Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Coolest Things I Learned in 2018

perell.com

Post reply on HN
Print view View original

141–148 of 148 posts

Re: Coolest Things I Learned in 2018

#141

Post by jacobolus »

wenc wrote: post #99
ergothus wrote: post #48

Earlier quoted context omitted.

I must have expressed myself poorly: When I try to pick up a new skill, I would start by asking "what are the higher-level rules?" or some variation. Generally speaking, I've had a very, VERY low success rate at getting answers. When I did get answers, they always relied on a context I didn't have, because I was brand new. Some areas I pick up intuitively, others take work, and others I completely suck at, but in non…

Yes. It's like learning a language by first learning the grammar -- that works if you already have semantic tree of grammar that you can patch in from a related language family (say Portuguese to Spanish), but for most people this doesn't work. Most people need to start by memorizing phrases by rote first. It's only after we are familiar with the language that the underlying structures emerge. This applies to learnin…

> Most people need to start by memorizing phrases by rote first.

The most effective for most people is to hear a large amount of the language being spoken, bootstrap by memorizing a few words, and then reliably get some amount of practice speaking and a lot of additional exposure listening.

Similarly, the best way to learn to write is to do occasional bits of writing with feedback and tons and tons of reading.

Re: Coolest Things I Learned in 2018

#142

Post by mikhailfranco »

mcv wrote: post #39
twodave wrote: post #3

> Instead of arbitrarily memorizing things, look for the explanation that makes it obvious. This is a higher learning method that I’ve always found most natural. I don’t memorize stuff, I try to understand it instead. That said, there is still great value in the classical learning model—if you don’t know the terminology of the subject it’s difficult to reason about that subject.

I've had the same thing. It was impossible for me to remember complex mathematical equations in school until I figured out how to derive them myself. That didn't help me with integration, sadly. Same with history: years are meaningless. What's interesting is how historical events follow from earlier events and lead to later events.

I strongly identify with the parent comment: I have poor rote learning of isolated facts, but strong spatial abilities (sporting and visualization), relational memory and explanatory insight through the implied graph of connections.

When I was young, I played with construction toys (Lego, Meccano) and board games, including Snakes and Ladders (Chutes & Ladders in the US). Later, when learning history, I found I was not remembering dates by their year number, but visualizing their position in the decade and century on the zig-zag of the S&L board. I implicitly grew a spatial layout of multiple centuries, often with certain viewpoints that made sense for topics of study. I excelled in history because of my relational understanding laid over this 3D spatial view of the world timeline.

Perhaps you can guess my field of academic specialization and the sweet-spot in my career as a programmer? ... physics degree, then 3D graphics and information visualization.

Re: Coolest Things I Learned in 2018

#143

Post by mikhailfranco »

n4r9 wrote: post #13

> In a Big Craze, Be the Arms Dealer This is clearly true but I wonder how much it's an artefact of the Western economic model. I came to a similar insight in the last couple of years, which was "the companies skimming the most cream off the top, are often the most abstracted away from providing a concrete benefit to society". It's like there's this inherent recursive aspect to shareholder capitalism. The profitable…

It is basic economics 101: in the long run, Perfect Competition tends to yield zero profit. It is nothing to do with capitalism or meta-services, it applies to all economic activity. As Peter Thiel controversially but truthfully claims: competition is for losers.

In any value chain, you should seek the role that has the most control with the least competition. Often this is based on constrained resources: the supply of shovels in San Francisco is much more constrained and easy to monopolize than the area of land or length of rivers in California that might have some gold.

Creating or preserving the monopoly will also benefit from barriers to entry, such as conspiracy (oligopoly), regulatory protection (licensing, tariffs) or just plain old intimidation (mafia, law suits).

Another result is that when one constraint or barrier is removed, profits fall in that process, and shift to the next most constricted step of the value chain, much like the maximum pressure and minimum flow in a network of pipes with various diameters and valves. In general, this means that as artificial constraints are reduced (free trade, cartel busting...), the control of some underlying physical resource becomes most profitable, often property (assuming planning laws are never relaxed :), or certain scarce commodities.

For example, in tax-free and low tax economies (HK, SG, GCC) the profit often shifts to property (commercial and residential), so that what a worker gains in net income, they pay in extortionate rents, based on restricted land (HK, SG) or oligopolistic control of land and property (GCC). Low taxes are often a way of government forgoing revenue to the benefit of local landlords or property developers. Startups in Silicon Valley are ways of exploiting geeks to work hard chanelling money from VCs to local landlords.

Re: Coolest Things I Learned in 2018

#144

Post by screye »

mikhailfranco wrote: post #142
mcv wrote: post #39

Earlier quoted context omitted.

I've had the same thing. It was impossible for me to remember complex mathematical equations in school until I figured out how to derive them myself. That didn't help me with integration, sadly. Same with history: years are meaningless. What's interesting is how historical events follow from earlier events and lead to later events.

I strongly identify with the parent comment: I have poor rote learning of isolated facts, but strong spatial abilities (sporting and visualization), relational memory and explanatory insight through the implied graph of connections. When I was young, I played with construction toys (Lego, Meccano) and board games, including Snakes and Ladders ( Chutes & Ladders in the US). Later, when learning history, I found I was…

Damn, you described by way of thinking to near perfection. Spatial memory and relational graphs.

I too spent a huge amount of my childhood with Lego and board games.

I never became great with history, but found my way to computer vision and ML through a background in mech engg. (basically applied physics)

Still in school, so don't have an end to my story, but it is great to know others whose thought process coincides so well with mine.

Re: Coolest Things I Learned in 2018

#145

Post by mikhailfranco »

mikhailfranco wrote: post #142
mcv wrote: post #39

Earlier quoted context omitted.

I've had the same thing. It was impossible for me to remember complex mathematical equations in school until I figured out how to derive them myself. That didn't help me with integration, sadly. Same with history: years are meaningless. What's interesting is how historical events follow from earlier events and lead to later events.

I strongly identify with the parent comment: I have poor rote learning of isolated facts, but strong spatial abilities (sporting and visualization), relational memory and explanatory insight through the implied graph of connections. When I was young, I played with construction toys (Lego, Meccano) and board games, including Snakes and Ladders ( Chutes & Ladders in the US). Later, when learning history, I found I was…

By pure coincidence, just noticed this thread today:

https://news.ycombinator.com/item?id=18854838

Re: Coolest Things I Learned in 2018

#146

Post by beat »

nickles wrote: post #134
beat wrote: post #124

Earlier quoted context omitted.

As long as we're playing whataboutism... the Great Leap Forward occurred roughly 10-20 years after WWII, which led to a similar number of deaths in Europe. Without looking up data, I think it's safe to assume Europe's economies contracted substantially during WWII as well. So there's plenty of "clear failure" to go around. At best, one might argue that the US recovered from WWII faster, because it was the only major…

> As long as we're playing whataboutism... I'd prefer not to. > Without looking up data, I think it's safe to assume Europe's economies contracted substantially during WWII as well... At best, one might argue that the US recovered from WWII faster, because it was the only major nation left undamaged. I'd argue you can't compare Great Leap Forward with WWII. The GLF was domestic policy (like Holodomer) and ultimately…

The Great Leap Forward and WWII are absolutely comparable. China had been fighting colonial domination (first European, then Japanese) for almost 50 years by the time WWII (and Japanese occupation) ended. This was followed by several years of civil war that finally ended in 1949 with the exile of the nationalists to Taiwan. They had a population of illiterate, impoverished peasants, a demolished urban infrastructure, and a whole lot of local corruption to root out in order to consolidate power. A large-scale famine was inevitable. This was exacerbated by a total lack of international trade, due to anti-communist action in the West and an untrustworthy Soviet Union to the north. It was hard times.

As for Argentina... as you noted, Argentina was doing better than some European nations until the early 1960s. What happened? Well, back to the geography problem. Argentina is a remote nation, with limited ports. Spain, Italy, and Austria were part of Europe, possibly the most geographically gifted place on Earth (except for maybe the continental United States). And the US was dumping resources into Europe via the Marshall Plan, to rebuild the infrastructure and stabilize the governments. It's not surprising that they would eventually pass Argentina.

That said, the current state of Argentina is not bad. It's not rich, but it's on par with its peers in Chile and Brazil, and the debt from bad government in the 1970s/1980s is mostly paid off.

As for the cited paper... I really don't give a shit if it's been cited 450 times. It's demonstrably wrong. Developing nations with immature physical and legal infrastructures grow measurably faster than the wealthiest nations, with mature infrastructure. If you treat growth as an S-curve, this is totally an expected result. Double-digit growth rates in the developing world are not unusual. When's the last time a western nation had that kind of economic growth?

Re: Coolest Things I Learned in 2018

#147

Post by dwaltrip »

hobofan wrote: post #129
dwaltrip wrote: post #111

Earlier quoted context omitted.

Wasn't the Amazon factory thing a bit overblown? Or are the abuses actually on-going and not rare in number?

The extreme happenings are probably not that extreme anymore from what I've heard. Working conditions in the warehouses are still bad though. They are bad in a lot of shipping warehouses, but Amazon has been among the ones leading the race to the bottom. Apart from that, being an employee at Amazon even outside the warehouse is generally not that great from all the accounts I've hear over the years. A big part of tha…

Fair enough, I see where you are coming from. I am curious how widespread the poor working environments are at Amazon outside the workhouse, as it is difficult to extrapolate from a few examples. Thanks for the response.

Re: Coolest Things I Learned in 2018

#148

Post by DoctorOetker »

the "hold a mirror up to nature" quote attributed to Tom Hanks is in fact a classic Shakespeare quote (I don't know if he was the first however, perhaps he in turn got it elsewhere)

You may also wish to read this page on mirrors and Shakespeare (hard to imagine for us, and now mirrors or reflective decorations look kitsch, but during the introduction of quality mirrors they were high tech, so back then this was almost science-fiction play)

https://extra.shu.ac.uk/emls/08-1/kellglas.htm

Post reply on HN