What works about on demand knowledge is that it is pull based (the knowledge you need, when you need it) and comes in digestible chunks. Unlike MOOCs, which are consumed far in advance of the knowledge being applied, Wikipedia and StackOverflow are the knowledge you need, now. Humans are lazy and working ahead requires discipline and foresight, which makes on demand knowledge far more appealing to most.
You may need to read the whole post. I thank Ben Casnocha for the pointer.
It has struck me that the the traditional notion of a course and the medium of online learning may be misaligned. Try to imagine what would happen if you got rid of courses. What would you do instead in order to provide students with direction in their learning?
I take issue with the humans are lazy part. What is more effective, a yellow brick road, or a tank tread? The current education system is like the yellow brick road.
Arnold, assuming you tend to agree, how did we get so far away from the virtues of on-the-job training and on-demand learning? Is it 3rd party payer?
I have studied the Toyota production system and I have experience in applying lean thinking to software development. One thing I have learned is pull almost always is better than push. Thinking back to my masters (20+ years ago) I think it could be improved considerably. I bet it is thought the same way today and that is not much progress.
I agree, for most cases.
But there are the rare exceptions like the apple i-phone, where the customer doesn’t know what he wants until he sees it.
Don’t Cambridge and Oxford with their tutor based systems provide long standing example of how to get rid of classes.
It reminds me of one similar (I guess) experience. Probably the best class structure I ever had was a lecture, then a giant homework problem that would take you all week. Then we had to write the solution on the blackboard to be learned/critiqued by the other students and professor. Not wanting to look like a jackass focuses the mind.
The funniest story from that class is I went to ask the professor about a problem and had to sit there awkwardly while he fumbled through it for an hour.
Youtube.
There are some apprentice like experiences to be had by self directed study (e.g. working on what you want to do or get paid for) backed by studying general sets of videos by workers in the field.
For some topics, certain high quality bulletin boards, practicalmachinist.com being an example, can fill a similar role.
The risk of all of this is missing “deep theory” that a course might focus on. What confidence-interval or significance testing really mean, the proven information theoretic limits of sorting algrorithms, all of the rare but important diseases that might *also* cause a certain rash, etc.
Is there any evidence or even hypotheses as to whether this deep theory is a profitable investment in real productivity?
As I always say, I was taught a ton of deep theory, then I got a job doing completed, like, totally different stuff. The deep theory for that I either didn’t need or picked up over time and on my own.
I think we are moving toward school being even more for testing than it is already and youtube and wikipedia for learning (education).
Is pull the apple to push’s orange?
Stackoverflow and Wikipedia mean I can work out all sorts of practical problems outside of my main fields of knowledge, and yes, they can both remind me of things that I should know but forgot. The traditional technology for that stuff was not the university course, it was the Big Black Handbook.
I am not sure if people without a good push-based education can make much use of the pull technologies. That doesn’t mean that Universities are the only way to get the background, it just means that StackOverflow and Wikipedia are not replacements.
Instruction should be couched in meaningful games that illustrate the power of the concepts.
> You may need to read the whole post.
Instead I’ve pulled some parts of the text that are relevant to me and discarded the rest.
Courses are generally designed to be shallow but thorough to expose the student to a body of knowledge. That approach always struck me as poorly designed to create mastery or wisdom. Instead it creates a shallow and quickly forgotten command of information. A different approach is to create mastery or skill or understanding. Almost no textbooks in economics do that. The obvious way to get there from here is problem-solving which has the potential to hone skills and insights that might apply to a range of other problems.
As usual I agree. My field is of expertise is software development. I find the recent college grads are only as good as how much time they put into non-course work, open source projects or game development. Books by design are not interactive and CS courses are not much better. Computer programming is highly interactive and when done well informative.