Live data from Hacker News

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

news.ycombinator.com

371–380 of 839 posts

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

#371
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…

there are some animations that show how fast molecules and proteins go around in a cell, it's basically a bunch of extremely fast collisions and interactions going at random that end up falling into proper configurations. The way Science is taught in molecular biology (as in, visually, with proteins binding to receptors just like if it were fate) is usually completely wrong.

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

#372

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…

Short answer: there isn't an easy answer. Yet. (Give QC another 50 years).

Proof? Just look at all the replies you got: each one is dozens of pages of complex (imaginary) math, control theory, and statistics.

The hardest part of QC is exactly what you described: how to extract the answer. There is no algorithm, per se. You build the system to solve the problem.

This is why QC is not a general purpose strategy: a quantum computer won't run Ubuntu, but it will be one superfast prime factoring coprocessor, for example (or pathfinder, or root solver). You literally have to build an entire machine to solve just one problem, like factoring.

Look at Shor's algorithm: it has a classical algorithm and then a QC "coprocessor" part (think of that like an FPU looking up a transcendental from a ROM: it appears the FPU is computing sin(), but it is not, it is doing a lookup... just an analogy). The entire QC side is custom built just to do this one task:

https://en.wikipedia.org/wiki/Shor%27s_algorithm

In this example he factors 15 into 5x3, and the QC part requires FFTs and Tensor math. Oy!

Like I said, it will take decades for this to become easier to explain.

For fun, look at the gates we're dealing with, like "square root of not": https://en.wikipedia.org/wiki/Quantum_logic_gate

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

#373

If I buy a stock, does the price at which I agreed to buy it become the new share price on the stock exchange? Every article on "Where do stock prices come from?" seems to just talk at a high level about supply and demand. But where does the price come from at a nitty-gritty level? Is it an average of all existing offers or something? Do different exchanges and stock-ticker websites have different formula for calcula…

It’s not dissimilar to a commodity like Gold with some extra parameters like number of stocks. Comes down to an open market and buy/sell demand. There are other parts to your example such as depth of order book (orders on the buy sell side) which provide liquidity. So if a low volume stock is at $4 and you offer to buy at $100 you would probably end up with 100/4=25 shares. If was more illiquid you could possibly move the price up depending on order quantity.

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

#374

Earlier quoted context omitted.

The short answer is this. Suppose thread OP gave you and I a box each. Each box has a some knobs to choose some settings, and if one chooses a setting and presses the GO button, a number pops up on the screen. We go far away from each other so that, due to the finite speed of light, it takes quite a while for any information to travel between the boxes. We do this, because we don't want the boxes to communicate live,…

So what I don't understand is, if both of us take an identical Gemalto token / Yuvikey, we can be light years apart and get the same sequences, no? Is the explanation that these will have one type of distribution vs if they had real "spooky movement at a distance" they have a clear different distribution? EDIT: what about this one: https://arxiv.org/abs/quant-ph/0301059

The distance is part of the proof, but not really part of the mystery.

I'm going to throw out an analogy that gets at what's observed and why it's surprising, but doesn't relate to the physics of spin, momentum, position or anything that's actually under observation in these experiments.

It's as if we have a pair of dice, and I throw my die and you throw your die many times. In a classical world, if I throw a three, it has no influence on what you throw; you're equally likely to throw 1-6. But in the quantum world it's as if when I throw a one, your die still has the expected uniform distribution, but when I happen to throw a three, you're a little bit more likely to throw a three. Your die is fair if I happen to roll a one, but it's weighted if I happen to throw a three.

Back in the real world, this is the strange behavior that is observed in experiment. Schroedinger's equation predicts the probabilities perfectly. But Bell shows that it's far from intuitive.

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

#375

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 honestly recommend the following: 1. Pick up the standard textbook by Nielsen and Chuang, Quantum computation and quantum information. read the first two to three chapters. 2. Solve the exercises for the Q# programming language.

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

#376

* Lie algebras and Lie groups - I still don't understand this, what they're used for or how to use them in any practical sense. * Galois Theory - I have a basic understanding of abstract algebra but for some reason Galois theory confounds me, especially as it relates to the inability of radical solutions to fifth and higher degree polynomials * "State-of-the-art" Quantum Entanglement experiments and their purported s…

Probably the best place to get started with Lie groups is to wrap your head around SO(n), which has both a nice geometric interpretation as rotations of n-dimensional space as well as a concrete representation as the space of orthogonal (determinant 1) nxn matrices. With a little matrix calculus you can work out what the tangent directions are at the identity matrix: they’re precisely the skew-symmetric matrices. Thi…

I’m sorry but you used so many big words so close together, I could find no purchase with the things I know. Perhaps you could tell me what was the problem that Lie groups are supposed to solve. I read something about an infinitely symmetric transformation- the rotation of a circle is a Lie group. But then what are all the letters about- the A,B,C etc all the way up to E 8. And why did it take 77 hours on the Sage supercomputer to solve E8.

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

#377
post #83
post #56

I find most explanations of the Equivalence Principle that lies at the foundation of General Relativity to be very lax. To wit, the idea is that you cannot distinguish whether you are in an accelerated frame or in a gravitational field; alternatively stated, if you’re floating around in an elevator you don’t know whether you’re freefalling to your doom or in deep sideral space far from any gravitational source (thoug…

You're right that this is glossed over in popular explanations, but the point you make is exactly the starting point for all formal courses and textbooks. The first part of the argument is that for single point particles falling, the effect of gravity is the same for all particles. This suggests that we should model gravity as something intrinsic to spacetime itself, rather than as a field living on top of spacetime,…

> it should reduce in the Newtonian limit to something that looks like a tidal effect, i.e. a gradient of the Newtonian gravitational field.

Can you please explain to me how you went from"looks like a tidal effect in the Newtonian limit" to "a gradient of the Newtonian Graviational field"?

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

#378
post #243

Yoneda lemma

As with most things in category theory, the way to really understand the Yoneda lemma is to sit down and prove it yourself. Break the statement down into the underlying definitions, draw lots of diagrams, and convince yourself that it’s true.

The other thing you can do is think about what it means for particular types of categories. For a posetal category, it says that an element of a poset is uniquely determined by the set of all elements that come before it in the ordering. For a group, it says that every element is uniquely determined by its action on the group. (This is basically Cayley’s theorem.) See this MSE post for more intuition: https://math.stackexchange.com/questions/37165/can-someone-e...

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

#379

If I buy a stock, does the price at which I agreed to buy it become the new share price on the stock exchange? Every article on "Where do stock prices come from?" seems to just talk at a high level about supply and demand. But where does the price come from at a nitty-gritty level? Is it an average of all existing offers or something? Do different exchanges and stock-ticker websites have different formula for calcula…

The price is generally the last price at which a transaction happened, modulo people intending to cheat the system.

There are people who put limit orders on the exchanges. Say that the price of TSLA is $500. I think it's overpriced, and its likely to go down, but then grow in the future. I can say, "I'm willing to buy 100 shares of TSLA at $420." Someone else holds TSLA and thinks its likely to go up, but not hold it's value, so they say, "I'm willing to sell 100 shares of TSLA at $690." The sum of all of these limit orders forms the market depth chart.

The more common way to interact with the market is to say, "I want to buy a share of TSLA at the current market price." In the above example, the only option is to buy TSLA for $690, even though the last transaction was $500! This is a example with very little market depth. In the normal case, you'd buy your share for $500.02 or something like that. (Same, but reversed, for selling at market price.)

For more information, but with a crypto focus, see https://hackernoon.com/depth-chart-and-its-significance-in-t...

For your example, you would put in a market order, and buy the stock at the lowest price that someone was willing to sell it at. If the last price was $4, but the lowest limit order that currently existed was for $100, and you bought it for $100, then yes, the price would go up to $100. (In real life, those sharp upticks don't happen much. It's more likely that a sharp downtick happens, where suddenly everyone wants to sell oil futures at the same time, but almost no one is willing to buy them, so the price ends up negative.)

Note that whenever people defend high-frequency trading for "providing liquidity to the market," this action of setting buy and sell limit orders that are close to each other is what they are talking about. There are algorithms that will see TSLA at $500, and offer to sell TSLA at $500.02 and buy TSLA at 499.98. If both orders go through, they make $0.04. If you operate fast enough to get out ahead of any big market moves, you can make a lot of money. But if you ever accidentally buy a bunch of TSLA for $499.98 right before the price plummets to $420, then you just lost a lot of money. This is why HFT and other trades with similar risk profiles are sometimes referred to as "picking up nickels in front of a steamroller."

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

#380

If I buy a stock, does the price at which I agreed to buy it become the new share price on the stock exchange? Every article on "Where do stock prices come from?" seems to just talk at a high level about supply and demand. But where does the price come from at a nitty-gritty level? Is it an average of all existing offers or something? Do different exchanges and stock-ticker websites have different formula for calcula…

If you see a single share price listed, that is the price of the last sale that occurred.

Now that can either mean that someone bought a share that someone else was selling or that someone was selling a share to someone who was offering to buy.

The shares are listed as a series of buy and sell orders in what's called an order book.

If the price a share was sold at was 100$ and you think it will go a bit lower, you could place a buy order at 90$. Should enough people sell shares to reach your price and order, your order will be filled and you will own the share at 90$.

If someone wants a million shares at 91$, you may not get your single share at 90$.

To go back to your example, if you were to place a buy order at 100$ for a 4$ priced share, how much the price moves depends on how many sell orders are in place from 4$ to 100$ and how much you are buying.

If it's only one for share, your order will probably get filled at something like 4.01$ if the spread is low (the spread being the difference between the highest buy order and the lowest sell order).

If you're buying 1000 shares and it's a low volume stock with a "thin" order book, maybe it could go up a few dollars instead but for it to go up to 100$ you have to buy every single share between 4$ and 100$.

Post reply on HN