Live data from Hacker News

Ask HN: Do you recall any book or course that made a topic finally click?

news.ycombinator.com

121–130 of 515 posts

Re: Ask HN: Do you recall any book or course that made a topic finally click?

#121

I didn’t understand probability, until I read E T Jaynes Probability Theory: Logic of Science It defines the base blocks of probability very, very slow. And never hand-waves anything. But it’s the “bayesian” view of probability; but it’s honestly the easier one to understand. https://www.amazon.com/Probability-Theory-Science-T-Jaynes/d...

What is the prerequisite to read this book?

Re: Ask HN: Do you recall any book or course that made a topic finally click?

#123
post #64

It's strange how Grant Sanderson's (aka 3Blue1Brown [1]) name hasn't come up so far. [1] https://www.youtube.com/c/3blue1brown

While I enjoy 3blue1brown's videos they definitely fall under the umbrella of current gen "edutainment" where the goal is to make you feel like you've learned something without have anything really click. Pretty much everyone I know who claims that 3blue1brown's video helped them understand a topic... doesn't really understand that topic at all. 3blue1brown's video can be fun for topics you already know, providing a…

I can't speak generally about all of 3blue1brown's content, but the essence of linear algebra made concepts like eigenvectors and values and what they are click for me, there is no question.

Re: Ask HN: Do you recall any book or course that made a topic finally click?

#124

I had a very poor model of what people meant when they spoke of divinity, god, and religion until reading the Tao Te Ching. Whenever I heard god, I just thought of sky daddy. The key is to understand that religion is a map, not the territory. All religions are essentially different paths to the same destination, mixed with varying levels of human interpretation.

> The key is to understand that religion is a map, not the territory. All religions are essentially different paths to the same destination, mixed with varying levels of human interpretation.

Given that a large fraction of people who self-identify as religious would vigorously disagree with this, I think this insight's of limited—though perhaps not no—utility in understanding religion.

Re: Ask HN: Do you recall any book or course that made a topic finally click?

#125
When I studied C in college everyone recommended the K&R C book.

I found it the be of no value to me for learning the language. Instead I found King's C Programming: A Modern Approach much more enlightening.

I know K&R is widely well-regarded, so I found it interesting how difficult it was for me to learn from it.

Re: Ask HN: Do you recall any book or course that made a topic finally click?

#126

The E-Myth Revisited: Why Most Small Businesses Don't Work and What to Do About It (Michael E. Gerber) What it taught me is that effective organizations invest heavily in training newcomers how to do their jobs / processes. I see this everywhere; when I walk into a business and something's "just off," it's almost always because management is dropping the ball on training newcomers how to do their jobs. In my specific…

You can't train people if you don't standardize first! This could be as simple as deciding what elements a big report should contain, or as complex as a multi-page standard operating procedure (SOP) for a knarly process.

Re: Ask HN: Do you recall any book or course that made a topic finally click?

#127
post #103

Bartosz Ciechanowski: https://ciechanow.ski/ I especially loved the one on mechanical watches: https://ciechanow.ski/mechanical-watch/

The mechanical watch article / walkthrough is one of my favorite finds on HN. Really great design and explanations, both visually and conceptually. I’ve shown it to several people because it’s that cool

Re: Ask HN: Do you recall any book or course that made a topic finally click?

#129
Statistics.

It made data, functions, web applications, and REST API finally click for me.

I took two stats courses-- a simple one at a community college where we dabbled with a stats analysis app. And a more complex one, where we wrote scripts in the language "R" and uses rStudio.

I began to realize things I should have realized long before that:

- Functions typically intake data, and output data. What made it click was inputs into a statistical model, such as multivariate linear regression.

- A program runs on data-- data is basically the "currency" (i.e. monetary currency) of an application.

- Data comes in various shapes and sizes, and collections. Programs have to be compatible with those shapes, sizes, and depths within collections.

Once I realized this, I began reflecting more on the typical 3-tier web application model, and HTTP, and REST API, including looking at diagrams on google images. It became so much clearer once I had a solid understanding of data & functions.

Post reply on HN