« first day (1558 days earlier)      last day (1768 days later) » 

12:01 AM
@DavidZ and if the OP decided that I solved his problem, why do you decide that I didn't solve? 15 points went on your this decision. I think that at least the decision if I solved or not the problem belongs to the OP.
 
This is, I think, a fine example of a conceptual question.
 
@AlfredCentauri and @DavidZ if I ask a question a certain person, I deserve the respect to be answered by that person, and to my question. And the question was, who decides here if a question, conceptual or else, was answered.
 
@Sofia Who told you that? You don't have a claim on the time of the other volunteers on this site.
 
12:16 AM
@dmckee I don't understand. Please be more clear.
 
What makes you think that the fact that you asked a question puts a onus on anyone to answer it. We're all doing this on a volunteer basis.
 
@Sofia I don't have the slightest idea what you're referring to. I saw a question, made a post here to attract attention to it, and, with respect to whatever you may have posted earlier, I have not read it nor does it any relevance whatsoever to my post.
 
@ChrisWhite It's been a while since it came up on meta, but at the time people seem to feel that the field was physics.
3
Q: Should We Allow Econophysics Questions?

GravitonMy question on econophysics is closed for being 'off-topic'. But at this stage-- when there is no econophysics forum and the boundary of Physics SE is less defined and rigid-- it should not be closed. Also one must understand that econophysics is physics, with some economics bent. Much like how e...

 
@dmckee, @AlfredCentauri, @DavidZ I have the feeling that I ask one question and everybody answers me something else.
The OP decided that I solved his problem. Who's decision is if the problem was solved or not?
 
@Sofia It's the OP's privilege to award the green check-mark, and other user's privilege to vote.
 
user54412
12:21 AM
@dmckee Ah I remember that post now. And apparently I was the only one who upvoted your answer.
 
You are arguing that your answer is good for the asker. I don't think that anyone would disagree that the asker thinks that, but some folks feel that answering those questions is bad for the site.
 
@dmckee : O.K., but you deleted the answer, just don't delete the fact that I solved the asker problem and the points he gave me for this. And, in general, a too detailed answer can be adjusted, if you'd advise me. Only the basic principles can be left. The advise help, the punishment is only the whip, but teaches me nothing.
 
It wasn't me who deleted it (though I approve), and the moderator who did left you a comment explaining why he did it. The comment says that the deletion is "temporar[y]" and that it is in accord with our policy on this matter as set on meta.
You answer will be undeleted in a couple of weeks (assuming of course that the whole question hasn't been reaped by then).
 
12:42 AM
@ChrisWhite Econophysics papers are published on Phys Rev Lett rather than the J Finance, and by that alone I'd say the field is interesting to physicists and actually probably closer to physics than finance or economics. What you most typically have is a sort of an Ising model, where instead of magnetic spins you have maybe people. You still look at phase transitions and use a very physics-oriented toolset and language.
I should add that this is distinct from the math style writing with theorem-lemma-proof often used in mainstream economics.
 
@alarge Now that is a pertinent fact that I hadn't known.
 
@ACuriousMind I kind of disagree with this, actually.
If a bad question gets a good answer that tells me that the question should be edited.
I guess closing/holding is supposed to invite OP to improve it, but it doesn't seem to go that way in practice.
Difficult to say what is the right course of action.
 
@DanielSank If the question is salvagable, that is of course preferable to closing!
 
@ACuriousMind Sure, but it seems like all too often folks just VTC without even trying a little bit to improve the question.
This reeks of the VTC bandwagon.
It's actually a pretty good question, IMO. The problem is that OP doesn't know enough to pose the question clearly.
 
12:58 AM
If the question is salvageable by editing in a way that is appropriate for other people (not the OP) to edit, then sure, edit it instead of closing it. But sometimes, what it takes to make the question acceptable is a drastic edit that changes the apparent meaning of the question. In that case, that's when you vote to put it on hold.
 
The pattern of asking a question poorly for lack of precisely the knowledge that question's answer would give you is extremely common.
I think we should be mindful of that.
I can argue this two ways:
1) If your goal is to help people learn, then it's obvious you shouldn't VTC questions which lack clarity for the reason I described above.
2) If your goal is to create value for the site, then if a question has a good essense but poor execution then you are clearly adding more value by editing.
"But sometimes, what it takes to make the question acceptable is a drastic edit that changes the apparent meaning of the question."

Not sure I agree with this, for the reason I give above.
It is extremely common to ask a question poorly precisely because you don't know the answer.
I think it's the would-be answerer's job to help out there.
 
@DanielSank I don't think so, or at least I don't think it's as common as you do. I think in many cases, what happens is that the asker lacks the background knowledge to even understand the question or the answer, not just the knowledge that would be provided by the answer.
Remember that this is an expert-level site.
 
@DavidZ People throw that sentence around every time this issue comes up.
I'm an expert, and I still suffer the pattern I describe!
 
@DanielSank so are you saying you believe that it is never the case that a question can be fixed by a drastic edit?
 
"Expert" is totally relative.
 
1:03 AM
@DanielSank The question still seems to me to ask something different and broader than what you answered. The question asks: "Why do macroscopic processes take well-defined times if all quantum processes only have transition probabilities?". Your answer then shows that transition probabilities are "classical" rather than "quantum".
 
@DavidZ Oops, misrepresented what I meant there. What I mean is that, yes, it takes a "big" edit, but that's ok.
 
I'm really a bit confused that OP accepted that answer without further comment
 
I also don't think the necessary edits are as big as you're making them out to be, in many cases.
@ACuriousMind Huh?
What my example shows is that the transition probability is
1) Dependent on coupling strength to the environment.
2) Only a result of decoherence which happens because you ignore the environment.
 
@DanielSank sure, it's okay if a question takes a drastic (not necessarily big) edit to become acceptable. But those questions should be put on hold unless and until that edit is made.
 
Ah. I see. Yes, this is sensible in principle. However, I have my usual grip:
Suppose I find a question which is really interesting but badly worded. Suppose it is on hold. Now I have to edit it and then convince other people to un-hold it, and then *remember to check back later in order to give an answer*.
 
1:06 AM
@DanielSank Okay. And how does it answer the question? I'm still not even sure what the question actually is, even after your edit.
 
@ACuriousMind The question is a little bit complex, I think. Quantum mechanics is "random" in some sense. Macroscopic processes proceed at well understood and repeatable rates. OP is trying to reconcile those two facts, roughly.
 
@DanielSank yep. As far as I'm concerned, that's not too much to expect of people who really really want to give an answer to a badly posed question.
 
I agree that my answer is not stellar, because I didn't explain how the average of jump processes gives e.g. exponential decay, but I think because I gave so much detail and laid it out OP can now figure things out him/herself.
 
And the system does help getting edited questions reopened. They automatically go into the review queue after being edited.
 
@DanielSank I believe SE as a whole thinks you are right, that's why they introduced the explainer-refiner‌​-illuminator badges a few months back.
 
1:09 AM
@DavidZ I can't remember the last time I saw that in the review queue.
@DavidZ I have the same problem with this as I do about down votes (although I've started using them a little bit more).
 
@DanielSank I review about 3 or 4 questions a day in the reopen queue that are there because they were edited.
Unfortunately, most edits are either trivial or do not address the close reason at all.
 
This must be a thing where I don't have enough rep for that.
@DavidZ ...namely that it's far easier to VTC/VTH than to remember to come back later and give an answer that I couldn't give before.
I'm not saying we shouldn't ever VTC/VTH, just that we should keep in mind where we're spending our effort.
 
1:29 AM
In unrelated news, if I had known that I'd get 8 upvotes on this answer, I would not have written it. It saddens me that effortless answers to effortless questions get disproportionally more upvotes than elaborate answers to well-posed questions.
 
1:39 AM
@ACuriousMind That's the nature of the beast, I'm afraid.
 
@ACuriousMind I upvoted that answer to teach you this lesson, CuriousGrasshopper.
 
Thanks, I guess
 
@ACuriousMind it saddens me too, but that doesn't seem like a good reason not to have posted the answer in the first place.
 
2:07 AM
@DavidZ and CG, it is simply a fact of reality that, on occasion, one is unexpectedly 'rewarded' for that which one thinks of as a trifle and, frequently, unexpectedly 'unrewarded' for that which one thinks of as a masterpiece (or a close approximation thereof). Of course, a genuine masterpiece is its own reward.
 
@ACuriousMind I think this is more evidence of the need to cherish every upvote for your good answers and ignorance of the ones that are effortless.
Also, what's kinda sad is that the question has garnered a single upvote to your 8 upvotes.
 
@KyleKanos Yeah, I was puzzled by that, too.
I mean, I know why I didn't upvote it, but I wouldn't have voted up my answer, either.
 
@alarge what were u sayin 'bout tight-binding models?
I just realized, I dont know what basis i am in @alarge .. or maybe I don't know what types of states the matrices act on?
@alarge o yeah.. it has to do with two observables... correct? so I use tensor product.. then I can work with matrices..
btw, the matrix for number operator is {{1,0},{0,0}} ...correct?
hey... didn't see @vzn .. u there?
 
@KyleKanos I think that's spot on.
 
0
Q: Should the engineering close vote be changed in light of the new engineering SE site

Chris MuellerQuestions are regularly closed here because they are considered engineering questions. The current close vote option for this reads: This question is off-topic because it appears to be about engineering, which is the application of scientific knowledge to construct a solution to solve a spec...

 
2:22 AM
@TAbraham Are you working with the two site or the one site model? If the latter, do you remember when Mark told you that you should be looking into 4x4 matrices?
Not latter, former.
I can't edit when I'm on my phone.
 
@alarge yep.. BTW, thanks for unblocking me.. I will not annoy you anymore and if I do, I understand if u don't respond.. it's very appreciated that you are still helping me...
@alarge ah.. I can on my ipad i think.. that would be suprising if you can't on an iphone (bcos same OS)..
@alarge I am wrting a 4x4 matrix.. and since it seems the math to find the eigenvalues and eigenvectors gets complicated, am using wolframalpha.. @MarkMitchison recommended to also use a computational program.. However, the main problem is writing the matrix.. I got the first terms.. i think.. it's just that the subscripts is messing me up... u see, I think I have something confused.. that's wat I want to clear up.. then, I think everything else will b fine...
 
Before you proceed I just want to make sure: You know how to solve the two-state, one site system with an analogous Hamiltonian analytically, right?
 
2:38 AM
@alarge one site? I am pretty sure I can do that.. MM never told me to do that.. said it's too easy, I think.. I think so too..
 
To be fair, you said you knew Pauli matrices. If you do, the two state general Hamiltonian is probably something you will have covered already.
 
@alarge oh yeah.. of course!
 
You can refresh your memory by working out the wiki derivation, and then move onto the 4x4 matrix (which ought to have more interesting things going on as you have the Vij terms)
 
2:53 AM
@alarge but aren't there creation and annihilation operators?
 
user54412
Moderators and other interested parties: should I go and mark the engineering faq answer I wrote as accepted? I don't think anything there is too controversial.
 
3:11 AM
@ChrisWhite I don't see what the issue would be in accepting it.
 
 
1 hour later…
4:30 AM
@alarge I am confused.. the subscripts confuse me...
hey @vzn .. want to talk?
haven't seen u in a while @vzn
 
 
2 hours later…
6:52 AM
0
Q: Structure of Metallic Hydrogen

Joseph HankelI am simply curious what the theoretical arrangement of metallic hydrogen would be. I found this article which gives a picture, but I don't understand what the labels mean (e.g. is H5 an isotope, some kind of charge, or just denoting the layer?). I would like to know: 1) If the image is the ...

on topic?
@ChrisWhite sure, it's been around long enough and is top voted
so that should be pretty uncontroversial
(I think)
 
 
3 hours later…
9:26 AM
Hi Physics StackExchange, I asked this question: physics.stackexchange.com/questions/163861/… but I did not get a clear answer. I made a chat room but the question wasn't really answered from it (I still have some questions). How can I improve the question to get a good answer?
I'm used to TeX.SE (which is smaller I think?) and they always answer really fast so I think I did something wrong.
I'll be back later.
 
9:55 AM
0
Q: How much cheaper and smaller on average is SiPM technology over conventional PMTs?

Still ThinkingI've heard it said that some of the major advantages of silicon photomultipliers are its low cost and compactness when compared to widely used photomultiplier tubes, but haven't found much information on the exact magnitude of the difference.

Off-topic?
 
 
2 hours later…
11:51 AM
@Qmechanic I think so
@user55789 answers don't come so quickly here. The only thing you did wrong was applying your expectations from another SE site to this one.
That being said, it's not really clear to me what exactly you're asking in that question.
 
 
2 hours later…
1:43 PM
hey guys i have a question from optic...
about laser and the reflection and refraction of it
When a laser beam is aimed at a wire, a circle of light can be observed on a screen
perpendicular to the wire. Explain this phenomenon and investigate how it depends on
the relevant parameters.
 
1
A: Laser, wire, and a circle of light

Tao TienSo I did it myself and figured it out. It's just really simple reflection. Basically the wire acts as a mirror, albeit a curved one. Because the light from a laser is straight, we won't have messy light. When a laser reflects off of a flat surface, it keeps going straight. However when it refl...

 
thanks a lot @KyleKanos <3
 
2:13 PM
the theory of this phenomenon isn't complete in this link. can u make it more clear for me plz?! @KyleKanos
 
2:40 PM
^ "<3" Seriously ?
 
@Gaurav Valentine's day is coming, love is in the air ;)
 
@ACuriousMind Ha !
@ACuriousMind Unfortunately over here, the sadists scheduled semester examinations to start exactly at V-day. Maybe your situation is better.
 
@Qmechanic That kind of data feeds into experimental design, but there isn't real a the answer. Just a whole bunch of "what I learned from looking through Hamamatsu's catalog this month" answers.
 
@Gaurav Heh, I've got my last exam exactly one day before it
 
@ACuriousMind Lucky you. You guys have a two-year post grad course right ?
 
2:52 PM
@Gaurav I'm never sure how this exactly translates into other models, but I think the two-year masters course I'm doing is a grad, not a post grad course
 
@ACuriousMind Oh good, I'd thought you'd completed that stage. So we're on more relatable terms here. How far through the course are you ?
 
@Gaurav It's just the end of the first semester of it now.
But the second year of the course is light on lectures, anyway, and puts emphasis on working in a group and writing your thesis.
 
 
1 hour later…
4:09 PM
nobody answered me :(
 
4:51 PM
@FarinazTalebi I'm not all that interested in doing your work for you. I think the link provides enough information to at least start you on the right path.
 
5:06 PM
Apologies for my potentially confusing messing around with comments on this question.
I was trying to edit my previous comment but you already posted an answer.
But ended up deleting it and redoing because it had been 5 mins...
Not the first time I have come afoul of the annoying 5 minute limit.
 
@MarkMitchison Heh, it's alright, I think were done there now :)
 
:)
It may also be that there is a physical reason why it only makes sense to have Lie gauge groups in fundamental field theories, in which case that would be a better answer to the question than my suggestion about effective low-energy symmetries. But I don't know nearly enough about YM theory to know what that could be,
 
AvZ
I wonder who comes up with the name for all these Chat rooms...
 
9
Q: What should we name our chat room?

KevinMany other StackExchange sites have a name for their chatroom. MSO has "The Tavern," SU has "Root Access," SF has "The Heap," etc. Our chatroom here is "Physics." Lame. We need a better name. I propose "The h Bar."

The community did
 
@MarkMitchison The formalism of gauge theories simply needs a group that is also a manifold, in particular, you need access to the Lie algebra as the tangent space. It will be difficult to come up with a "physical reason" for this because "physically", you don't see the gauge symmetry at all, only its strange results like the A-B effect.
Or rather, to define "local gauge transformation", you need to be able to write down the notion of a smooth map into the group, and groups with smooth structure are Lie groups
@KyleKanos "The coupling"...that sounds more like some physicist dating service
3
 
5:20 PM
@ACuriousMind I was thinking something very similar
 
@ACuriousMind oh god
@ACuriousMind So perhaps this means that quantum physics on continuum space time can only have a continuous gauge symmetry. I only know of discrete gauge symmetries being usefully applied to lattice Hamiltonians, although this may just be ignorance on my part.
 
@MarkMitchison That makes even more sense! The idea of "local transformation" demand that the group be "modeled" on the same object as the spacetime - in the continuous case, this is $\mathbb{R}$, but if spacetime is a lattice, one could probably also work with discrete group that are "modeled" on lattices.
 
Well then I guess that may be the answer to the question, if you want to write it up :)
 
The problem is then that you don't have a notion of derivative
And the crucial point of gauge theories is that they modify the usual derivative into the gauge covariant derivative
The discretized version of gauge theories on continuous spacetimes also usually retains the ful Lie group as symmetry group
What is changed is what the gauge field is - given a 2D lattice, the gauge field "degenerates" into an assignment of group elements to the edges of your lattice
Hm, the idea sounded interesting, but I think the fact that discretizing usual gauge theories doesn't break the continuous character of the group means that it is not really feasible to think of discrete gauge groups on a lattice in this sense
(Though you may of course have discrete symmetries on the lattice that are "unphysical" - they just aren't gauge symmetries in my narrow sense)
I'll stop think out loud now, I think.
 
@ACuriousMind Right, but this is because you need lattice YM to agree with continuum YM in the limit.
As you say, "unphysical" models don't have this restriction so that you may consider any group of internal symmetries
 
5:36 PM
@MarkMitchison That is also true. So, yes, if I do not demand that there is a continuum limit, the link variable on the edges are not constrained to come from a Lie group.
Alright, I'll think a bit about it and write it as an answer if I don't find anything wrong with it (and noone else is faster)
What I'm thinking about here is something different from the discrete-gauge-from-SSB from the paper you linked, though, @MarkMitchison.
 
@ACuriousMind Yes, I think that's true. I guess that the mathematics developed in those lecture notes to describe the properties of the states can be useful in the context of the low-energy lattice Hamiltonian scenario I described. I suppose this only because I know someone who has used it (or rather who had it on their desk!) and who works on lattice spin models supporting topological excitations (in the context of quantum computation).
It was just the only reference I could remember off the top of my head. But if you search "toric code discrete gauge symmetries" several references come up from people like Kitaev or Lahtinen who work in the same field as my friend.
 
6:43 PM
I've been staring at this for a few minutes. It's sorta interesting, but I think the real question is, Why would you want to do this?
0
Q: Chocolate dynamics

kryomaximNow I have found a possible model on how to describe chocolate when it is chewed. It has to do with geometrical transformations when a curve $\gamma$ intersects a manifold $M$. The chocolate is simply the manifold $M$ at the beginning. When the teeth that propagating along the curve $\gamma$ cro...

 
 
1 hour later…
7:44 PM
@ACuriousMind My answer to someone who names himself Physicsaholic : the Curious Mind answer makes sense, and there are all sort of exchanges between the system and the apparatus of measurement, but the apparatus doesn't measure itself. It doesn't give to the system energy, if it is energy that we measure. The apparatus won't falsify the system. Do you agree?
 
@Sofia An energy measurement forces the system into an energy eigenstate, thus changing its energy expectation value to the eigenvalue of that state. The "collapse" into a certain state that measurements induce does not obey any of the usual conservation laws because that collapse is not a consequence of the physical laws within the system (in particular, it is not induced by the time evolution of the isolated system).
 
@ACuriousMind No!!!!!!!!!!!!!!!!!!!!!! The apparatus doesn't measure itself, it doesn't touch the observable it measures. It touches the dual of that observable. The apparatus doesn't tell the system into which state to collapse.
 
@Sofia It doesn't tell the system which energy eigenstate to occupy, but it forces it to occupy an energy eigenstate
(I don't know what you mean with "dual of an observable")
 
user54412
@DavidZ I don't see why not. I'd hesitate to call it chemistry in the traditional sense, given that we've never made metallic hydrogen and whatever "bonds" it has are so far removed from the ones we quantitatively understand that we hardly even know the order of magnitude of the phase transition.
 
user54412
Actually I suppose the OP isn't clear whether the question is about metallic or graphene-like hydrogen
 
7:55 PM
One should really think of energy as the Hamiltonian, and the conservation of energy as the statement that the expectation value of the Hamiltonian is conserved under time evolution
 
@ACuriousMind : Regretably, even this you don't know. (I mean we don't know.) Please don't argue about this, just wait, I will tell you sometime. Which apparatus forces the collapse? Alice's apparatus or Bob's?
 
@Sofia Ah, I "know" this because this is what a von Neumann measurement does by definition.
 
I don't understand, please be more clear!
What has von Neumann's postulate to do with Alice and Bob?
 
I actually wrote an answer regarding the von Neumann measurement:
10
A: Simple example showing why measurement & interaction are different

ACuriousMindThis is not a settled question. Just as it is still debated whether or not there is wavefunction collapse, so is it debated what exactly we should understand by a measurement. In the following, we will go through the ideas behind the von Neumann measurement scheme, which is one way to try and tal...

A von Neumann measurement is the simplest way to meaningfully describe what we expect a measurement apparatus to do on the quantum level.
Regardless of the details, a measurement is an interaction with the measurement apparatus, and hence the system that is measured is inevitably disturbed
There's something like a "weak" measurement where you don't really disturb the system, but that's a special case
 
No, no, no. Which one of the apparatuses, Alice's or Bob's causes the collapse? The answer is both.
 
8:00 PM
@Sofia Who are Alice and Bob now?
I also don't claim "collapse" happens, which is why I wrote it in quotation marks
 
Two observers that have to do with an entanglement, Alice measures her particle, Bob his particle.
 
Wait, we are talking about this question, right:
2
Q: What is the energy of a superposition of energy eigenstates?

Physicsaholic$\newcommand{\ket}[1]{\lvert #1 \rangle}$Suppose I have a system say SHO in a superposition of energy eigen states $\ket{n_1}$ and $\ket{n_2} $ given by $\ket{\psi} = \frac{1}{\sqrt{2}}\ket{n_1} + \frac{1}{\sqrt{2}}\ket{n_2} $. If I measure the energy of system there is equal probability of gett...

 
@ACuriousMind yes, just you made a correction to an answer that I gave to someone, and I have objections.
@ACuriousMind I leave the decision to you, shall we answer to our nice ancestor first?
 
@Sofia Okay. My objection to your answer is that you are not using the proper notion of "constant of motion" in the quantum setting. The classical conservation laws are implemented in the quantum theory as operator constraints, which in turn induce only conservation of expectation values
You seem to disagree with this statement, yes?
 
@ACuriousMind it displeases me to disagree with you, but I have a bad feeling about this. By the way, I don't know why the continuity eq. goes wrong, so I take this back. But I don't like the issue with the average energy as constant of motion. Of course it is constant, but I a constant of motion should be an eigenvalue. Can we say for the particle passing through the 2slit that the mean path is passing through the middle of the wall?
 
8:13 PM
@Sofia Yes. An average does not need to be actually realized by anything in the data set.
 
@ACuriousMind you see, it even seems to me that pedagogically it's a bad idea. It encourages people to believe that though there is some sort of constant energy. And there is none.
 
It's just another counterintuitive thing when passing from the classical to the quantum theory, but it is true - my claim that the classical conservation laws lead to conservation of expectation values is just a special case of Ehrenfest's theorem
@Sofia But if you do not say it like this, then you would not have a notion of conservation of energy although the theory is time-translation invariant
...which would render Noether's theorem meaningless on the quantum level, but it is not!
 
@ACuriousMind one step at one time, please!
Emmy Noether didn't have in mind QM - the Schrodinger eq. appeared 10 years after Emmy Noether. But you will not consider this a serious argument.
@ACuriousMind Ehrenfest's theorem(s) are not worth much unless we have to do with quite localized wave-packets.
 
@Sofia What? Ehrenfest's theorem holds for all states
 
@ACuriousMind of course holds, but for non-localized states it is as in my example with the average path being in the middle of the wall.
 
8:25 PM
Well, you can choose to not view the conservation of expectation values as a meaningful thing to say. It is nevertheless the case that whenever someone talks about conservation laws in the quantum setting, they mean precisely the implementation of the classical laws as operator laws, which in turn implies the conservation of expectation values.
 
@ACuriousMind "quantum setting" ??? Which setting? We mean in the quantum theory?
 
Yes, I mean in a quantum theory
I understand that "conservation of expectation values" is not what our intuition understands as conservation of a quantity, but our intuition is not worth much when reasoning about quantum theory, really.
 
@ACuriousMind yes, for a classical mind the averages are more intuitive. This is what I say that you gave them. Something classical on what to rely, a classical refuge. I wouldn't have done this.
@ACuriousMind I notice that I answer you before you send me your thoughts.
 
So, to summarize - we do not actually disagree about any fact of the theory, you just don't like the usage of the classical language of conservation for the quantum version of it?
 
@ACuriousMind we disagree about pedagogy.
 
8:33 PM
Yeah, that's another way to state it
Coming from a QFT background, it is just that I see the importance of understanding already in QM how classical constraints on the theory are implemented on the quantum level
I don't really care for any ontological meaning of the procedures - whether or not it is a "good picture" to say that the expectation value is conserved matters far less than the fact that the precise mathematical meaning of that statement leads to the correct theory
 
@ACuriousMind my dear one, I am a phenomenologist. Correct theory you say? Let's sometime have a real talk. I want to tell you a couple of things, i.e. QM vs. logic. And I'll want to see your reaction. QM is illogical from the point of view of human logic. And I don't mean that it is not intuitive, this difficulty is too small.
@ACuriousMind : no logic classical or not, is satisfied by QM.
 
@Sofia There's a wonderful answer by Valter Moretti that shows that quantum logic, or rather, the quantum probability theory is indeed very different from classical logic. I am aware of that.
But the logic of QM is a valid logic - it is just not the logic our minds have evolved to apply
 
@ACuriousMind my dear, I don't think that you are aware, by whatever I saw from your answers until now. Let's talk sometime.
@ACuriousMind no, no. A talk would be better. And now, I go to a movie. Sleep well tonight and sweet dreams !
 
@Sofia Heh, thanks, and have fun, then!
 
9:25 PM
So I don't understand the paper enough to determine if it's actually any good -- maybe somebody here can shed some light on the validity? arxiv.org/abs/1404.3093v3
No big bang would be kind of a big deal
 
@tpg2114 The "previous" paper to that one is this: arxiv.org/abs/1311.6539
If the reasoning in that one is valid, the newer one is also correct
 
Hrm... also outside of things I understand
 
@tpg2114 that paper is based on Bohmian mechanics, I'd be very skeptical already, also that quantum Raychaudhuri equation I've only seen in loop quantum gravity (it may be my ignorance), so that's two huge issues tbh :)
 
I get stuck at "the wave function of a quantum fluid", actually
 
Why do people have issues with Bohmian mechanics? Doesn't it make the same predictions as the Copenhagen intepretation?
 
9:35 PM
Even before they begin with the "Bohmian" stuff - why should a fluid have a "wavefunction", and why should it only depend on one coordinate?
Is the "fluid" a particle?!
@KyleKanos Yeah, but it does a lot of special pleading to get there, it isn't amenable to be generalized to QFT without much work, and stuff like symmetries, spin and other things are a hassle to work with
 
I've always wanted to dig into quantum turbulence
Quantum turbulence is the name given to the turbulent flow – the chaotic motion of a fluid at high flow rates – of quantum fluids, such as superfluids which have been cooled to temperatures close to absolute zero. == Introduction == The turbulence of classical fluids is an everyday phenomenon, which can be readily observed in the flow of a stream or river. When turning on a water tap, one notices that at first the water flows out in a regular fashion (called laminar flow), but if the tap is turned up to higher flow rates, the flow becomes decorated with irregular bulges, unpredictably splitting...
But I haven't had the time
 
The essential step of both papers seems to be the claim that the correct quantum corrections to GR are obtained by replacing the geodesics by Bohmian trajectories.
 
@bolbteppa It must have some validity right? Otherwise it wouldn't be published in peer reviewed journals?
 
But they literally give no argument for that - they say, Bohmian mechanics is equivalent to normal QM, and so classical geodesics must be replaced by Bohmiann trajectories
It sounds like a total non-sequitur to me
That Bohmian mechanics is valid does not obviously imply that the quantum notion of geodesic is a Bohmian trajectory
@tpg2114 It is a valid, equivalent formulation of quantum mechanics, and it is very instructive to see that a non-local hidden variable theory can be the same as a local probabilistic theory
 
9:45 PM
@ACuriousMind no it isn't valid, did you read Lubos's explanation of it? The theory most definitely isn't valid, he has gone over the problems with it at least 10 times on his blog, how can you defend this stuff?
 
@bolbteppa You have to take everything Lubos writes with a grain of salt
 
Plus it seems that he's arguing it's kinda useless, not that it's wrong
 
^that
For him, Bohmian mechanics is just an exercise for people who do not want to accept that the world is probabilistic in nature
And I kinda agree with that, but Bohmian mechanics is precisely engineered to be indistinguishable from QM
But the papers don't even really do Bohmian mechanics
 
Here is a bullet point critique of Bohmian mechanics by him motls.blogspot.ie/2009/01/… if you got the impression he was saying it's useless, because he said "Bohmian mechanics is at least vaguely defensible in the non-relativistic quantum mechanical models only; in more general theories, it collapses completely."
 
They just say: "Hey, let's pretend that geodesics are Bohmian trajectories, and just add the terms that miss to make the geodesic equation a Bohmian guiding equation (which, btw, in this case is just the Hamilton-Jacobi equation which also is valid in standard QM)-
 
9:50 PM
Doesn't the assumption of quantum trajectories necessitate Bohmian mechanics?
Or imply it?
 
No offense to people who are way smarter than me, but I would trust something that appears in a peer reviewed journal over something that appears on somebody's blog
And again, physics may be different from what I'm used to
 
@bolbteppa Again, you can't take everything Lubos says at face value. He is not The Physics Committee, he is a guy with his own opinion.
 
@KyleKanos Yeah, it would. But I'm not even sure what they're claiming. I mean, it's not as if every geodesic has a particle travelling along it, so what exactly do they mean when they say "let's make the geodesic a Bohmian trajectory"?!
 
You shouldn't, you should study the foundations of Bohmian mechanics and check for yourself, use the critiques to guide it
 
I'd say that, no matter the state of Bohmian mechanics, these papers don't actually make a lot of sense
 
9:53 PM
So are they then saying that the Nobel prize was mistakenly awarded a few years back?
 
@KyleKanos Yeah, probably. Their quantum corrections to GR eliminate the singularities
 
Because the award was given to Perlmutter, Reiss & Schmidt for showing the accelerated expansion of the universe, which kinda requires the Big Bang
 
Due to the rather trivial fact that Bohmian trajectories do not cross, and hence cannot converge
 
The only reason I found that paper was from a news-media report about "Physicists prove the big bang didn't happen" and I have zero faith that it was an accurate representation of things. But I don't know enough to evaluate if their results are found under really restrictive assumptions or something
 
It's really a neat trick, if you don't watch closely - they replace geodesics with something that cannot converge into a singularity, and then celebrate that they have found that their "quantum geodesics" don't converge into singularities!
 
9:57 PM
Is that a trick though? If you can show your singularities-are-impossible formulation predicts everything the other one does with singularities, then isn't that proof that the singularities aren't valid?
Again, don't know enough to say whether that's what the deal is here
 
@tpg2114 Yeah, except they don't do that - they wave their hands saying "Bohmian mechanics is valid QM" and just add the "quantum correction" to the equation.
I see literally no derivation of their result in the papers
They just write it down as if the validity of Bohmian mechanics makes it somehow obvious that geodesic must be replaced by Bohmian mechanics
If quantum gravity were that easy, I kinda think someone else would have done it by now
 
@ACuriousMind Maybe, maybe not. Sometimes things are not so obvious beforehand, but then you go "Oh duh" once discovered
Although I will say that nothing about QM or relativity is at all an "Oh duh!" thing. Nothing remotely intuitive about either
 
The paper gets stranger every time I read it. Their claim is not actually that the singularities of Hawking-Penrose aren't there, they say that don't matter because particles follow quantum trajectories, not geodesics
Which requires anyone to take this seriously to put a lot of faith in the ontological meaning of Bohmian mechanics, and not just in the math
 
10:12 PM
I'm not going on Lubos' random opinion, he gives very clear arguments, for example his point about the failure of Bohmian mechanics to quantize variables, he shows how they basically need Bohr's ad hoc quantization of angular momentum rules or something equivalent to it as an assumption, that's kind of a major issue, ALSO look what they say about spin, they basically call it a fake variable, you guys want to turn this into an opinion thing :(
That's a really awful thing to say that you should take what he says with a grain of salt :(
 
It is not a matter of opinion that Bohmian mechanics reproduces QM predictions. It is designed to do that. Lubos is so hostile to it because the Bohmians essentially just know what they want to reproduce and fiddle with their theory until it fits.
Occam's razor needs to become Occam's chainsaw when you take it to Bohmian mechanics
2
 
@bolbteppa One should take everything everybody says with a grain of salt until it is independently verified multiple times
It's not a personal thing about anybody. It's the foundation of science.
 
There's a lot of special pleading, and no conceptual simplicity like in QM. It's a bad theory, aesthetically. But it is a theory.
Even Lubos never catches the Bohmians in making wrong predictions - because whenever he would, they just summon a new postulate out of nowhere that allows them to predict the correct thing
The best argument against Bohmian mechanics is that QFT doesn't need it. Ever.
 
@ACuriousMind you said it was "a valid, equivalent formulation of quantum mechanics" a minute ago.
 
10:29 PM
@bolbteppa I do not retract that statement. And I don't think I've contradicted myself anywhere.
 
That is the antithesis of science what you just claimed, that they can just summon new postulates whenever their theory runs into trouble, that is literally the antithesis of science :( He very clearly and explicitly showed how it doesn't make sense because of the spin issue, let alone the relativity issue or the quantization issue. This is something that fundamentally lies about humongous concepts like spin, how can that be "a valid, equivalent formulation of quantum mechanics"?
 
@ACuriousMind It depends on how you define "best". Bohmian mechanics violates Occam's razor worse than many-worlds (which in turn violates it worse than Copenhagen), and that's what you have to make precise for a good argument against it. (Against Bohmian mechanics as an explanation for anything. It is of course a valid theory which shows a system for which the axioms of QM are satisfied, and has a certain value in this sense.)
 
@bolbteppa You see, if you were presented with the finished Bohmian theory, and didn't know that they made up their postulate specifically to match QM, you could not claim that it is "the antithesis of science". It's just a theory, and it doesn't matter why it was designed the way it is (this is also something Lubos doesn't really get). And it is a theory that makes the correct predictions.
 
user54412
@tpg2114 Maybe... But it's also ArXiv's gr-qc cross-posted into astro-ph. In my humble opinion, most gr-qc/astro-ph stuff is pseudoscience. It's filled with a bunch of failed mathematicians pushing around Feynman diagrams until they claim they can explain complex emergent astrophysical phenomena with their pet theories.
 
user54412
On any given day you'll see things like how sneutrinos explain solar convection or how axions predict neutron star cooling curves.
 
10:39 PM
@ACuriousMind that's definitely not true, I never accepted the postulational quantum mechanics in something like Sakurai either, you guys might be happy with believing authority, but I didn't believe it, the only book that convinced me of quantum mechanics was Landau, that's the only book I've found which has an internal logic and consistency to it, it's clear you guys are going down blind alleys not following this path to be quite honest :( I'm so glad I read those crazy books
This is exactly what I was talking about yesterday, if you're not careful about the foundations of your theory you'll end up doing something stupid, yesterday I said you'd probably end up falsely merging QM & GR because of those Einstein-Hilbert tensors, today we have people claiming the big bang never happened because their misinterpretation of QM is based on sneaking classical mechanics into it, it's shocking,
 
@ThomasKlimpel With "best" I mean it shows that Bohmian mechanics has no heuristic. It's predictive power is designed to match normal QM, but QFT naturally arises from trying to quantize field theory the same way normal QM quantizes the theory of particles. Bohmian mechanics does nothing of the sort, it's a dead end.
 
@ACuriousMind something as silly and apparently trivial as the unit dependence of entropy could lead you to something like this if you're not careful, that's not my opinion that's just the logic of physics, by ignoring this it's leading you to consider logically flawed stuff like Bohmian mechanics and classical statistical mechanics as 'theories', I mean that's two clear warnings on two separate days, if you ignore them I can only hope you don't run into trouble
@Qmechanic thanks for the help on the representation theory question
 
@KyleKanos The real question should be why people can't explain their issues against Bohmian mechanics in sufficiently simple terms. And I don't think that you have to invoke QFT for that, even so QFT confirms that Bohmian mechanics doesn't really help.
 
user54412
@bolbteppa Really, you need to step back, relax, and examine your priorities. You've lately accused a number of people in this room of being stupid and/or irrational. And your claim about stat mech amounts to dismissing an enire, well-founded, empirically verified field as useless. You should really consider that maybe you're the one missing something.
 
@bolbteppa Again, you make very bold claims without substantiating them.
@ThomasKlimpel The trouble is that the Bohmians designed their theory so there'd be no simple objections ;)
 
10:46 PM
@ChrisWhite it's not my claim about statistical mechanics it's in Landau and Lifshitz, I didn't use any of those words :)
 
user54412
In particular, you keep insisting that I cannot apply simple statistics to Newtonian billiard balls, that somehow classical Newtonian physics is not self-consistent, and that quantum mechanics is the end-all, be-all of physics, without which we could not build physical theories. That's a bit fringe, to say the least
 
@ACuriousMind Yes, but there are many people who would have sufficient knowledge to appreciate why Bohmian mechanics doesn't really help, without having much knowledge about QFT. My own knowledge about QFT is in the "shut up and calculate" state, which means that I can appreciate it as an effective theory, but couldn't engage in foundational discussions about it.
 
@ChrisWhite that is the most safe/mainstream claim to make that QM is the be-all and end-all of physics, of course it is, classical mechanics is a limiting case of it, when you are using classical statistical physics you are really using quantum mechanical boltzmann statistics, they are the same mathematically obviously, the fermi-bose distributions reduce to Boltzmann statistics when you take the right limits, farside.ph.utexas.edu/teaching/sm1/lectures/node82.html
@ACuriousMind what have I not substantiated? I've given links to every claim I've made that needs justification, I have not made one novel claim lol I'm basically just quoting Landau & Lifshitz most of the time, or Lubos
I spent like a year playing with a form of Bohmian mechanics based off of Schrodinger's first paper, my posts are on stack to prove it lol I'm not making snap judgements, just trying to point out that this is dangerous territory :)
 
@bolbteppa That classical physics is a limit of quantum physics does not mean that it could not exist on its own. I have made this objection repeatedly, and you have evaded it by just reasserting that classical statistics does not make sense because the underlying theory is quantum. And I get the stuff about unit dependance and so on, I just don't follow your idea to throw out the baby with the bath water, as they say.
 
@ACuriousMind There are people who feel uneasy about many-worlds, even so many-worlds reproduces orthodox QM predictions without any special tricks. So one should at least try to "explain" how Bohmian mechanics is even more objectionable than many-worlds (for example it also has to assume that the wave function exists in an ontological sense).
 
10:59 PM
@ThomasKlimpel I, personally, don't care for quantum interpretations at all. They're not even different theories, they are just attempts to imbue the math with different ontologies.
 
I don't understand, you are telling me that on a quantum level entropy exists as a well-defined fundamental physical concept for a gas of Fermi/Bose particles, but (for example) when we assume that the distance between particles is so big that we can treat it as a classical ideal gas, all of a sudden the very meaning of entropy just completely changes, now only entropy differences are fundamental, all of a sudden some variables in our theory acquire a unit dependence?
 
@bolbteppa No. No. No. I require you to forget that the world is quantum. Assume there is no underlying quantum theory from which you could make that argument. Your whole argument is like saying Newtonian gravity is wrong because it doesn't recognize that there is an upper speed limit for observer frames.
Which is to say, of course Newtonian gravity is wrong in its predictions, but it is a consistent theory
 
@ACuriousMind If you don't care at all, then why do you talk about it? That's something I also don't understand about "shut up and calculate". How can people who "shut up" be so vocal? Doesn't this contradict their claim that they have "shup up"?
 
@ThomasKlimpel That's a bit like asking an atheist why he even bothers to deny something that doesn't exist, isn't it?
But, as an answer: I am quite vocal about shutting up because of all the effort wasted in debating empirically irrelevant interpretations and claiming it is physics just infuriates me
 
Not necessarily. An atheist doesn't say "personally I don't care at all". He gives rational arguments why he doesn't believe in a higher being.
 
11:07 PM
@ACuriousMind I know what you're saying, but you can't deny that Newtonian theory is wrong, of course it is, you know that, however we can take a Taylor series expansion of the relativistic action and reproduce the Newtonian Lagrangian. Logically if you just magically plug some electromagnetic forces into newton's F = ma laws, as intro books do, you are mixing a Lorentz invariant theory with a Galilean invariant theory, of course it's wrong...
This simple observation is at least half of the reason why Einstein came up with special relativity in the first place, trivial things like this actually really matter
 
@ChrisWhite I don't consider ArXiv to be peer-reviewed publications, but the Phys Review and Phys Letters are -- that's what I was referring to.
 
@ThomasKlimpel True, bad comparison
 
in Discussion on answer by ACuriousMind: "Reality" of EM waves vs. wavefunction of individual photons - why not treat the wave function as equally "Real"?, 15 hours ago, by Thomas Klimpel
@ACuriousMind I find this answer with its overemphasis on the global phase and the comments talking about $|\psi(x)|^2$ depressing. We can measure $\psi(x)\psi^*(y)$ at least in theory, from which we can reconstruct $\psi(x)$ except for the global phase. I find some consolation in the fact that Mark Mitchison has written an answer which conveys some of the real reasons why the wave function of QM is considered to be less real than EM waves by many physicists.
I'm a bit frustrated by your insufficient explanations. There are simply better reasons then this trivial issue with the global phase.
 
user54412
@tpg2114 Good point. But gr-qc/astro-ph cross-lists leave such a bad taste in my mouth I can't ignore it. As a good Bayesian, I take into account my prior experience with similar articles, and as a result I discount how much I believe the article.
 
user54412
All too often esoteric theoretical physics papers ignore vast amounts of observational and computational astrophysics literature that at least partially solves the problem they claim is unsolved.
 
11:12 PM
@ThomasKlimpel First, I've already said that I don't consider this notion of "reality" particularly coherent. Second, the part of my answer where I say that we also cannot exactly determine |psi(x)|^2 is not about the global phase, but about the fact that you need ensembles to get the wavefunction, while a single test charge is enough for the EM field
 
@ChrisWhite That's why I caveated the "peer-reviewed" comment with "maybe things are different in physics." It's possible that the nature of rapid publications for "peer-reviewed" things in physics means that the review basically consists of "Is this insane/garbage or could it be legit"
 
@ACuriousMind okay if I forget the world is quantum then we could probably have a debate sure, I'd have to think about it, but honestly I'd just try to find any reason to justify what will lead to quantum mechanics lol (trying doesn't mean I wouldn't listen to reason of course hehe )
 
It would be similar to a conference paper in my field for example. Where somebody reads it quickly and says "Does it sound interesting?" and if so, it gets published
 
user54412
@tpg2114 I'm told certain branches of theoretical physics are becoming more like pure math -- peer-reviewed publications matter very little; all that counts is what you have on arxiv and people's general sense of your abilities
 
user54412
There's a hint of this in astro too -- after all, the most exposure any paper of mine gets is when it's posted to arxiv. Very few will follow up on the peer reviewed version, except to see if numbers changed
 
11:14 PM
@ThomasKlimpel I also agree that MarkMitchison raises a good point I hadn't thought of, by the way.
 
user54412
(Not that anyone's reading my papers anyway...)
 
@ChrisWhite Maybe because there's so many players in the aerospace engineering/fluid dynamics world, but we're lucky if we get 1 citation that doesn't come from our own research group
Heck, sometimes we're lucky if we get 1 from our own group...
 
user54412
Well, I hope my exposure is greater than my citation count.
 
user54412
It helps that many astro departments have a habit of gathering in the mornings and reading ArXiv over coffee. I think the field is just small enough for this to be feasible.
 
@ACuriousMind Your point with the "ensemble of equal states" is probably also a good one. However, if one elaborates this point, then one ends up with interpretational questions about normal probablity theory. There are indeed different interpretations of probability theory, and these can be important for knowing how uncritically (or carefully) one can apply results from the theory in different situations.
 
11:31 PM
@ThomasKlimpel I already linked somewhere further upthread an answer by Valter Moretti on the different structures of the probability theories of classical and quantum theories. It's an interesting read.
 
Then I will read it before I digest Qiaochu Yuan's blog post about non-commutative probability, Poisson algebras and QM. I already had the impression before that Valter Moretti has written some interesting answers.
 
vzn
tpg et al saw your ref to the "no big bang" theory. that is now bouncing all over the internet due eg to reddit. eg No Big Bang? Quantum equation predicts universe has no beginning / physorg
 
vzn
11:55 PM
re bohmian mechanics. not an expert. but have read some of the rationale. think that bohm did not intend for it to be definitive. more as a demonstrable alternative that could point to new ways of thinking.
it was a mathematical implementation of his philosophical ideas on implicate vs explicate order.
bohmian mechanics creates an implicate order for the explicate order of QM theory.
suspect that bohm never asserted bohmian mechanics was the only possible implicate order for QM theory. maybe it was more an "existence proof" only to demonstrate that at least one (implicate order) exists.
 
My vague understanding is that this paper used Bohmian mechanics (which assumes a path actually exists, contrary to the Heisenberg uncertainty principle) to derive some explicitly loop quantum gravity equation, the quantum-corrected Raychaudhuri equation, is that right?
 

« first day (1558 days earlier)      last day (1768 days later) »