Hacker News Reader

Comment by solinent | Hacker News Reader

I think quantum mechanics is well-understood, the mathematics behind it provide us with excellent predictions.

I think because at its core it is a probabilistic (or perhaps there are many worlds, but the implications are identical) theory--there is some inherent uncertainty in the predictions. This uncertainty builds up over time if left unmeasured, due to the chaotic nature of the system. We won't be able to result this uncertainty until we have more precise and accurate measurement devices.

I was thinking, the only real way for us to progress was to get the results out of CERN we were expecting--either smaller particles or some detection method which allows us to peer even deeper into the atom.

I think certain physicists are tired of the search for a unifying theory since it probably won't help us make better predictions. We need to get out there and get new empircal data.

View on HN

I did my M.Sc. in quantum physics, and I'll be the first to admit I don't understand it.

Yes, the equations do seem to predict what we observe in experiments very well. Shockingly well, given how strange some of the predictions made by them are! However, we lack an intuitive understanding of why the equations work. Matter and energy behave very differently on the quantum scale than they do on the macroscopic scales we're used to observing. If you're trying to model a pendulum, your intuition about how objects behave in the real world can be of immense help when performing sanity checks on your equations. Your intuition is worse than useless when it comes to quantum systems. It's not just flat-out wrong much of the time, it can lead you to ignore things that should not be ignored.

Quantum physics has all sorts of English language, short-hand descriptions for mathematical functions like "tunnelling", "non-local", "spooky action at a distance" or, as is pertinent to this article "observation". You may think you understand what these mean, but you're probably at least partly wrong if you're thinking about them in terms of human language, natural human intuition, or even the very specific short-hand terminology of physicists. We are continually teasing out nuances in the math that language simply does not capture. If we lack the language to adequately describe reality, can we really understand it?

Feynman was very correct to observe that people who think they understand quantum physics on a deep, intuitive level, probably do not. Doing quantum physics with a human brain evolved for bashing stones together is like trying to breathe in outer space. It's possible with considerable effort and the right tools (e.g. a space-suit), but if you ever forget that it's a deeply unnatural thing to be doing you're probably going to get yourself into trouble.

I ask you because I definitely do not have an M. Sc. In quantum mechanics - at the lowest level of any system isn’t it inevitable that there is no deeper understanding. At a certain point you have hit the base level and things just “are.” Correct me if this is erroneous!

This is a bigger question than physics can answer. The project of hard science is essentially reductionist and based on predicting events based on hypothesis. Your question is more logical or philosophical.

Logic may have something to say about open or closed systems, but mapping that onto the real world will always be a following question.

It seems to me the limited hypothesis "We have reached a point where no further progress can be made in understanding why quantum mechanics works the way it does" is falsifiable and thus not purely logical or philosophical. (i.e. quantum mechanics is the lowest level of physics).

It is only the broad hypothesis that is only answerable in philosophy: "There is a point in physics where no further progress can be made in understanding why things work the way they do and we will be purely limited to describing how things work." (i.e. there is some lowest level of physics)

Your 'limited hypothesis' can never be proved, because you can't prove things you haven't observed i.e. the proof works only up until counterexample, which by definition of unknowns, counterexamples (or a deeper understanding) would always be a possibility.

Falsification is not end end-all-be-all, it's important, but it's not the only thing that matters.

What scientific law or rule can be proved?

what is the point shkkmo is making?

That the limited hypothesis is a scientific hypothesis because it is falsifiable and testable while the broad hypothesis is a philosophic question because it is not falsifiable.

This is relevant as it is sort of ambiguous as to which hypothesis jshaqaw was asking about.

based on the specific wording and vocabulary jshaqaw used, i assumed he was referring to the latter

Yes, and I assumed that was your assumption. :D

ok good night :)

I think it was actually the former I was asking about but again I recognize I am outside my zone of competence in this area.

Consider the equation: 1+1=2

You don't have to understand the mathematical explanation for why this is so (one does exist) in order to have an intuition that serves you well in practically all situations. For most of us, there is no deeper understanding. It just is, and that's good enough. That doesn't necessarily mean there is no deeper understanding, but let's ignore that for now.

What I was trying to get across is that basic intuitions we have about the macroscopic world that we can see and touch are often flat-out wrong on the quantum scale. Likewise, human language has developed to describe macroscopic phenomena and struggles to capture the meaning of quantum phenomena. Intuition and language are crucial parts of understanding, but we find ourselves on treacherous ground when we try to apply human language or intuition to quantum systems. The confidence that comes with intuition and mastery of language are wholly counter-productive in quantum physics. You need to constantly reexamine what you think you know in the light of both new discoveries and the math. Both are sources of continual surprise.

Hence, if you think you understand quantum physics, you probably haven't given it enough thought.

Yes and because we live in the macroscopic dimension we develop the intuition this way and thinking in terms of another dimension makes out heads swirl. Perhaps we’ll never build an intuition at the quantum scale, just some tools to abstract it away and get some benefits out of it.

Feynman's diagrams (like the orbital model of the atom) were attempts to visualize a world we can't experience.

Those of us who aren't deaf and blind begin to make audio-visual models of our macro world in childhood. We can agree with confidence that "that's a rock" (if not its physical properties). Even in our macro world, our models have limits. (What do radio waves 'look like' as they're emitted by different antennas? Go around buildings? We draw pictures.)

But we're nearly deaf and blind to the quantum world. We have collections of indirect observations and math. Our models of it are necessarily limited by our inability to experience it directly. Those with math 'intuition' are in a better position ... but still.

> Feynman's diagrams (like the orbital model of the atom) were attempts to visualize a world we can't experience

They are a convenient notion for defining path integrals.

> [...] quantum physics on a deep, intuitive level, probably do not.

Intuition and understanding are two different things. Our intuition behind quantum mechanics is obviously wrong--quantum mechanics doesn't even obey regular logic! Quantum logic gets rid of a rule, for example.

So it's probably impossible to develop an intuitive understanding if that is your criteria.

As we've progressed, we slowly just get used to new abstractions. The concept of zero, negative numbers, complex numbers, rationals, reals, they were all very non-intuitive to the people at the time but were required to understand the science of the time.

> However, we lack an intuitive understanding of why the equations work.

I think the word mechanism captures what you are describing. We have a set of equations we know work, but we have absolutely no idea what underling mechanism gives rise to those equations. It's like the position ancient seafarers were in: the could use the sun and stars to navigate, but because they had no idea of the mechanism behind their movements they were happy with the idea the world was flat.

It's not a good state of affairs because understanding the mechanism yields far better predictive powers than the mathematical equation describing what happens. For example, we had good equations describing rechargeable battery degradation. But until we understood the mechanism, we were just stumbling about in the dark on how to improve it.

> quantum mechanics is well-understood

No it is not, we don't even understand what causes quantum waves to collapse. We call it "observation" but we don't know what that is. So you can't say we understand quantum mechanics until we have a theory which can accurately predict when wavefunctions will collapse.

When two quantum systems interact their wave functions either gets entangled with each other or they cause each other to collapse, it seems like the first happens in small systems and the second happens in large systems, we don't know much more than that.

From what it seems, we don't actually even know if "collapse" as such is a thing, nevermind what causes it.

We are one experiment away from placing serious doubts even on counterfactual definiteness.

Observation = interaction

But what kind of interaction? Is it a tensor field? Change of spacetime? Particle mediated? Wavelike? Nonlocal string or brane vibrations? Does it interact with other known fields and particles and in what way?

What is its range and speed? Does it work same in very low or very high energies? Does temperature affect it? Does it change at relativistic speeds?

Etc. The tool we have to probe it is currently only entanglement and secondarily tunneling. Maybe some calorimetry and charge detection.

None of that is a problem if you accept MWI.

The mental benefit of MWI is that instead of worrying about both entanglement and observation/wave function collapse, you only need to worry about entanglement because what Copenhagen describes as collapse is really just entanglement of the observed system with the observer.

A group of particles can interact without it counting as an observation which is why I mentioned entanglement.

https://en.wikipedia.org/wiki/Quantum_entanglement

Entangled particles share a state, they do not interact: https://en.wikipedia.org/wiki/No-communication_theorem

Particles can entangle when they interact, ie they go from not sharing a state to sharing a state. Your link states that wavefunction collapse is random and can thus not be used to communicate information, it doesn't say anything about interactions resulting in shared state.

I would bet ssijak thinks in MWI, where observation is not a separate thing from interaction and there is no such thing as "counting as an observation" in physical reality.

The "observations" that other interpretations cause endless confusion about are simply interactions between the observed system and the observer, but "observed system" and "observer" are concepts without a corresponding physically reality. They're simply labels assigned by humans.

Note that there's no way to observe a particle without some form of entanglement, though.

> I was thinking, the only real way for us to progress was to get the results out of CERN we were expecting--either smaller particles or some detection method which allows us to peer even deeper into the atom.

This cannot be more wrong.

Science advances with one unexpected experimental hypothesis failure at a time.

Quantum computers working (or not working) would be another important test for quantum mechanics. Because quantum computer with n qubits to work requires amplitudes to behave the way quantum mechanical equations predict to the precision of 1/2^n which for 256 bits is much larger than the precision amplitudes can be measured in principle (e.g. for a spin ~1/planck length)

Would higher precision of measurements help here?

Certainly. Effects can show up as qualitatively new things such as a new kind of particle, but also as tiny discrepancies between theory and experiment. There might be interesting new physics in every digit that gets added to the fundamental constants, for instance.

quantum mechanics is well-understood if you don't question the axioms and blindly calculate probabilistic outcomes. That doesn't mean we understand why those calculations work in the first place, what their limits are, what they represent, etc...

ptolemy might object, 'but we can calculate the answer with arbitrary precision!!' but his 'model' didn't capture reality in a meaningful way dispite the accuracy.

The same could be said of quantum mechanics

Ultimately you have to accept axioms of how the universe functions... we didn't question the non-probabilistic newtonian outcome metaphysically when it was presented, why is QM now being held to that standard?

We questioned non-probabilistic newtonian physics when we asked harder questions than it could answer or it gave wrong answers when pushed to a higher regime.

Something similar is going on now in QM.

And do you really think we have discovered the "axioms of the universe" already? That is quite haughty. No we don't have those.

We have models that perform well in certain regimes. Push those regimes, and the models break, then we find new and better models.. rinse and repeat. That is science. https://youtu.be/ka9KGqr5Wtw

* We have 'axioms' (I would not call them that. I would say models) that work in a limited regime e.g. quantum field theory or Maxwell's equations, but we don't know if they are truly fundamental - in that sense they are not axioms , which in math have a precise meaning.

> And do you really think we have discovered the "axioms of the universe" already? That is quite haughty. No we don't have those.

That is also haughty.

It is possible--probable, even--that our current models of physics are incorrect. We should certainly continue to probe them and look for something better. I would bet money that before my lifetime is out, some part of QM or relativity are overturned just like newtonian physics.

However, you cannot just assume that that is the case. It is no certainty. We do not have certainty. The universe is sticky; it very well may be that we can never reduce it to something as simple as rule 30. It may be that our current theorems are axiomatic--at least, as far as the universe can be said to have axioms.

(Or, as I think is more likely in the universe where we never discover any more physics, the difference between our laws and 'truth' is smaller than is possible to detect. That is, any instrument we use to measure will inherently produce too much noise to produce a statistically significantly different result.)

I agree that we should push our models to regimes where they fail. I disagree that the questions you were asking above will lead us there.

I've no idea about nuffink, but

> We need to get out there and get new empircal data.

To do this it's necessary to formulate the right questions first, as gathered data (which are in effect, answers) provide no information alone.

Edit: "provide no information alone" is a bit too strong. What I'm trying to say is more precise questions tend to produce more precise answers in normal discourse, I'm generalising that.

> excellent predictions

Sometimes QM provides us with no prediction whatsoever as to what will happen next (eg. photon splitting experiment.) I would not call this "excellent."

> chaotic nature

Chaos theory is something else.

> results out of CERN

Particle physics is also not particularly relevant to this discussion. Fundamental particles is a separate issue to interpretations (foundations) of QM.

Reply on news.ycombinator.com