Tyler Cowen’s latest blog post gives advice for learning about modern China. He says that “books about sequences of dynasties are mind-numbing and not readily absorbed” and recommends finding other entry points before reading about dynasties.
Find an “entry point” into China of independent intrinsic interest to you, be it basketball, artificial intelligence, Chinese opera, whatever.
In a podcast interview—sorry, I no longer remember which one—Cowen talked more generally about finding entry points or onramps for learning big topics. The blog post mentioned above applies this specifically to China, but he gave other examples of coming to a subject through a side door rather than the front entrance. If I remember correctly, he mentioned learning the politics or economics of a region by first studying its architecture or food.
I’ve stumbled upon a number of intellectual onramps through my career, but I haven’t been as deliberate as Cowen in seeking them out. I had no interest in medicine before I ended up working for the world’s largest cancer center. I learned a bit about cancer and genetics from working at MD Anderson and I’ve since learned a little about other areas of medicine working with various clients. Right now I’m working on projects in nephrology and neurology.
Applied math is my onramp to lots of things I might not pursue otherwise. As John Tukey said, you get to play in everyone else’s backyard.
There are many things I’ve tried and failed to learn via a frontal assault. For example, I’ve tried several times to learn algebraic geometry by simply reading a book on the subject. But I find all the abstract machinery mind-numbing and difficult to absorb, just as Cowen described his first exposure to Chinese history. If I’m ever to learn much algebraic geometry, it will start with an indirect entry point, such as a concrete problem I need to solve.
When people ask me how to learn programming, I tell them to try to write a program to solve a problem they want to solve. I’ve read a number of books on programming, but nothing sticks like learning how to solve the problems you’re actually interested in solving. A C++ book I tried to learn from used people as a class, then inherited that class as an employee and I could not care less. I ended up learning C++ by solving problems I needed to solve at work or wanted to solve in my spare time.
I think people think it would be neat to learn programming because they think they should know how to do it, but learning something you’re not interested in because you think you should learn it is a great way to get bored. Worse, it could teach you learning isn’t worth the trouble. Learning to program because you think you should know how, instead of learning it in order to solve a problem you’re interested in, will lead you to be bored with computing and even possibly hate it.
There’s nothing wrong with not learning something. We have a finite amount of time each day and are guaranteed to die at some point. Use your time wisely. No one dies regretting not having learned to program.
Ask a teacher. K-12 education has undergone massive evolution over the past 2-3 decades (yes, and a minority were fads). One key change was finding relevance, to the STUDENT, between what the student doesn’t know and what the goals of the class are.
This is time-intensive, but worth the effort when the student launches into becoming a self-learner. Many tutors have known this forever. But it has taken decades to percolate upwards. It’s entering universities now. Before long, these new graduates will be able to synthesize it for themselves.
As for us older codgers, we may need to ask them to teach us how.
Algebraic geometry was developed by people noticing cool things about polynomial equations, and this is a good way to learn it. But that didn’t work well for me, because I wasn’t very interested in polynomial equations. So I had to find my own onramps:
https://twitter.com/johncarlosbaez/status/1081234511625564160