Live data from Hacker News

Ask HN: What scientific phenomenon do you wish someone would explain better?

news.ycombinator.com

611–620 of 839 posts

Re: Ask HN: What scientific phenomenon do you wish someone would explain better?

#611
post #63

Automatic differentiation. It's useful to so much computational work, but most people only get a cursory introduction to the topic (a rough intro to the minimum they need to know), whereas really understanding it seems to open up a lot of research.

Oh man, I read a super cool article about that about a year ago. It provided an algorithm for automatic differentiation using an imaginary number such that it times itself equaled 0, but wasn't equal to zero itself. I'll try to find the link.

I don't know if this was it, but an explanation nonetheless https://medium.com/@omaraflak/automatic-differentiation-4d26...

Re: Ask HN: What scientific phenomenon do you wish someone would explain better?

#612

Macroeconomics. Central banks are "creating" a trillion here, a trillion there, like nobody's business. But what are the consequences? What is the thought process that central bankers have gone through to make these decisions? Also why, exactly, are they buying the exact assets that they are buying (govt. debt, high-yield bonds, etc..) and why not others (e.g. stocks or put money into startups)? And then, what happen…

Not sure I'd class it as a science, but here's my take... though I'm no expert and certainly agree this stuff could be better explained!

> Macroeconomics. Central banks are "creating" a trillion here, a trillion there, like nobody's business. But what are the consequences?

Nobody quite knows; it's still hotly contested between left wing lovers of Keynes and right wing believers in austerity.

> What is the thought process that central bankers have gone through to make these decisions?

Probably largely a political one. Central banks may be trying to fulfill a remit set by law (e.g. bank of england: keep inflation below x%) and are trying to deliver on that. (why? too much or too little inflation both cause problems, I guess we somehow reached consensus on a "sane" amount that keeps pace with genuine growth of wealth within the economy)

> Also why, exactly, are they buying the exact assets that they are buying (govt. debt, high-yield bonds, etc..) and why not others (e.g. stocks or put money into startups)?

I think this is about distributed decision making. The central bank does not have the expertise to decide which stocks or startups represent the best investments. The examples here involve lending money to government, presumably the idea being the latter is better placed to decide what to do with the money. Another example is buying assets from other banks, which are again better placed to decide which businesses/homeowners/etc represent a more sound investment as they do it on a daily basis (from a profit/loss point of view ... of course we debate whether or not that's the case on a societal level).

> What would happen if a country defaults on a loan given by a central bank?

Internally it would depend on laws/balance of political power within the country. Between countries, depending on the currency the country could do crazy stuff like print excessive amounts of money to repay the loan (Germany did this in the 1930s leading to hyperinflation) or they could just as you say, default. The country's credit rating would then be downgraded, making it harder for them to raise credit in future.

> Would the central bank then go ahead and seize and liquidate assets of the country under a bankruptcy procedure to pay off the debt (like it would be standard procedure for individuals and companies)?

Not the bank, but the country making the loan, may first negotiate some debt relief with strings attached e.g. preferential trade agreements. Beyond that, I have no idea what precedent exists.

Re: Ask HN: What scientific phenomenon do you wish someone would explain better?

#613

Avogadro constant - I can accept the number, if someone can show me how, over a hundred years ago, without having every really seen a molecule, this number could be derived. Obviously no one ever sat there and counted them but I find it hard to believe, they could have decided this without being able to see it.

Here's how to estimate the size of an oil molecule just using water and an oil drop: https://spark.iop.org/estimating-size-molecule-using-oil-fil... (They use some other stuff, but you get the idea) You can back out Avogadro constant starting with this experiment.

Thank you. I do get the idea. And now I can go back and finish High School chemistry. I appreciate you help in understanding this.

Re: Ask HN: What scientific phenomenon do you wish someone would explain better?

#614

Quantum Computers. Not like I'm five, but like I'm a software engineer who has a pretty decent understanding of how a classical turing machine works. I can't tell you how many times I've heard someone say "qubits are like bits except they don't have to be just 1 or 0" without providing any coherent explanation of how that's useful. I've also heard that they can try every possible solution to a problem. What I don't u…

I had an aha moment with quantum computers a few months ago when reading an article that explained it as probability distributions. I don't think I have the complete understanding in my mind anymore and I wish I had saved the article, but looking into how quantum computers essentially serve as probability distribution crunching machines might help with your understanding.

Like probability distributions, but they don't just sum when you combine them, they interfere (probability is the square of amplitude, which can be negative).

Quantum computing is all about finding ways to hack the interference process to compute more than you otherwise would have.

Re: Ask HN: What scientific phenomenon do you wish someone would explain better?

#615

I wish I could get a clear explanation of why faster than light travel breaks causality. I have seen the math but intuitively I am having trouble grasping it especially when you have only two reference frames (most explanations will use three reference frames to show the violation)

The simplest answer is that events that happen further apart in space than in time can be observed in different orders depending on the frame of reference. If an effect can happen before its cause, causality is violated.

The three-reference frame example is the easiest, because you can start with a frame where two events, A and B, happen simulateously. A reference frame (say, a spaceship), flying along a line in the A to B direction will observe A happen, then B. A ship flying the opposite direction will experience the opposite, B then A.

So whose observations were correct? All them are perfectly valid. The problem is if we allow A to cause B, in which case the B then A frame has the effect happen before the cause.

Re: Ask HN: What scientific phenomenon do you wish someone would explain better?

#617
post #119

I would like to understand how cellular biology processes actually work. Like, how do all the right modules and proteins line up in the right orientation every time? Every time I watch animations, it seems like the proteins and such just magically appear when needed and disappear when not needed [0]. Sometimes it's an ultra-complex looking protein and it just magically flys over to the DNA, attaches to the correct sp…

They don't. This is a pet-peeve of mine, and it's reinforced by animation after animation. Everything is being jostled around randomly. The molecules don't have brains or seeker warheads. They can't "decide" to home in on a target. The only mechanisms for guidance are: diffusion due to concentration gradients, movement of charged molecules due to electric fields, and molecules actually grabbing other molecules. It's…

I once saw a video that purported to showed the jittering for some simple chemical reaction, it was indeed very enlightening.

Re: Ask HN: What scientific phenomenon do you wish someone would explain better?

#618

Earlier quoted context omitted.

I recently started taking insulin. Check out the molecular structure for that. It blows me away how complex it is.

By molecular structure alone Insulin is one of the simplest proteins, even (though it's complex in ways you don't see by looking at a static picture of it, lifecycle, oligomeric interactions)

Compared to something that isn't a protein, it's pretty complex.

It's like how the source code to `ls` is simple because it's one of the most basic Unix programs, or something like that.

Re: Ask HN: What scientific phenomenon do you wish someone would explain better?

#619

Earlier quoted context omitted.

They don't. This is a pet-peeve of mine, and it's reinforced by animation after animation. Everything is being jostled around randomly. The molecules don't have brains or seeker warheads. They can't "decide" to home in on a target. The only mechanisms for guidance are: diffusion due to concentration gradients, movement of charged molecules due to electric fields, and molecules actually grabbing other molecules. It's…

It would be great if they showed just one animation up front of the chaotic mess that actually represents reality. They could then show the simplified version so that we can actually see what is going on.

https://www.researchgate.net/profile/Nicolas_Bellora/publica... is one example of the chaotic mess. What that shows is many RNA polymerase molecules walking up a gene. The horizontal line across the middle is DNA. The vertical tails hanging off it are RNA being built as the DNA is transcribed.

What that image drove home for me is:

1) that DNA transcription isn't something that happens rarely, or once-at-a-time. DNA is constantly being transcribed; proteins are constantly being built. The scale and rate isn't something I'd ever been taught.

2) How RNA polymerase works must be taking into account a hell of a lot of congestion. Polymerase molecules must constantly be bumping into each other.

3) How the picture would make no sense whatsoever unless you already know what the mechanism is.

I think it does make sense to start with the idealised process, as long as you follow up with messy reality.

Re: Ask HN: What scientific phenomenon do you wish someone would explain better?

#620
post #293

Crypto and practical security. I get tired of the circular “don’t roll your own crypto unless you’re qualified”. How does one become qualified? I don’t feel like I know how to evaluate many of the arguments people make for or against technologies people argue about on HN, such as Signal or different password managers. I feel like “security through obscurity” is a bad thing, and “layers of security” are a good thing,…

> I get tired of the circular “don’t roll your own crypto unless you’re qualified”.

It's true, but you need to realize that you're qualified enough only when you understand that you shouldn't roll out your own crypto.

In my opinion, the only person who has credibly demonstrated being able to roll his own crypto is djb (http://cr.yp.to/)

> but isn’t all security obscuring something,

Keeping a secret isn't "obscuring" something, it's hiding it entirely. Security through obscurity is bad because it relies on attackers being dumb. The smartest person in the world cannot be expected to guess a well chosen and kept secret.

Post reply on HN