It’s time to admit quantum theory has reached a dead end • Hacker News Board

Re: It’s time to admit quantum theory has reached a dead end

#91

Post by XorNot » Wed, Mar 09, 2022, 1:14 AM UTC

8bitsrule wrote: post #28

Even on the macro level there are lots of tricky questions. A bar magnet is surrounded by magnetic flux lines. Apparently it takes no energy to maintain that magnetic field. If I heat the bar in boiling water for long enough, it becomes demagnetized. On the macro level there's no visible/measureable change in the bar. We need a lot of magnification to see that the 'particles are no longer aligned'. On the other hand,…

Nobody ever asks this question about a spring supporting a mass against gravity.

There's nothing actually holding atoms together, it's all electric fields. So why is a magnet different? (it's not).

People also don't ever ask this question about gravity itself.

Re: It’s time to admit quantum theory has reached a dead end

#92

Post by mathematicaster » Wed, Mar 09, 2022, 1:18 AM UTC

user3939382 wrote: post #78

fennecs wrote: post #64

Amusing to see all the pessimistic comments form programmers who have zero understanding beyond high school maths thinking somehow there is a magic solution to all the problems in physics like you can just refactor everything. Quantum field theory is definitely not stuck in the 1920s. Modern physics has a lot more going on than 19th century mathematics. We understand quite a lot about the mathematical structure under…

> Quantisation is a nearly well understood process The layman’s explanation I received for this, perhaps incorrectly, is that you fill in variables for factors that you don’t understand or can’t measure with whatever value makes the rest of the theory / equation work. Is that at all correct?

All correct.

Oh ... and if things are still a bit floppy, just fix the gauge and you are good.

Re: It’s time to admit quantum theory has reached a dead end

#93

Post by nwallin » Wed, Mar 09, 2022, 1:21 AM UTC

geijoenr wrote: post #5

For some time already we are in an "epicyclean phase" of physics, trapped by the extraordinary predictive success of Quantum Electrodynamics and still using mathematical methods devised in the 19th century (variational calculus). This has lead us to the current situation, with extremely complicated theories at the limit of human understanding that bear no new results. It will take a modern day Copernicus to come up w…

Copernicus' model had the planets orbiting the same in circles at uniform speed. This does not match reality, so Copernicus just chucked in some epicycles to correct the error. Copernicus's model had all the complexity of epicycles, but he hard coded the first epicycle into the system by having the planets orbit the Sun instead of the Earth.

We wouldn't need a Copernicus to solve this problem, we would need a Copernicus, then a Kepler, (ellipses & non-uniform speed) then a Newton, (gravity causes ellipses and non-uniform speed) and then an Einstein. (gravity is warping of space-time)

If quantum field theory is as wrong as Ptolemy's geocentric model was, we're hopeless. Because QFT very well predicts the observations; our observations have no ellipses in them that invalidate circular orbits, our observations have no anomalous Mercury precession that invalidates Newtonian gravity, no speed of light being consistent in all directions to invalidate luminiferous aether. To say that we simply need a smarter theoretical physicist is simply wrong -- our current theories do not contradict the things we are able to observe.

We know that general relativity and quantum mechanics do not play nice at small scales and high local gravity. But we cannot observe this conflict. And that's nothing to go on.

We would need an observation which shows that general relativity or QFT is wrong about something before we could conceivably make foundational progress on making new or different theories. And every few months there's a new article about "Einstein is proven right again" or "LHC experiment shows all readings are nominal".

Re: It’s time to admit quantum theory has reached a dead end

#94

Post by kayamon » Wed, Mar 09, 2022, 1:23 AM UTC

chii wrote: post #74

jiggawatts wrote: post #4

Earlier quoted context omitted.

To me, this always felt like programmers excusing abhorrent code with "but it works". Quantum theory isn't. It's not a single, cohesive, consistent theory.[1] There is no recipe you can apply, it's just a bunch of guessworks and heuristics that happen to produce the numerically correct equations if you keep trying long enough. This isn't secret or some sort of external criticism, you'll find this front-and-centre in…

> because approaches like this often have virtually no explanatory power. the famous feynman have produced a response to this train of thought - usually from the laymen - where they demand an "explanatory theory". The answer is usually "don't know"; why gravity work the way it does? - don't know, but we can calculate orbits to great precision; why does photons or electrons travel the way it does? - don't know, but we…

Feynman must have driven his wife mad.

  "Honey did you take the garbage out?"
  "When you say garbage, have you considered if garbage can even really exist?"

Re: It’s time to admit quantum theory has reached a dead end

#95

Post by Iwan-Zotow » Wed, Mar 09, 2022, 1:24 AM UTC

bricemo wrote: post #58

lacker wrote: post #47

Earlier quoted context omitted.

It will take a modern day Copernicus We might already have a modern day Copernicus. Could be string theory, could be something else. The problem is that we don't really have any experimental data that can't be explained by the Standard Model. What we really need is a modern day Galileo that can perform some sort of observations, like finding Jupiter's moons, that don't fit in with the existing conventional physics.

27% of the universe is totally unknown (dark matter) and another 68% is totally unknown (dark energy), leaving only 5% of our universe that is explained by the standard model. Feels very epicycle-ish and ripe for a major shift.

they are both called dark for a reason

not very observable

Re: It’s time to admit quantum theory has reached a dead end

#96

Post by wahern » Wed, Mar 09, 2022, 1:25 AM UTC

dr_dshiv wrote: post #84

JumpCrisscross wrote: post #65

Earlier quoted context omitted.

> we don't really have any experimental data that can't be explained by the Standard Model What was the evidence Copernicus saw that epicycles didn't explain? (Honest question.)

Nothing. His model actually fit the data worse than epicycles. He just thought it was a more sensible approach that matched the Pythagorean opinion. It wasn’t until Kepler that a geocentric model involving ellipses actually was more predictive. Might be worth returning to the Pythagorean interpretation (which Bohr discussed): All is number. The world is made of math, not stuff.

> It wasn’t until Kepler that a geocentric model involving ellipses actually was more predictive.

You mean heliocentric?

Re: It’s time to admit quantum theory has reached a dead end

#97

Post by dfdz » Wed, Mar 09, 2022, 1:25 AM UTC

The quotes from scientists at the end of the article nicely summarize my opinion on this article:

> Deutsch, too, seemed impatient with my dissatisfaction over our understanding of the nature of reality. “Someone might equally well say: We may know what dogs look like and how they behave, but we don’t know what a dog ‘actually is’,”

Re: It’s time to admit quantum theory has reached a dead end

#98

Post by fennecfoxen » Wed, Mar 09, 2022, 1:29 AM UTC

sidlls wrote: post #87

user3939382 wrote: post #78

Earlier quoted context omitted.

> Quantisation is a nearly well understood process The layman’s explanation I received for this, perhaps incorrectly, is that you fill in variables for factors that you don’t understand or can’t measure with whatever value makes the rest of the theory / equation work. Is that at all correct?

I hope that's a misunderstanding or miscommunication. It might be useful in some limited contexts, but it would point to gaps in the theory (at minimum).

There are 19 free parameters in the Standard Model. These include such values as the mass of various particles (electron, muon, tau, six quarks, Higgs), Higgs vacuum expectation value, the strength of several interactions, and the mixing angles and phases for certain interactions.

You can rearrange some of the parameters so that they're not free, and redefine each of these in terms of some other related parameter — your choices here are pretty arbitrary; these values being chosen as the 'free' ones does not represent some fundamental truth.

We do not have some underling deeper theory that indicates why these numbers are the way they are. Why is the electron mass ~511 kEv while the up quark is 2.2 mEv? Why not 1.9 mEv? Why not 2.3 mEv? The Standard Model doesn't generally explain it, nor does it even really intend to explain it, just to describe it.

Re: It’s time to admit quantum theory has reached a dead end

#99

Post by hither_shores » Wed, Mar 09, 2022, 1:33 AM UTC

user3939382 wrote: post #78

fennecs wrote: post #64

Amusing to see all the pessimistic comments form programmers who have zero understanding beyond high school maths thinking somehow there is a magic solution to all the problems in physics like you can just refactor everything. Quantum field theory is definitely not stuck in the 1920s. Modern physics has a lot more going on than 19th century mathematics. We understand quite a lot about the mathematical structure under…

> Quantisation is a nearly well understood process The layman’s explanation I received for this, perhaps incorrectly, is that you fill in variables for factors that you don’t understand or can’t measure with whatever value makes the rest of the theory / equation work. Is that at all correct?

No, that sounds more like a (very distorted) description of regularization and/or renormalization.

Regularization techniques take effective field theories like the Standard Model, which are known to be inaccurate at high enough energies, and isolate their low-energy behavior.

For example: classical electrodynamics, interpreted naively, says that the total energy stored in an electron's electric field is infinite, since the field strength blows up to infinity at the position of the electron itself. As this isn't actually true, we know there must be physics going on very close to the electron that our theory doesn't account for. But even if our theory isn't the Final Truth, it's still capable of making perfectly good predictions far away from the electron, and we can't just sit around waiting for quantum mechanics to be discovered.

So we make the structure of the electron a parameter of our theory: it can't tell you what the force between two electrons is, but if you posit that the electron's charge is distributed like so, it can tell you what the force between two electrons would be.

We then use empirical data to reabsorb our new degree of freedom: this is renormalization. Instead of trying to predict the force between two electrons, we measure it, and then work backwards to figure out what the electron's charge distribution would have to look like to produce that force. In this particular case: the electron behaves as if it were a sphere about 10^-15 meters across. This is of course not the actual structure of the electron, but it reproduces the same low-energy classical physics.

Re: It’s time to admit quantum theory has reached a dead end

#100

Post by jml7c5 » Wed, Mar 09, 2022, 1:35 AM UTC

You are led into a room. The far wall is covered in levers, buttons, dials, flashing lights, oscilloscope displays, buzzers, and so on. The person who brought you there says "we need you to describe the mechanism behind this wall. You're not allowed to look behind the wall. Good luck!" They leave the room.

You walk up to the wall and press a button. A few feet away, a light flashes green. You press the button again. The same light flashes green. "Aha!" you think. "I have figured out what this button does: it causes that light to flash green." On a table to the side of the room is a notepad and a pencil. You jot down your discovery.

You walk back to the wall, and press a different button. Once again, the same light flashes green. You begin to wonder if all the buttons affect that one light. You press the original button to confirm your suspicions. The light flashes blue.

---

Months have passed. You have covered the three previously unadorned walls with notes, with diagrams, with layer upon layer of discarded theory. But you're so close! The machine you've designed perfectly matches everything you've observed. It's so vivid in your imagination that you can practically see it through the wall. There's just a single lever you're unsure about; you can't decide if it's part of a complex mechanical linkage, or if it simply closes an electrical circuit. It would all work either way. You draw both options and place a question mark between them.

You wander down the hall to an office, pop your head in and say to the person seated at the desk: "I think I've figured it out. Could you come take a look?"

---

THEM: "So you're not sure which one of these it is?"

YOU: "It could work either way, so I drew both. I couldn't find any way to determine the answer."

THEM: "Well, that's not very satisfying. Are you sure you tried everything? What about this button, how many times did you press it?"

YOU: "That one? Tens of thousands of times, cumulatively."

THEM: "Have you tried pressing it repeatedly? Over and over again, I mean, without anything in between."

YOU: "I have the notes over here... yes, one hundred times in a row to confirm my theory was correct."

THEM: "Only a hundred? What if it does something different after a thousand presses? You should try that; it might solve your dilemma."

YOU: "I... could, but that seems awfully time-consuming. None of the other buttons did anything different after more than a dozen presses."

THEM: "Yes, but couldn't this button be different?"

YOU: "Maybe, but there's a point where I have to assume I've got it right. After all, there's no way of knowing whether it does something different at ten thousand presses, or a million presses!"

THEM: "All the same, we would really like to know what's behind that wall. It's a bit disappointing to only know what will happen but not know what's really going on back there. Could you try it the thousand presses? Perhaps twiddle some other widgets at random, see if that creates any discrepancy?"

YOU: "But that might never end! If nothing comes up after a day of effort, you could ask me to do it for a week. If nothing comes up after a week, you could ask me to do it for a year. Can't you just accept that either option works?"

---

The moral of the story: there's a limit to knowledge gained by inference. The problem is even worse than the story described, because even if you resolve the dilemma of the lever:

-It's impossible to know that the 200 sextillionth press (and only the 200 sextillionth press) of that button does something slightly different. Your design is wrong, and you'll never discover the discrepancy.

-It's impossible to know that the first time you pressed the button (and only the first time), it lit up a yellow light in the corner that you didn't see. Your design is wrong, and the information needed to resolve it is lost to you forever.

In the real world, maybe we'll get lucky and produce a theory that has only a single interpretation. Maybe there is a discrepancy that rules out all but one solution. (And maybe we'll get extra lucky and it will be easy to visualize.) But even then, we can never be sure. I feel as though the author of the article isn't clear enough about this. You can never know what's behind the wall.