[LINK] Mastering Linear Algebra in 10 Days: Astounding Experiments in Ultra-Learning
Scott Young completed the four-year MIT computer science degree curriculum in less than one year. This is a post about how he did it.
During the yearlong pursuit, I perfected a method for peeling those layers of deep understanding faster. I’ve since used it on topics in math, biology, physics, economics and engineering. With just a few modifications, it also works well for practical skills such as programming, design or languages.
Here’s the basic structure of the method:
Coverage
Practice
Insight
While I find what Scott Young is doing very impressive, a lot of his presentation is reminiscent of [a type of] marketing that makes my skin crawl. I half expect him to star in his own infomercial where he points and shouts aggressively at the viewer, surrounded by people doing sit-ups while reading Learn C++ In 21 Days.
EDIT: clarity.
Is marketing always bad? (Non-central fallacy.)
That was “is reminiscent of [a variety of] marketing that makes my skin crawl”, not “is reminiscent of [all] marketing [which] makes my skin crawl”. Edited for clarity. My skin has no reaction to marketing in general.
From the discussion on Hacker News, the linear algebra final he passed. I can’t say I’m terribly impressed.
EDIT: The rest of his final exams are better, but not exceptional.
If I took all the time I spent playing video games when I was in college and instead spent it doing schoolwork, I probably could have done something similar, as I was generally able to learn a semester’s worth of stuff during the “crunch time” before exams. Playing video games was a lot more fun, though, so I have no regrets.
Crunch time motivation is very high quality and not trivially replicated. So I’d be impressed if you managed to pull this off in practice.
(BTW, I recommend students plan to do things during crunch time by default, or at least experiment with this. You’re going to have an extremely high-quality source of motivation if you just wait a while—why not take advantage of it? If you want to work, and you have no imminent deadlines, either work on whatever you feel like working on AutoFocus-style or, if your energy level is high enough, work on some independent project that has no deadline—your opportunity costs are lower this way.)
-- Bill Watterson, Calvin and Hobbes
Because things learned during crunch time seem to fade from memory much faster.
Well, it’s not like it’s forbidden to revise stuff after an exam (e.g. using SRS) if you care about it for reasons other than the exam itself.
Fair enough.
Oh no. Now I have a perfect, bulletproof excuse, that I actually buy, for my habit of procrastinating so badly with assignments that I typically end up doing them in a modafinil-powered all nighter on the night before they’re due.
John_Maxwell_IV, what have you done?
Note that, in practice, the point at which crunch time begins depends on your estimate of how long it will take you to learn the stuff you haven’t learned yet. Depending on my courseload, it would be as long as a month or as short as two weeks.
Scott Young is also currently (i.e. started yesterday) running a “bootcamp” email course on productivity. First instalment currently publicly visible here, but that link will go out of date.
Your comment prompted me to sign up to his list, which got the link to his ebook on holistic learning (PDF). He posits holistic learning against rote learning, which appear to be the Programmer’s Stone concepts of mapping versus packing.
Does he mean natural languages? I’m willing to believe that the guy found a super effective method of gaining deep understanding but in language learning there’s no such understanding to be had. There’s no grand insight from which you can recreate the entire dictionary.
By the lack of an Oxford comma, I infer Scott means computer languages.
Since we’re talking about a computer science degree, I assumed he was talking about computer languages (though I had to think about it for a bit too).
What about individual IQ? It’s not at all clear that learning methods yield uniform results across the bell curve. What might work for a 130+ IQ individual may not work for a 110 IQ individual—and vice-versa.