TsviBT comments on What should normal people do? - Less Wrong

23 Post author: seez 25 October 2013 02:28AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (93)

You are viewing a single comment's thread.

Comment author: TsviBT 25 October 2013 06:38:39AM *  0 points [-]

Specific advice: [ETA: If you decide that it is worth your time and effort to work directly on improving your general thinking skills, then one difficult but effective way to do that is to] learn to program and/or to learn math. Use google to find resources. Don't be embarrassed by books/articles with "for beginners" or "introductory" or "elementary" in their titles, and especially don't be embarrassed if even those are too hard (in fact, beware of the good old "elementary" math text, meaning "elementary... for grad students, hahaha!"). Just keep looking until you find something you can dig into. Do lots of exercises; basically, do most of them until they get boring, but learn to feel the difference between being bored because it's easy and bored because you have no idea how to do it. Find someone who can check your understanding, or find books with really good worked out examples.

Learning a technical field (meaning proof-based math, programming, and to a lesser extent other sciences) - even the very basics - will help you think faster, hold more things in your head all at once, follow long and complicated arguments, notice flaws or vagueness in arguments, distinguish crucial structure from irrelevant details, etc. Accordingly, it will take a huge sustained effort.

General advice: find someone smarter than you, and ask for life strategy help. E.g., do you think you could do a non-glamorous but well-paid job and donate to charity? If so, brainstorm with google and/or a smart person for jobs like that. Maybe check out 80,000. I haven't really looked at that site much but it seems like it has some helpful stuff.

Comment author: Creutzer 25 October 2013 11:14:42AM 3 points [-]

Specific advice: [...]

This is supposed to be for a person of average intelligence? ...

Comment author: TsviBT 25 October 2013 08:28:11PM 0 points [-]

Yes it is. There was a big additional assumption I was making in my head, I've edited to clarify. Now does it make sense?

Comment author: Creutzer 26 October 2013 10:12:24AM *  0 points [-]

I don't think so. Now it basically reduces to the general claim "learning math and programming improves general thinking skills" - which, by the way, I'm not convinced of in full generality -, but has nothing to do with the average person. The problem is that learning programming and math takes so much time and effort, if it is at all possible, for the average person (and with no easily identifiable returns at that) that the antecedent of your conditional is unlikely to ever be true, thus rendering your advice largely irrelevant.

Comment author: bramflakes 25 October 2013 03:40:54PM 2 points [-]

You're significantly overestimating how a) easy and b) enjoyable the average person finds math-related subjects. Most people don't get past algebra.

I think a better option would be to gain a gut-level understanding of comparative advantage, and how much to value your time, so that you can get paid for what you do best, and outsource what you're bad at or find boring to other, more competent and enthusiastic people.

Comment author: TsviBT 25 October 2013 08:36:53PM 0 points [-]

b) enjoyable

In my head, I was assuming motivation, edited to clarify.

Most people don't get past algebra.

Yeah I know, that's why I commented. Even basic facility in proof based math is an extremely powerful mental technology, as I tried to say. I would not recommend calculus. I am talking about combinatorics or graph theory, or discrete math in general, where you can see the basic building blocks of proofs and proof strategies. This is worth years of effort.

Comment author: Gurkenglas 27 October 2013 03:27:44AM *  2 points [-]

Maybe proficiency in proof-based math is not a cause of mental superiority, but an indicator?