How To Build MOOC's that Fail

Having started a half dozen MOOC's in the recent months, I have found most of them tend to share a common trait. Many MOOC's currently represent a sort of parody of higher education's worst practices, its most spectacular delusions about itself.

And thus they tend to fail--some rather spectacularly.

(In the interest of protecting the guilty, I won't name specific courses. I have no interest in insulting people who are surely earnest and well-meaning in life--they just happen to lack any experience putting a course online, let alone a MOOC.)

For the sake of neatness, I'll organize my thoughts here on four's.

For the same of keeping the reader interested, I'll frame everything ironically: trying to articulate the unspoken assumptions which make so many MOOC's so very dreadful.

I'll start with the Four Delusions of Higher Education. These underwrite the Four Rules for MOOC Failure.

  1. Sink or Swim. This is simple. Provide no guidance. Don't tell students where things are. Make the goal of the course as mysterious as possible. Give the students "freedom"--like plunking a traveler down in the middle of a strange city.

  2. The Heliotropic Professor. The professor is the center of the learning universe. Everything revolves around the professor. The professor is the sun, and the students are tiny tiny planets--really cosmic dust, basking in the glow of the professor's expertise.

  3. Nothing happens without the professor's instigation. The professor must frame everything, explain everything. Students must do 'exploratory learning'--so they can then find out what the professor thinks, which of course is always right, since assumptions and standards can never be explained. (That would undermine the professor's mysterious sole access to True Knowledge.)

  4. Go Figure It Out. No matter what is said, no matter what is asked for, no matter how unclear or obscure, ultimately the student will just have to Go Figure It Out. After all: they're learning, aren't they? We can't make it too easy for them--like explaining what they should be learning. Since in the end, we are all just lonely particles colliding against each other randomly, why not just make the student responsible? Surely they will thank us later.

  5. The Piehole Illusion. Anything that comes out of the professor's piehole can be learned. The professor can say "2 + 2 = 4" and that will be learned--without the student needing to know if she is learning a fact, a rule, a concept, an allegory, etc. The student may have to listen again and again, and that is all to the good, because what the professor says is so very rewarding.

If you accept all four of these precepts, it will be very easy to make a MOOC that fails utterly.

Fails to help a thoughtful person learn anything--because nothing is specificed in the way of learning: nothing about what is to be learned, to what standard, or how.

But if you need to operationalize this knowledge--and here I go further than almost any MOOC instructor does--you may follow three simple rules.

  1. Conceptualize your course as content. Just imagine all the things you need to tell someone. Then record yourself saying them. That should be enough. If you then add some articles people can read, surely no more is necessary. You will then add a commentary 'explaining' everything--so students understand fully that the professor's view is always the right one.
  2. A course, after all, is just a pile of facts & ideas. The student will just have to go figure out what it all means, how to fit it all together, and above all, how to learn it. (Helping the student learn can't possibly be the instructor's job! The instructor deigns to share his wisdom, and his job ends there. Teaching is like grace: you don't ask why.)

  3. Don't plan any learning activities. Since your course is just content, it can't possibly matter what the student does to learn. Learning is the student's job. So just give them a sandbox and say--go learn there. Don't tell them what to do, what they'll need to practice, nor how anything need be done. Throw up a discussion board and say "talk amongst yourselves!" Done!

  4. Don't consider pre-requisite knowledge. There's no point in worrying if students are ready or not. It's all sink-or-swim, so just throw them in. No pre-tests. No lists of things they might need to know or be able to do. No expectations.

  5. Assume everyone taking your course, no matter where she lives on the planet, is exactly like your current undergrads who pay tens of thousands of dollars for your institutions courses. Assume that nothing about the learner makes any difference.

In short: univeralize utterly your tiny corner of bourgeois North America. Whatever you do: don't reflect on social, national or cultural difference, nor on how well- or ill-prepared your learners might be.

Maybe one addendum.

  • Don't hire any instructional designers. No good can come from carefully selecting and arranging carefully designed tasks in a sequence so that the learner is prepared to succeed. (Remember, it's better for the learner to fail--since this proves how terribly complex the subject matter is.) In the 21st century engineers will solve all our problems--you know, like psychoanalysts did in the first half of the 20th century, and nuclear engineers did in the second half of the 20th century. Humanity, you see is not involved. The mind is just a bunch of wetware: so surely engineers know and can do everything that's needed.

Clearly the subtext here is: an effective learning experience, especially one mediated by time, distance and computer technology, really needs to be designed.

If you throw a bunch of content out and say "learn this," the learners will get out exactly as much effort as you put in--which is next to none.

So what you have in front of you now is a really a plea for instructional design. And I know that has issues.

When professionals get together, the thing they talk about is: how everyone needs them so terribly badly. Dentists wish everyone flossed. Doctors warn about germs. I once had a landlord who was a plumber: he insisted a drain should be cleared with baking soda--lest you harm the terrribly delicate lead pipes.

Such is the definition of a "profession." It's a specialized kind of knowledge. And it has to value itself and therefore to devalue anyone who doesn't have its special knowledge.

But if you really want to make a MOOC that makes a difference, if you actually want to share knowledge for others to learn it, not just to show how classy your institution is that it can give away courses (really meaning: lectures), just do the opposite of what most MOOC's do.

  1. Conceptualize your course as things people will be able to do afterwards--with standards attached. A gradated series of tasks and activities is even better. But a final demonstration, with lots of small steps leading up to it will suffice.

  2. Plan activities which will support learning--which let the learner practice and get feedback. Be aware that not everyone will do these activities. Consider making them inherently interesting. Begin with tasks that can be done without specialized knowledge. Maybe even use tasks that have real-world implications--which people find intriguing. And be aware that the students can't possible mentor each other, since their expertise has never been assessed.

  3. Help the learners discover if they are prepared to take the course. You might use a pre-test or a checklist. If this material even links to preparatory activities, you can kill two birds with one stone.

  4. Consider hiring actual instuctional designers: not engineers, not programmers, not someone with a few education courses. Really take seriously that you are designing something. Plan. Make a proof-of-concept. Start small. Scale up. Do all the things that are considered "design thinking"--even if you are not a designer of any sort.

In short, consider that plunking things on a web site is not design--any more than throwing furniture around a room is interior decorating.

Or do what everyone else is doing: record a few lectures, throw up a discussion board, and break your arm patting yourself on the back.

--Edward R. O'Neill


  1. Hi Edward,
    I blogged about my mooc 'issues' recently ( Having studied with the open university here in the UK, your point about preparing students struck home because the current MOOCs so obviously don't. I've enrolled on a few of them now that are badged as 'introduction to ...' and state no prior knowledge required or 'maths at level x' needed, and then completely ignore that once the course starts. I've shared your post via Google+.

    1. I like your blog post, Duncan. I've received emails from Coursera professors saying, in effect: please ignore when I make a mistake.

      It's difficult, because lectures are oral! They're not perfect. But maybe then they shouldn't be the *primary* content in a course....

      Your UK open university approach is older than the MOOC and thus has built up some experience.

      Sadly, what's "new" is often just a forgetting of what works....

  2. I am taking my first Coursera-based MOOC (out of curiosity--I'm a retired college professor whose livelihood does not depend on it) and everything spoofed here is true for this course.

    I do enjoy the course, in part at least, because I have met some very interesting people on the forums. But that cannot be the sponsoring university's goal (UMich in this case)--it's like saying, "I took a cooking class. I didn't learn how to cook, but met a lot of women."

    1. Your analogy is brilliant. Made me laugh out loud. Hope you don't mind if I share it. I won't claim ownership :-)

  3. I wonder how/whether it is possible to build technology specialized for MOOC that takes into consideration the best of instructional design for perhaps a limited subset of use cases. That way the instructional design is baked into the course. I think the merge of education technology and instructional technology could be powerful.

    1. @E Aboyeji That would be wonderful. Sadly, some of the most prominent MOOC producers seem to have no clear idea about what the best use cases would be.

      LMS's also tend to be a Swiss Army Knife--one tool containing many tools--but most of the tools are mediocre.

      I actually think a clever solution (which goes back to the earliest MOOC's) would be to 'squat' on existing social media platforms, and discuss there!

  4. Navigation, navigation, navigation. I've started (but not finished) two MOOCs and spent most of my time being lost in poor LMS or website navigation. I love the "Heliocentric Professor" image. I still think MOOCs have the power to give access to some wonderful teachers and subjects (particularly practical ones - show-jumping design, anyone?) and it's early days...

    1. Imogen.

      Your navigation point is spot-on, I think.

      If I can't figure out where things are, learning will be hard.

      I will say: *some* MOOC's have better communicators in charge. So I'm just exaggerating the bad trends.

      I also hope some good content comes out!

  5. The Web is full of Good and Bad content, MOOCs are no different, or should I say, why should MOOCs be different? some will fall and some succeed! Take a look at OLDs MOOC, a collaboration between the OU and JISC, now thats a MOOC :)

  6. So if I get this right, instructors, teachers and professors are just useless egomaniacs and the world should revolve around content and instructional designers. We already have the perfect MOOC by your definition - it is called Pearson's MyMathLab.

    1. Dr Chuck, I don't think that's what this post is saying at all. In trying to create a MOOC, one cannot overly on traditional modes of teaching. With new technology comes new opportunity. The idea is not necessarily to throw away the baby with the bathwater but if you try to keep both the baby and the bathwater you miss the point altogether./

  7. One thing that was not mentioned here is that due to the nature of a MOOC, with hundreds and hundreds of students, there should be a very strong way of integrating the community so that they can help each other out.
    When we created StudyRoom ( we had that as our goal and with the first study room that we offered for a Coursera class the engagement of the students has been stunning now that they finally have a place where they can meet their classmates and learn together.

    1. Very true, Emerson! An earlier generation of open courseware emphasized the social layer--and let it be *anywhere*. I don't think Coursera has a handle on this. But I hope someone figures it out!

  8. Edward, thank you for this post. Especially your Four Delusions of HE, cheered me and a few fellow suffers up no end. Shared pain. I would like to reference (shamelessly plagiarise) your notes. I will cite my source. Regards Peter email

    1. Thank you, Peter! I've been wanting to publish those Four Delusions for some time. *Not everyone shares them.* But I think they're pretty deadly for learning.

  9. Thanks. Shared here:

  10. I would add in the delusion that Love of Learning and Desire to Increase One's Knowledge is enough to motivate thousands of adults to give up personal and professional time for your course, without any other motivational factors. Even if it is a rather arbitrary certificate of achievement or something similar. I worked my tail off in Harvard's CS50x MOOC (shameless blog plug: ), where there were required assignments and deadlines, and I would get a certificate if I completed everything. Now I'm in a cafeteria-style "Take what you want and leave what you don't" MOOC that I can't find the time to engage in, because there is no bite to it. I'm also not getting anything out of it, other than learning about some things that I could probably find on my own if I really wanted to. MOOCs, like any other course, should have accomplishment goals, deadlines, and nominal rewards.

  11. Fantastic post! Thank you for the insight, and of course the quality of the writing


Post a Comment

Popular Posts