6 ms·

Atomic Spins Evade Heisenberg Uncertainty Principle

73 points by okket 10y ago

tbrownaw 10y ago
Heh. Ask only what you want to know, and nothing more. Good for QM because it gives you a sink to dump uncertainty, good for hiring because it helps you not get sued, good for avoiding miscommunications in general, ......
- mikekchar 10y ago
  In a related way, I often talk about the Heisenberg Uncertainty Principle of Metrics. The more you measure with the intent of adjusting the process, the more people intentionally and unintentionally drive their behaviour based on the metric. So it's a real skill to minimise measurement so that you are able to control the areas you need to control without creating incentives for people to do bizarre things in areas you didn't think about. It always drives me crazy when managers start measuring things "just in case I want to use the information in the future". It pretty much guarantees unpredictable behaviour of the workers.
  - abbeyj 10y ago
    This sounds like Goodhart's Law (https://en.wikipedia.org/wiki/Goodhart%27s_law https://en.wikipedia.org/wiki/Goodhart%27s_law)
  - philipov 10y ago
    This reminds me of a recent observation by (I think) some Princeton ethical philosopher, that any theory of human behavior becomes invalid once it becomes known by a significant proportion of the population, because people's behavior anticipates expectations of their behavior.
    - kordless 10y ago
      This is simply thinking about thinking, which is allowed (at least in Zen) as long as one doesn't think about thinking about thinking. In other words, holding irrational beliefs and spreading those beliefs in a viral way, either in the population, or inside one's own head, creates suffering if there is no means to end holding the irrational beliefs.
randomaxes 10y ago
This is a technique called squeezing, it in no way violates our fundamental understanding of physics. Heisenberg uncertainty is generally: 1/2ћ ≤ ΔxΔp. If you measure something in a way that you have complete uncertainty in the quantity p then you can have arbitrarily fine resolution in x. It's interesting, but squeezing is a common technique, and this headline misrepresents the importance of this experiment.
- jessriedel 10y ago
  It's not just the headline. Sentences like this are false: > ...measure the spin precession rate much more accurately than previously thought possible. since no one thought different.
photon-torpedo 10y ago
Clickbait headline, I'd say. The experiment fully agrees with known quantum mechanics. (Nevertheless, it is an impressive technical achievement.) The authors of the original Nature article say that "spins obey non-Heisenberg uncertainty relations", which is true if "Heisenberg uncertainty relation" is (narrowly) defined as the uncertainty relation for position and momentum, Δx⋅Δp ≥ ħ/2. But since spins don't have position and momentum, they are not expected to observe this uncertainty relation anyway. The more general uncertainty relation for any two observables A and B reads ΔA⋅ΔB ≥ |⟨[A,B]⟩|/2 (where [A,B]=AB-BA is the commutator, and ⟨⟩ is the expectation value) and for spins which have three components Sx, Sy, Sz the commutator is [Sx,Sy]=iħSz etc, therefore the uncertainty relation for two spin components reads ΔSx⋅ΔSy ≥ ħ/2⋅|⟨Sz⟩|, hence it does not seem so surprising that it can be made rather small if the expectation value of the third spin component can be made small. Edit: typo
- monktastic1 10y ago
  Also: "This is because to measure its position you have to disturb its momentum...". Another article in the long, illustrious tradition of mis-explaining HUP as being due to measurement "disturbing" some property or the other.
  - jeremysalwen 10y ago
    A tradition started by Heisenberg himself!
    - Xcelerate 10y ago
      That's what's funny about the whole thing; Heisenberg mistakenly thought his own equation was due to the observer effect. There's a good article on him almost failing his PhD exam because he had difficulty with a related concept (https://www.aps.org/publications/apsnews/199801/heisenberg.cfm https://www.aps.org/publications/apsnews/199801/heisenberg.c...). It's amusing because he was later awarded the Nobel prize for "the creation of quantum mechanics".
  - ulucs 10y ago
mathw 10y ago
The article's introduction explains the uncertainty principle as the observer effect, where the process of measurement changes a system. Which it's not, quantum uncertainty is a fundamental consequence of the underlying physics, it has nothing to do with influence from the measurement process. And that's about as far as I understand it. The whole thing seems intended to make my primitive brain hurt. But any experiment which starts off by cooling things down to "a few microkelvin" definitely has my appreciation.
- parenthephobia 10y ago
  > Quantum uncertainty is a fundamental consequence of the underlying physics, it has nothing to do with influence from the measurement process. If you can prove that, your Nobel prize awaits. :) The truth is that nobody knows why there is quantum uncertainty. The common belief is that the uncertainty is "real", but non-local hidden variable theories have not been proven to be impossible. (In a recent survey of physicists, 47% favoured explicitly real uncertainty - https://arxiv.org/pdf/1612.00676.pdf https://arxiv.org/pdf/1612.00676.pdf)

73 points by okket 10y ago

Heh. Ask only what you want to know, and nothing more. Good for QM because it gives you a sink to dump uncertainty, good for hiring because it helps you not get sued, good for avoiding miscommunications in general, ......

mikekchar 10y ago
In a related way, I often talk about the Heisenberg Uncertainty Principle of Metrics. The more you measure with the intent of adjusting the process, the more people intentionally and unintentionally drive their behaviour based on the metric. So it's a real skill to minimise measurement so that you are able to control the areas you need to control without creating incentives for people to do bizarre things in areas you didn't think about. It always drives me crazy when managers start measuring things "just in case I want to use the information in the future". It pretty much guarantees unpredictable behaviour of the workers.
- abbeyj 10y ago
  This sounds like Goodhart's Law (https://en.wikipedia.org/wiki/Goodhart%27s_law https://en.wikipedia.org/wiki/Goodhart%27s_law)
- philipov 10y ago
  This reminds me of a recent observation by (I think) some Princeton ethical philosopher, that any theory of human behavior becomes invalid once it becomes known by a significant proportion of the population, because people's behavior anticipates expectations of their behavior.
  - kordless 10y ago
    This is simply thinking about thinking, which is allowed (at least in Zen) as long as one doesn't think about thinking about thinking. In other words, holding irrational beliefs and spreading those beliefs in a viral way, either in the population, or inside one's own head, creates suffering if there is no means to end holding the irrational beliefs.

randomaxes 10y ago

This is a technique called squeezing, it in no way violates our fundamental understanding of physics. Heisenberg uncertainty is generally: 1/2ћ ≤ ΔxΔp. If you measure something in a way that you have complete uncertainty in the quantity p then you can have arbitrarily fine resolution in x. It's interesting, but squeezing is a common technique, and this headline misrepresents the importance of this experiment.

jessriedel 10y ago
It's not just the headline. Sentences like this are false: > ...measure the spin precession rate much more accurately than previously thought possible. since no one thought different.

photon-torpedo 10y ago

Clickbait headline, I'd say. The experiment fully agrees with known quantum mechanics. (Nevertheless, it is an impressive technical achievement.) The authors of the original Nature article say that "spins obey non-Heisenberg uncertainty relations", which is true if "Heisenberg uncertainty relation" is (narrowly) defined as the uncertainty relation for position and momentum, Δx⋅Δp ≥ ħ/2. But since spins don't have position and momentum, they are not expected to observe this uncertainty relation anyway. The more general uncertainty relation for any two observables A and B reads ΔA⋅ΔB ≥ |⟨[A,B]⟩|/2 (where [A,B]=AB-BA is the commutator, and ⟨⟩ is the expectation value) and for spins which have three components Sx, Sy, Sz the commutator is [Sx,Sy]=iħSz etc, therefore the uncertainty relation for two spin components reads ΔSx⋅ΔSy ≥ ħ/2⋅|⟨Sz⟩|, hence it does not seem so surprising that it can be made rather small if the expectation value of the third spin component can be made small. Edit: typo

monktastic1 10y ago
Also: "This is because to measure its position you have to disturb its momentum...". Another article in the long, illustrious tradition of mis-explaining HUP as being due to measurement "disturbing" some property or the other.
- jeremysalwen 10y ago
  A tradition started by Heisenberg himself!
  - Xcelerate 10y ago
    That's what's funny about the whole thing; Heisenberg mistakenly thought his own equation was due to the observer effect. There's a good article on him almost failing his PhD exam because he had difficulty with a related concept (https://www.aps.org/publications/apsnews/199801/heisenberg.cfm https://www.aps.org/publications/apsnews/199801/heisenberg.c...). It's amusing because he was later awarded the Nobel prize for "the creation of quantum mechanics".
- ulucs 10y ago

mathw 10y ago

The article's introduction explains the uncertainty principle as the observer effect, where the process of measurement changes a system. Which it's not, quantum uncertainty is a fundamental consequence of the underlying physics, it has nothing to do with influence from the measurement process. And that's about as far as I understand it. The whole thing seems intended to make my primitive brain hurt. But any experiment which starts off by cooling things down to "a few microkelvin" definitely has my appreciation.

parenthephobia 10y ago
> Quantum uncertainty is a fundamental consequence of the underlying physics, it has nothing to do with influence from the measurement process. If you can prove that, your Nobel prize awaits. :) The truth is that nobody knows why there is quantum uncertainty. The common belief is that the uncertainty is "real", but non-local hidden variable theories have not been proven to be impossible. (In a recent survey of physicists, 47% favoured explicitly real uncertainty - https://arxiv.org/pdf/1612.00676.pdf https://arxiv.org/pdf/1612.00676.pdf)