Stuff I Learnt in 2019 | Hacker News Reader

Stuff I Learnt in 2019

bollu.github.io

152 pointsyarapavan28 comments

For general relativity: I never miss an opportunity to recommend "Spacetime and Geometry" by Sean Carrol. The only book I needed for my GR course (plus Misner Thorne and Wheeler for occasional reference).

Your learning rate must be off the charts to accomplish this impressive list.

IIUC this entire blog is one long README.md, sorted by recency. I understand that it makes it easier for the author to post but it's not the most user-friendly.

Do you have a better recommendation for ease of use v/s easy to browse? What do you think of having an auto-generated "table of contents" page? Does that sound good?

That would definitely help. I understand you though: writing is hard and it makes sense to want as little friction as possible.

@bollu, Congratulations! Impressive year by any measure.

Could you go into some detail with your learning process?

Such as how much time do you have to pursue all these subjects? How you manage and allocate that time?

Maybe provide a case study of your approach to reading 'Information Theory, Inference, and Learning Algorithms'? how long did it take you to finish this book?

Do you typically complete all the exercises in the books you read? Do you try to reproduce each paper?

And any other relevant details you think might help others to "learn" quicker would be appreciative.

> I recently learnt that the Toeffili and Hadamard gates are universal for quantum computation. The description of these gates involve no complex numbers.

Hmm, these are matrices so essentially complex numbers?

Correct. You can pretty much always replace complex numbers with 2x2 matrices.

In case the author is reading, I'm curious to hear how you structure your day and what your learning routine is like. Any tips you can share?

I like this. I think I am going to try to keep a list next year, and maybe it might somehow even help keep me on the path of learning.

post the `https://` part too. I don't have https everywhere everywhere.

Learned quite a bit about: K8S, Kotlin, Android Architecture Components.

Maybe you could fit in some Category Theory in 2020.

Hey, author here! I do know some category theory, as I dabble with Haskell. Another motivation for me to study more algebraic topology & geometry is to see category theory in the setting where it was originally developed. Do you have recommendations for what results from Category Theory proved fruitful for you?

For me as a programmer, it helps me to think more abstract about the problems that come up. I apply CT similar to the way other programmers apply object-oriented design patterns. I think it's a great framework of thought.

What do I need to do to get on this level?

i started learning judo.

Fan fact - John Carmack is a judo and jiu-jitsu player [1]

1. https://www.youtube.com/watch?v=We1wlYOP0Xg

I need to study more

I hate “learnt”

What do you hate about it?

It sounds like an "ain't" version of learned. I just read this from Grammarly:

"Learnt and learned are both used as the past participle and past tense of the verb to learn. Learned is the generally accepted spelling in the United States and Canada, while the rest of the English-speaking world seems to prefer learnt."

Apparently I'm an egocentric American.

Yeah, we were taught "learnt" in school (British English).

Some kind of focus would do this young man well. Also thinking how to turn that all information into action or outcomes.

On the contrary, I think it is completely fine to learn for the sake of learning. It doesn't preclude future action, in fact, it greatly helps it. And at young age, with probably limited responsibilities? A no-brainer the way I see it.

What would you recommend I focus on? I personally work on compilers as part of research, so that's "focus". My current broad goal is to learn all the parts of math that I feel might wind up becoming useful for applied math in the next 50-60 years. So no arithmetic geometry, yes algebraic geometry.

Biology. Bio informatics is going to be important in the next decade.

This is a solid list.

Reply on news.ycombinator.com