Hacker News Reader

Comment by lisper | Hacker News Reader

> the universe as a whole is skewed in the forward direction, because its past endpoint was highly ordered, and its future endpoint is highly disordered.

This is highly misleading. The arrow of time is not thermodynamic, it's quantum. It's defined by increased entanglement, not increased entropy.

http://blog.rongarret.info/2014/10/parallel-universes-and-ar...

View on HN

This is highly misleading. The arrow of time is not thermodynamic, it's quantum. It's defined by increased entanglement, not increased entropy.

That's not at all what I remember from physics class when I was at university, and would seem to imply (incorrectly) that non-QM models can't produce an arrow of time.

Sorry, your grand statement matter-of-factly dismissing thermodynamic arrow of time needs a much better source than some computer scientist's armchair blog about QM.

I know it's been recently trendy for armchair physicists to post grand philosophical claims about the universe from some vague understandings of Bell's Inequality and/or entanglement, and blurring the line between philosophy of physics with physics. But your casual dismissal of entropic thermodynamic arrow-of-time time in favor of purely increasing entanglement ranks way high on the BS meter.

For context - I have a PhD in experimental condensed-matter physics, and TA'd thermodynamics and statistical mechanics for several years. I left the field of physics many years ago so maybe there were some notable recent developments in quantum information theory that I missed. But I skimmed your linked article and can't take it seriously. Would love a better source of you can provide one, eg peer-reviewed (in a respectable physics journal) or written by an actual career physicist at a reputable institution.

http://arxiv.org/abs/quant-ph/9605002

This doesn't discuss the arrow-of-time directly, but it's a pretty straightforward corollary.

Cheers for the article, quite interesting.

I'd still maintain the original comment in the original article is correct, that the arrow of time is still driven by 'thermodynamic' processes, where I'm allowing thermodynamic really means statistical mechanical process, and statistical mechanics includes (as it always has) quantum statistics of quantum processes.

Ie, your arxiv link gives further quantum detail of the underlying macroscopic measurements (eg, the schrodinger cat example).

But the irreversibility and arrow-of-time come from the enanglement of the particle with the 10^23 atoms of the macroscopic observer system. And in that sense the irreversibility is more due to the large size of the system, just like a classical example (such as shaking a bottle of originally separated black/white marbles) than the fact that there is quantum entanglement between the particle and the measurement system.

So while Cerf and Adami identified interesting quantum processes to describe underlying interactions, the irreversibility still comes from a statistical consideration of the interaction with a macroscopically large system. Just like the traditional arrow-of-time statement that you had originally refuted.

I respectfully disagree. I maintain there is a fundamental difference between the Cerf & Adami process (which is really just decoherence) and classical thermodynamic processes. That difference is: for ordinary thermodynamic processes, the odds of observing a decrease in entropy with increasing time is very small but nonetheless non-zero. In the decoherence model, the odds of observing a reversed entanglement is exactly zero because measurement is entanglement. As a practical matter it amounts to the same thing, but in principle it is as different as GR is from Newtonian mechanics.

Your arxiv link disagrees with you. From the paragraph after equation 6.6 :

>This irreversibility is completely equivalent to the irreversibility in classical mechanics. Indeed, classically, to reverse the microscopic time evolution, it is necessary to invert the velocity of all the particles, the practical impossibility of which gives a macroscopic irreversible aspect to time evolution. In quantum mechanics, it is necessary to undo any unitary evolution associated with all interactions that particles have undergone, so that reversibility is practically impossible if a macroscopic number of particles have been involved. We are led to conclude that irreversibility is not an inherent feature of quantum mechanics.

I know that looks like a disagreement, but it's actually not. Note that I said the odds of observing a reversed entanglement is exactly zero, not that the odds of it actually happening were exactly zero. The odds of it happening are comparable to the odds of thermodynamic reversal. The difference is that if a thermodynamic reversal were to happen, it could be observed. If a quantum reversal were to "happen" (I put "happen" in scare quotes because the whole concept of something "happening" becomes a little murky here) it could not possibly be observed, because any attempt to observe it would "prevent" it from "happening."

lisper, you're obviously a pretty smart guy or gal. But your nit-picking here over 'observing' vs 'happening' is so far from your original very bold claim that the arrow of time is not thermodynamic but quantum. You even went so far to say the original article was 'misleading' because of this.

I had to call you on this. Bold claims require strong confirmation. And even the arxiv paper you linked to makes the same statement that I defended - that irreversibility of purely quantum processes is still due to the macroscopic (ie large) system size, which immediately implies the same broad thermodynamic (and statistical ensemble) considerations we've been assuming for the past century.

We may just have to agree to disagree on this. IMHO the difference between the arrow of time being due to entanglement and the arrow of time being due to the 2nd law is analogous to the difference between gravity being a force and it being an artifact of the curvature of spacetime. Not a big difference as a practical matter but it's a huge difference conceptually. And in an article entitled "The quantum mechanics of fate" this difference matters.

Also, this is HN comments, not Physics Review A. Not everything that gets said here needs to pass the highest standards of peer review.

I actually tried checking, to see if 'retro-casually', at around 1 us timescale, a more 'entropic' outcome of otherwise perfectly balanced source of physical randomness would show up more often than it should.

Used a 2-mbit source of randomness, been running the setup continuously to get a few terabits of data. Fixed a bunch of imbalances in the state-of-the-art commercially available source of physical randomness. Nope. Source code/results are available here: https://code.google.com/p/reasonable-deviations/source/brows...

What's the difference? I'm familiar with (and endorse) the Julian Barbour "time capsule" interpretation of time, and I just read the link. It cites the exact same facts, and provides the same explanations, that I do when demonstrating that the arrow of time is thermodynamic (ie time slices can only contain memories of "pastward" states as Gary Drescher puts it).

The QM explanation is not contrary; it's just a special case of the thermodynamic explanation; entanglement is one way entropy can increase, but it happens in the classical world too, which also has time symmetric laws.

So in what sense is someone erring when they call it a thermodynamic arrow of time?

This is another Nautilus article, by Julian Barbour, on the arrow of time. http://nautil.us/issue/9/time/the-mystery-of-times-arrow

and this from Paul Davies on the flow of time. http://nautil.us/blog/where-did-time-come-from-and-why-does-...

I was reading about computation the other day and stumbled upon a concept where something can't be computable if it requires more operations than there is energy in the universe to do the operations. While reading the post you linked to, it struck me that the speed of light is a similar barrier to us. We can't travel the speed of light. We always like to talk about it and what we would 'observe', but it's theoretically impossible for us to do.

Why be so dogmatic? You can define an arrow of time going away from some initial ordered state even in a purely classical universe. No QM needed. Although the relationship between entanglement and entropy is worth considering for sure.

Is this an accepted viewpoint? I've never heard this before, and though I'm not a physicist I've certainly been taught that the arrow of time is thermodynamic.

Accepted by who? It's probably not the mainstream view (yet), but that's because there's a ton of misinformation out there about QM. For example, many people (even many physicists) think that the "measurement problem" (http://en.wikipedia.org/wiki/Measurement_problem) is still a problem. It isn't, and hasn't been for a very long time (decades). The reason people think it's still a problem is that the solution requires accepting the fact (and it is a fact) that classical reality isn't really real, it's just a very good approximation, and a lot of people are (understandably) resistant to that idea.

So if time is the result of increasing entanglement, what does it imply that massive objects (i.e. black holes) slow time?

Good question. I don't know. I suspect that answering it would require (or lead to) the unification of QM and GR.

Does increasing entanglement explain the second law of thermodynamics?

AFAIK this is an open question. My guess would be that yes, it does. I'm also guessing that the first person who actually does the math to prove it will win the Nobel prize in physics.

Reply on news.ycombinator.com