other times, during breaks, while it might be difficult to advance on the harder stuff without a bigger more comfortable time frame.

Thanks for sharing! Is there anything like Anki out there, but it's a SaaS product and available both on my phone and on the web?

Quizlet, but I personally still use Anki because it's free and it works on my phone.

Anki includes a free sync feature to keep you desktop and mobile device synced. The only cost of Anki is the one time $25 cost of the iOS version. You can get around this by using the we version on your iPhone.


>Not checking with your instructors or classmates to clear up points of confusion.

Having spend much of yesterday grading, this one rings true. As a professor it is incredibly disappointing to get an assignment from a student who clearly didn't understand the assignment and decided to fake it only to do the assignment totally wrong.

Students, if you are confused: 1. read the directions 2. look on the syllabus (I get way too many emails with questions that are answered in the directions or the syllabus) and then 3. ask the instructor

Which type of learning are these meant for? They seem geared towards a very narrow range of subjects and a very particular testing style.

Example: "Don't repeatedly solve problems you already know how to solve." That's great if the test is only looking to see whether you know how to solve the problem. If instead you are going to be tested on how quickly you can solve multiple iterations of the problem, extensive repetition is a necessity.

There are also times when passive reading wholly absent understanding is not a bad things. This is particularly true in law and history. Any law student who properly understands an assigned reading prior to a lecture need not bother attending class. Often you must just read and retain material on the expectation that it will make sense later. Either it will be explained in person, or at some point you will attain a critical mass of knowledge. At law school that is normally at the start of the second year, when you start making links between the various disciplines and suddenly it all starts making sense.

My problem with online courses boils down to the "sticking around issue", as I like to call it. You start, you get mildly bored, most of the time you're too drained by work to really feel like you're learning.... aaand you drop it.

What's your solution to that? Enrolling into paid courses only? Approaching courses with a different mindset altogether?

A lot of things can be done, actually! First, figure out if you really want to take the course. Why are you taking it? Does taking the course align with the goals most important to your life right now? If not, then this resistance is just your minds' way of telling you not to waste energy on something not that important to you.

If it is something that will help you achieve one of your most precious goals, then think about why you have that goal. Then when it's time to watch the next video, or do the next piece of homework, think of that reason. If you're mildly bored and don't have a compelling reason to keep going, you'll lose focus quickly. If you have that "why," you have a reason to keep making progress even if you're bored or tired.

A support group or even just an individual can also help. Get someone to take the course along with you, and there will be tons of benefits. You'll see that it's possible, you won't want to fall behind on conversations with your friend, and you will enjoy it more. Failing that, at the very least have an accountability partner who can help remind you of why you have this goal, and check up on you. Having someone call you up and ask "did you finish lesson 3 this week like you said you would?" can be a great motivator.

Those are all short-term tips. If you want information on how to increase your willpower overall, I recommend reading Kelly McGonigal's The Willpower Instinct, a very action-oriented scientific look at willpower.

A big problem lies from the MOOC end, not the student end, in unnecessarily enforcing a schedule and penalizing those who fall behind. That's a huge detractor to catching up and finishing the course when "life happens".

If I could resume at my own pace, I'd have finished courses that I gave up on.

+1. I have found myself completing courses when I can choose my pace. There are times when I am done with project work, and I merrily chomp through a course or some difficult task during the weekend.

I wonder if this schedule is a carry over from how college courses are run. Perhaps there are those who need the discipline of a schedule. It'd have been nice to have a choice between a schedule as well as an "at your pace".

I'd like to see study communities built around topics and not specific courses. One of the best ways to stick with something is to have a groups of friends doing it as well (think of gym buddies). The problem with many MOOCs is that the community is only connected to the particular course (even a new version of the same course will have a separate forum), and if you fall behind a bit you end up being on your own (and then many people lose motivation and stop).

Working on my startup, I have thought a lot about this. Would the following be useful to anyone?

-create a linode, AWS or DO instance.

- Have trusted friend config wireless router to use new instance IP for DNS.

- Run modified nodejs dnsjack on new instance.

- Add my SmartMadre traffic control code to instance. Config oAuth or our APIanySite

- Now, code will cut off non allowed sites after X mins. SmartMadre code requires some amount of measurable progress to allow full internet back

- give trusted friend control of instance. You and friend could get email showing amount of daily progress.

- Suggestions welcome!

The problem then becomes that you get bored halfway through setting this up ;)

But you raise an important point. Doing it with a friend. It helps so much because you motivate eachother and there's also a bit of friendly competition and raising eachother's bar.

I watch standalone lectures, because there's no commitment to keep going. You trade a wider variety of content with less depth in one area.

My experience is that I don't remember much if I don't commit to the course (and actually, even then I don't remember as much as I'd like!). It really pays off to do the assignments and to pace of the course.

In my opinion, the challenges of self-teaching are to be able to study regularly over a longer period of time, and not only read/watch things but do exercises. Those things are enforced at school, but can be difficult to reproduce in a less constrained setting. For that, the MOOC are quite useful. My only grief is that I usually don't like the videos. I'd be happier with a book.

Yeah, I think that if I were trying to learn a new skill, exercises/books are definitely better.

I'm usually listening to things while cleaning, driving, etc, for entertainment rather than maximizing learning. My experience has been that if someone introduces a really novel concept or historical fact in a talk, it will sink in and stick with me for a long time.

I've found a lot of fantastic material in this university's lectures- http://www.gresham.ac.uk/

I'm working on a search engine to curate a wider range of these, although it's still very much a work in progress-


That's amazing.

There seems to be a bug with the filter on the left side: after you select one and the page reload, subsequent attempts to filter won't work.

I actually have been working on a similar kind of search engine, albeit being a more "proper" search engine + crawler combination rather than curation likes yours. It's great to see other works like yours for inspiration :-).

Awesome, thanks for pointing that out.

Just bookmarked both sites! Thanks :). Always looking for more material while driving, and podcasts are too hit or miss for me these days.

Would love the ability to search through these materials that are audio only.

It actually has that - if you're not seeing the Audio/Video facet Cloudflare might have an old copy of the javascript cached, I'll investigate.

I'm working on adding a feature so that every search result can be subscribed to as a podcast as well, so you can listen to everything on one topic by different people.

Oh yes, you are right. I missed it way up top the list of filters.

I'd love that podcast subscription capability, though listening to it in HTML5 means I can speed up the playback without having to download the audio file first (my podcast app, PocketCast, doesn't let me do this).

A while ago, I was thinking about doing a search engine on top of audio. The startup deepgram seemed to offer interesting tools, but didn't get a chance to play with it: https://www.deepgram.com/

Nice, thanks for this.

Currently I have it searching transcripts, but I want to get machine transcriptions in as well, because it should work better how often someone says 'um', which I'm going to try as one of the ranking factors.

I've currently finished Andrew Ng's course on machine learning, this worked for me:

1. Set a clear goal, e.g "By the end of September I want to work through the course solve all the exercises".

2. Tell another friend what I'm doing and make a bet, every time I don't finish the weeks assignments in time I'll pay 50€.

3. Plan the evening before at what time I'm going to sit down and do this course.

Find one or more people who will take the course along with you and commit to meeting regularly to discuss the class materials. By introducing deadlines and social pressure, it helps to get past the motivation issues.

I wonder when MOOC providers will start providing "study halls" using WebRTC video...

I had that exact problem with a statistic course on udacity. What I ended up doing was committing to spending a certain amount of time (about 2/hr/week) on that course, then made a beeminder goal to charge my credit card if I failed.

It worked in that I stuck with it, but the amount of time I spend on the original course didn't divide by two hours so I did also spend a little bit of time on another course.

/at this point I should clarify I am an unpaid shill for beeminder, not a paid one.

This is the main enemy of online courses, its not the competition, its the the lack of compromise from the user in part because everything is "virtual" and you don't feel any major regret for dropping out.

Coursera and all the others should focus on this part of the problem, not by asking for more money, that will just make them less likely to try again. Maybe calling the student after a few weeks of inactivity could be one way to help, another one may be to create and maintain local study groups (having free food is always a good incentive), another one is to be "paired" with another online student so you feel a little social pressure.

>another one is to be "paired" with another online student so you feel a little social pressure.

This is the biggest difference for me between courses that I have started and courses that I have finished. Meeting with someone in real life, even if it's just for an hour a week, easily creates the pressure to continue on that an email reminder simply doesn't do.

I don't think measuring completion is a good metric. I don't see the value in a certificate, so I never buy them. I've taken a lot of courses that I haven't fully completed and still got a lot of value out of them. Sometimes I get bored or busy and don't complete them, sometimes I'm only interested in parts, sometimes I don't want to do a pile of homework on a topic that I'm just looking for a refresher on.

I think the Kahn academy model is a much better one where you complete things in much smaller units and can pick up where you left off, but it doesn't really get into more advanced material.

I see people say there's value in completing the course with others, but I value the flexibility.

"The Achilles heel of the MOOC phenomena has been that while enrollments have been huge, the number of students who actually complete courses for credit has remained low."

Of course completion is low, there is no cost to signing up so many people will do it on a whim.

The real Achilles heal is certification, for it to be valued employers need to be convinced you aren't cheating and that the assessment is rigorous.

Also, I don't care about the entire course. I want the specific information that is relevant to my problem at hand. I'll be the judge of which lectures satisfy what I need.

I find it a shame that MOOCs these days are trying to duplicate the brick and mortar experience, because they really could be offering so much more.

I signed up for the course in question - I think it was my first interaction with a MOOC - and lost interest very quickly. Realistically, I need the abbreviated version rather than the intro videos and the opportunity to meet classmates and so on.

Does a MOOC have an interest in creating longer courses that seem like they're worthy of certification, rather than just teaching people quickly and effectively?

I signed up for a ton of courses and dropped the majority. For me, the quality simply wasn't there. I have no way of objectively verifying this claim, but imho if the quality was there completion rates would be much higher. As is the majority of these courses just aren't worth the time.

I avoid MOOCs lately as they are often too slow paced, or lacking in specific and/or advanced material and instead find a university with a public accessible calendar and cherry pick what I'm interested from there. For example this excellent course from CMU on the internals of modern database management systems. https://www.youtube.com/playlist?list=PLSE8ODhjZXjbisIGOepfn...

> The real Achilles heal is certification, for it to be valued employers need to be convinced you aren't cheating and that the assessment is rigorous.

I really don't think this is it either, because you have to be naive to think there isn't massive cheating that goes on in traditional universities. I would venture to guess that cheating is perhaps lower currently with the students who complete MOOC's because the fact that there is no current incentive to do so. Also it's not that it isn't rigorous either -- at least for some of the EdX courses they are almost 1:1 translations of the OpenCourseWare versions, which themselves are just the actual classes they teach at MIT.

A majority of the people simply don't like to learn anyways, even if they try. Even if there's a perfect MOOC out there most students don't complete the course because most people simply aren't students, or they're not cut out to be students. So the failure rate may not be the fault of MOOCs, but simply how many people in society work.

With that said I do believe that they should be valued and have a lot more credibility.

I took this course about a year ago and really enjoyed it.

The main things I took away were:

* focused and diffuse thinking - I now take short walks at work if I'm struggling with problems for a bit and often find I come up with an alternative approach when I'm not focusing as tightly

* repetition and recall - I take notes and then create flash cards to force recalling information

Additionally, learning the science behind why you lose information so fast after a cramming session has been helpful too. I really wish I had learned this info before attending college, it could have made a big difference for me.

I'm currently taking this course for the second time. The first time I enrolled I didn't get around to watching any of the lectures. Luckily, I ran across a reminder email from the course and it just so happened that I had enough free time to re-enroll and complete the first week's lectures. It's really difficult to feel invested when the cost to sign-up is $0. They do try to create a community around the course, but it's so distributed that I wouldn't consider it a cohort in the traditional sense.

What's your feedback on it so far? I saw the course a few weeks back and it seemed kind of gimmicky so didn't dive deeper.

I finished it a couple of weeks ago, and it's really worthwhile.

It did seem a bit gimmicky and superficial at first, but once the course progresses, you realize that the material itself heavily applies the techniques that they describe in the course (e.g. spaced repetition, visual associations).

> Do the hardest thing earliest in the day, when you are fresh.

I do the opposite. I do the easiest things first, so I can concentrate on the harder problems without the easy ones being on my mind.

Also, I just don't work that well in the morning. Later in the day is best for me for my most difficult tasks.

I took the course 2 years ago and really enjoyed it. I didn't take a lot of work, yet the content was relevant and good.

Do you still remember any of the course?

Not that much. I remember the importance of sleep when you're learning (well explained in the course), and also spaced repetition and the importance of recalling the material on your own.

I also remember that the instructors were very good - positive and enthusiastic, and that they explained the concepts very well.

In general, it's hard to remember contents of courses I think. I often try to write a review of the course, either as a blog post, or on one of the review sites (Course Talk or Class Central). The act of deciding what was the most important parts, and expressing it in words go a long way towards remembering better.

Examples of reviews:



I took it after the class has ended, still remember the content. It seems obvious, but it pays to hear and see it reiterated by other sources.

Get enough sleep.

Learn to chunk large bits of information.

Get a high level overview before diving into the small details.

Learn to use things you already know to make connections to new topics.

Don't over learn, cramming in one go rarely works, it's best to study, forget about it, come back at it, it will stick with more exposure. meaning, get the knowledge into your long term memory.

Use pomodoro technique, to beat procrastination.

I took this about 2 years ago too, and remember a little about the major topics and some random facts about the instructor.

Its hard to guage how it might help with motivation to get started with MOOCS, but I'd say it's more valuable than all the content just to remember for any topic: "don't believe you know it (or waste time passively studying it) unless you test yourself on it" and "use spaced repitition or you will forget it, but think you still know it."

I followed the course around the same time (i.e. watched and read the material but without doing any of the work).

I don't know that I can recall many specifics, but the most noticeable effect of the course for me was a greater awareness of what state of mind I'm in when approaching a task, how appropriate that state of mind is to what I'm trying to achieve, and how my choice of activities and working methods are likely to shape that state of mind.

The article fails to cover that this course is actually short and simple unlike the other MOOC courses which contributes to the success of this course. Maybe there is a lesson here for other MOOCs which is make your courses more byte sized and simple so people can actually complete them instead of putting your class room course as is online.

I looked into this couple weeks ago; how is the course compared to Mind for Numbers?

Popularity is an odd way to rank online courses. It really only speaks to what is the most universal interest, so it's not terribly surprising that's a course such as this (as opposed to an area of speciality).

I looked at the article written by Barbara Oakley which was cited in the current article. [1]

MOOC deniers are asking if they a) adequately satisfy the requirements from education, an important part of which is the actual completion of the course and b) are economically viable.

Let us leave aside the question of whether you decide if it is a good idea whether to gamble at Vegas by talking only to those who hit the jackpot. [6]

Barbara Oakley's defense seems to be:

1. "Look at how many people can now learn this stuff" [2]

2. "Look at all the new and amazing things which happen when students can play, pause, rewind and resume the lesson, or when teachers can use all the technologies at their disposal" [3].

3. To fellow instructors: "Please try a little better. The problem might be the way you teach"

And remember that teaching someone how to learn is not exactly at the same level of rigor or conceptual difficulty [4] of teaching someone calculus. If you have ever seen a screencast of some really dense material [5] from some really smart people and see them struggle to "hold the stack in their head", so to speak, as they present their thoughts in public view, you know the process starts becoming seriously challenging as the subject increases in difficulty. Here are some quotes from Barbara Oakley's articles, which I think are good examples of things which don't scale very well as the material becomes more rigorous.

"Good online courses make students feel professors are speaking directly to them. A teacher’s direct focus on the camera translates as personal attention in the videos. Students develop a sense of familiarity; we are often seen as friendly private tutors. It makes us more approachable and “listen-to-able.” It’s not that we’re replacing teachers in a classroom. It’s that we serve as additional personalized resources, despite the fact that we’re explaining at massive scales. And I should mention that every single video lecture I give in our MOOC is the best lecture on that topic I’ve ever given in my life."

"Terry and I made “Learning How to Learn” for less than $5,000, and largely in my basement. I had no previous film editing experience—in fact, I could barely click a camera shutter. Much of the moving imagery for the course was created using simple PowerPoint slides. So I would issue a challenge to MOOC critics. Make your own online course. Film the most interesting, most insightful lecture you’ve ever given in your life. If you don’t think your lecture is good enough, reshoot it until you’re happy. Make your video available for millions of students around the world, not just the privileged few in your classes. Come up with questions for a quiz on the mistakes you most commonly see in your classes. You will learn more than you know about the outreach and capabilities of MOOCs. More importantly, you will exemplify a wonderful openness for learning to students everywhere."

[1] http://nautil.us/issue/29/scaling/why-virtual-classes-can-be...

[2] "She cited a range of groups who are promoting the course from the California State Prison System, federal K-12 teacher certificate programs, as well as refugee camps in Somalia and Sudan, where she asserted that students threatened to overwhelm the meager Internet bandwidth available in those countries." - from original article

[3] "I would venture to say most MOOC deniers have little experience with creating and teaching online courses. The reality is MOOCs can be artistically and technically fascinating and can have terrific pedagogical advantages. This is particularly true in the fraught area of STEM (Science, Technology, Engineering, Math), where difficult explanations often cry out for a student to replay a portion of a lecture, or simply to take a pause while comprehension works its way to consciousness. "

[4] For example, I would like to know how the students assessed the satisfactory completion of the material to decide whether they are, indeed, better learners. And then compare it with whether their actual learning rate improved in their future endeavors.

[5] https://www.youtube.com/watch?v=nfoudtpBV68&list=PL4LJlvG_SD...

Edits: some edits for readability

[6] I mean this in two senses. One, of course, is the choice of subject, among thousands of possible ones - something as meta as "Learning how to learn" has a few built in advantages when it comes to measuring popularity. Two, I don't mean it in the sense of a skill-less person getting the luck of the draw to achieve the desired result. I mean in the sense of how there are bound to be a few jackpot winners in a large enough sample, and studying them is misleading because the takeaway messages are not universally applicable.

Yes, a 4 hour seminar on study skills is more of a supplement than a unit of progress toward mastering a subject. So? It fills an import gap for students who didn't learn study skills in traditional school.

Sure, "Most popular MOOC" is a gimmick -- CGPGrey or Veritasium could claim that any of their YouTube videos are the most popular MOOC.

To really evaluate this course, they should compare students who take this course plus other, vs students who only take others. (Which has blindness problems, but it's a start)

I think this is a good list to take good course, I'll save it only for take a few.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact