The h Bar - 2014-10-21

« first day (1447 days earlier) ← previous day next day → last day (3469 days later) »

12:40 AM

Anyone have experience with python multiprocessing (or other SMP packages)? I'm looking for an equivalent to C + OpenMP #pragma omp for, and I'm a bit appalled that it doesn't seem to exist - it looks like you need a whole lot of fussing around with multiprocessing to achieve what you can get with 2-3 short lines in C... (ducking out of chat for now, but will check back in a bit).

3 hours later…

3:36 AM

@Kyle only a little bit. It may not be as difficult as it looks, but really, Python is meant for scripting rather than number-crunching, so there's less incentive to come up with good multiprocessing frameworks.

@noahnu you're on the right track but let me suggest that you calculate how much energy the car has at the beginning, at the end, and at the middle. Then subtract. (Then take a look at your formula for $\Delta W$ and see if you can figure out what's wrong with it.) (P.S. it's \Delta, not \triangle)

@JamalS try contacting the college and see if you can get that requirement waived.

@JamalS On top of David's suggestion, offer to give them a talk on it. That is, give them a chance to judge for themselves if you know what you're talking about.

2 hours later…

5:36 AM

@DavidZ: The research supplement is not mandatory, but I'd like to do it to maximise my chances.

5:56 AM

@JamalS I know, I meant waive the requirement for the letter.

@DavidZ: Oh, I see. I'll contact admissions.

7 hours later…

12:55 PM

1

Q: Differential equation of a Green function for a dielectric with a charge

adding Suppose there are two infinite planes, one in $z=a$ and the other in $z=b$, with $a<b$. Between the planes, there is a dielectric medium with constant $\epsilon_1$. The differential equation for a Green function is (inside the medium): \begin{equation} \nabla^{2}G=\dfrac{\delta(\boldsymbo...

homework electrostatics boundary-conditions dielectric greens-functions

Think this can be improved?

@ManishEarth Not really, since the question makes no sense, as Green's functions are defined for differential operators, not for physical situations.

1:10 PM

Yeah

2 hours later…

3:17 PM

@ManishEarth IMO it could be improved but not by anyone other than the OP

3:59 PM

It's quiet today....

...................................

So, chat session time. What do we have to discuss today?

Terry Bollinger

4:11 PM

Hi @DavidZ, wow, it is quiet...

Yeah, it seems that nobody has anything to say

Terry Bollinger

Anyone else even here? I'm on my Android, it doesn't show the list.

I see about 10 people's avatars

but apparently they're not paying attention to the room

So.. no topic??

Terry Bollinger

Hey, here's a question I've been getting asked lately: Is it reasonable that the phenomenon of entanglement should always be viewed as embedded In 4D spacetime?

4:17 PM

Not really. The last big thing we had was the discussion about the homework tag, and whether to rename it, but that's been quiet for a few weeks. Of course we could always continue discussing it if anyone had some input.

@TerryBollinger huh, well some people have tried to compare it to wormholes in recent work. So that suggests it's not necessary, although I would still say it's reasonable.

Terry Bollinger

I like renaming homework...

Oddly, I think that that a "just the facts" embedding makes temporal aspects of entanglement less like wormholes, not more.

That seems plausible, though I haven't read the research so I can't really comment on it

Terry Bollinger

It just becomes classic quantum atemporal behavioral.

I didn't realize there was a new wormhole interpretation.

@TerryBollinger Why would you say such a thing? Entanglement is a direct consequence of the fact that the cartesian product embeds into the tensor product, but that this embedding is not surjective, i.e. that there are states in the product space that do not come from a tupel of one-particle states

user54412

@TerryBollinger what do you mean by embedding in spacetime?

4:26 PM

I don't see why we need to talk about spacetime at all

user54412

@ACuriousMind ditto

@TerryBollinger I think this is the paper I was thinking of

Terry Bollinger

@CuriousMind, how can you not? Detection of entanglement always requires at least 2 classical events separated by an interval.

@DavidZ, thanks, I was completely unaware of this wormhole thing.

@DavidZ Hi guys!!

@TerryBollinger :):)

hi

4:34 PM

I'm in school till about 6:30 every day nowadays

it's brutal

PM that is ;)

Terry Bollinger

Talk of that may have driven interest in some old answers I did...

Hi @Danu, tough schedule, wow.

@ACuriousMind did you read the question I posed earlier in chat?

@TerryBollinger I'm not in class 24/7 but I am unable to get the work done unless I hang around everyday till that late

...and I'm not smart enough to do the work completely alone

except for QFT and GR courses, those I can handle just fine

@TerryBollinger Yeah, detecting entanglement needs at least two measurements...but for its existence, I don't see how that is relevant, so saying it "should always be viewed" in spacetime is a far too strong statement, imo.

Terry Bollinger

(switching to a real computer...)

@Danu I'm not sure...do you mean the phermi problem?

4:38 PM

@ACuriousMind No, about domain issues of momentum operators

@Danu Ah, I see it

You know this paper?

@ACuriousMind many people are trying to relate entanglement to quantum gravity, that's why

user54412

@TerryBollinger I agree that usually we present the two entangled observations as spacelike separated, cementing the aura of weirdness everyone wants to assign to entanglement, but surely we can have timelike separated measurements where we're reasonably certain there wasn't some physical influence propagating from the first measurement to the second.

@ACuriousMind Is it relevant?

user54412

And really, I don't see anything wrong with colocating the two measurements at the same point in spacetime

4:41 PM

@ChrisWhite then it's just 1 measurement

@Danu If I recall correctly, it discusses these different momentum operators in one of the sections

user54412

@Danu how so? If my Hilbert space of interest consists of more than just spacetime locations, or even makes no mention of spacetime, what does the location of my measurement matter?

@Danu Hm, it has a bit on the momentum not being self-adjoint on page 11, but it doesn't seem to directly adress your question. (Nevertheless, it is a valuable read, though you might already know most of the stuff in it by now)

@ChrisWhite Because measurement presupposes some kind of existence [in a spacetime]?

@Danu You could have two different particles at the same point in spacetime and measure their properties simultaneously

Terry Bollinger

4:47 PM

Back again!

user54412

@Danu I would say the most we should require is existence in some Hilbert space. Surely I could formulate the quantum mechanics of a universe with nothing but spins and polarizations, with no locations?

@ACuriousMind Thats still one measurement, just a measurement on 2 particles

@ChrisWhite I don't agree that this should be the only requirement to have a theory of measurement

@Danu If you like, call it one measurement, but then we can measure the whole of an entangled state with one measurement ;)

Because all we need to do for that is to gather information about the state of the entangled particles - their separation doesn't really matter

@ACuriousMind I'm not sure what you mean by this. Of course, we can measure everything that is located at exactly the same point in spacetime at the same time(?)

Terry Bollinger

BTW, before I forget it: To me, any "wormhole" interpretation, iff it means GR, is just plain silly for entanglement. I agree with Dirac's argument that highly curved space is incompatible with long-range quantum effects, including by implication entanglement. (I've not looked at that paper yet.)

In short: Entanglement is a phenomenon for which very flat space is a prerequisite.

4:51 PM

@Danu Yeah, so (somehow) entangle the spin of two bosons, and shoot them thus that they cross paths at some point. Measure their spin. You will find the same perfect correlation as for every other measurement of entanglement

Terry Bollinger

@ACuriousMind, I'm not quite following you either...

@ACuriousMind Sure, but tyou're still measuring them. IN SPACETIME.

...yes, alright, but then that's a trivial argument that everything has to be thought of as in spacetime

Because all measurements are performed in spacetime

user54412

Don't worry, @ACuriousMind, I understand you. Then again, I haven't slept in... a couple days?... and I'm not really thinking straight... which may be the best way to think about QM anyway I suppose

So, I guess, either everything has to be thought of as embedded in spacetime, or this is not an argument that entanglement has to be thought of as in spacetime

Terry Bollinger

4:54 PM

The classic real measurements of entanglement always involve at least two interval separated points. I understand that you can always remap, but after a certain point you cease to be connected with experimental outcomes. Since in the case of entanglement those outcomes can no longer be explained by conventional causality, that form of representation remains to me very interesting.

@Chris, I think you may have uncovered the secret to understanding QM!

I think an important to make is that (hopefully we can all agree on this?) measurement is fundamentally classical

user54412

@Danu what do you mean by "classical"? :P

@TerryBollinger Ah, but this is not what the defining property of entanglement is! It is very true that, to see that local causal realism can't explain entanglement, we must have spacelike separation of the measurement events. Yet the state was no less entangled if the measurements were timelike separated, or if no measurement was made at all.

Non-Quantum

@ACuriousMind Watch out with ontological statements like this! ;)

Terry Bollinger

@Danu, yes! And when those fully classical points include relationships that do not abide by conventional causality, it's hard to get around the idea that "something" is linking them.

user54412

4:57 PM

non-quantum does not necessarily imply spacetime, relativity, or even Newtonian space/time

user54412

at least to me

@ChrisWhite I said no such thing! :) But I do think that it does

We all know that spacetime works 'is real' in classical regimes, I'd even call it fundamental

To me, entanglement is really just the observation that the tensor product is bigger than the cartesian product, and there's only something really interesting about it if you believe the world ought to be local, causal and realist.

Terry Bollinger

Hi @dmckee!

@ACuriousMind I think a statement like that is lacking in the physics aspect, although it is mathematically complete

user54412

4:59 PM

@Danu My views are probably shaped by the fact that, as someone who studies GR, I get defensive when I hear quantum people refer to GR as "classical," as though GR and QM aren't equally revolutionary ideas that happened at the exact same time.

@ChrisWhite haha, oh god. This was not at all the point (in fact I think I might be more interested in GR than anything else!)

Terry Bollinger

@ACuriousMind, I have no objection to the math at all! Just don't let the math erase interest in the observables, which are the only thing (thanks to Bell) that keep this from being an entirely metaphysical (as in Einstein-Bohr) conversation.

@ChrisWhite I always thought it had simply become conventional to use "classical" as the opposite of "quantum" and "Newtonian" as the opposite of "relativistic" (either kind)

4

user54412

@Danu I would claim physical completion of that idea simply requires the whole "a physical observable is the eigenvalue of a hermitian operator" postulate/definition

@DavidZ This was my impression as well.

5:01 PM

@Danu But what is lacking? I claim that QM with its Hilbert spaces and the tensor product as the natural operation of combining states of spaces correctly predicts measurement results. That is validated by experiment. The whole talk of spacelike separation is only necessary if I want to exclude that there is a classical theory doing the same.

Terry Bollinger

@ChrisWhite, have you ever read Dirac's "Lectures on Quantum Mechanics"? He really beats hard on the incompatibility of curves space and QM. A fascinating read.

user54412

@TerryBollinger I'll add that to my reading list :)

@ACuriousMind Like I said, it's mathematically complete. No argument there.

(NB chat session time is over... until the next one in two weeks)

Terry Bollinger

Uh... are @Chris and @ChrisWhite the same person here?

5:02 PM

(carry on)

Terry Bollinger

@DavidZ "Loooove is coming to us all" (sorry, too old)

user54412

@TerryBollinger Are you referring to names on the left side of the screen? They get truncated when you post short messages.

@TerryBollinger You only need to type the first three characters of a name to ping someone. Unless there's more than one person whose name matches those characters, then you only need enough to uniquely identify them. In case that's what you meant.

user54412

...

But not long ones.

Terry Bollinger

Yes, mine is a mix of both on the left, so it was a bit confusing.

I have a visitor, must sign off. This was great fun!

5:04 PM

See you later @Terry

@TerryBollinger Do you thus believe that things like QFT in curved spacetime are fundamentally flawed, or that Dirac wasn't completely right?

@ACuriousMind big difference between QFT and QM, no?

Terry Bollinger

More or less yes. I think everyone missed a huge opportunity by not taking that book by Dirac seriously enough. Dirac is never one to skip lightly over. Must go!

@Danu Yes, and Dirac couldn't have known, so I'm interested whether his arguments are particular to QM, or if they pertain to a more fundamental "quantum" aspect of theories

@TerryBollinger Bye! :)

@TerryBollinger Really? You don't believe in things like Hawking radiation, etc?

5:09 PM

@Danu But a mathematically complete description and correct predictions - what more is there to a theory?

@ACuriousMind Understanding its context, applicability etc

I'm afraid I can't make it very precise, sorry.

I've been having many conversations like this recently - I'm the most 'physics-oriented' physicists in my program, with many people holding your point of view or even more extreme things :)

user54412

Mach would argue context doesn't rely on spacetime

@ChrisWhite Let's not get Mach involved here ;)

user54412

actually I find it interesting that the ontological status/necessity of space and time was debated even between Newton and Leibniz - it's a very old thing to muse about

@Danu Haha...it seems our roles in such discussions are quite alike - here, I'm usually in the minority ;)

5:14 PM

@ACuriousMind Ah, you're in the wrong program my man (woman?). I honestly think you'd thrive here at LMU

I think that - although mathematics is very important [insert famous quote about cleaning windows] and all that - the way to really make progress in physics is by doing physics, and no that's not a trivial statement in my environment, haha

It depends on the notion of progress ;)

And what you mean by "doing physics" :D Surely the people there do not deny the value of experiments?

That's also true

@Danu I think I'd be fine there, but sometimes it's not the best environment where most people agree with you ;)

I like my ivory tower with a window open

Or you like it to be lonely at the top (of rigor) ;)

Anyways, regarding people in my program

I feel like many would rather be doing physical mathematics than mathematical physics sometimes :)

The line between the two seems quite blurry to me, even from afar

I mean, I know math professors that do more explicit physics, e.g. simulations or other kinds of modelling, than some of the theoreticians

5:30 PM

It's quite clear to me which of the two I prefer ;)

1 hour later…

6:44 PM

@DavidZ Oh I don't think it's particularly difficult to use multiprocessing, but farming out independent loop iterations to different processors seems like something that should be packaged in a one-line command. And sure, pure python is meant for scripting, but numpy crunches numbers pretty well... Anyway there seem to be solutions, just not brainless ones unfortunately.

7:04 PM

@Kyle because NumPy is written in C ;-) But anyway, yeah, my point was that e.g. there's no one-line command for farming out loop iterations because there's very little demand for it. People who want that sort of convenience don't use Python for it.

7:47 PM

Does any one know if in a given star system, moons and planets will all rotate in the same direction whether that be a planet around the sun or a moon around a planet?

im aware our planets move in same direction around our sun, but do all the moons? And is there a scientific reason that they all follow the same rotation

@Dave Given that we believe that solar systems from out of a gaseous rotating cloud, it is highly unlikely that you'll find planets orbiting in different orientations.

8:36 PM

@joshphysics: Joshua and Sandra Bullock! :)

@joshphysics: I'll get working on it.

@ACuriousMind and what about moons around the planets ?

does the same logic apply ?

@Dave I'm not certain, since I think moons can also be captured, and astronomy is really not my field. Better wait for someone who knows what they're talking about there ;)

9:05 PM

@JamalS ;) Gotta keep things interesting.

Good because it would be nice to have independent verification of my work :)

10:02 PM

@ACuriousMind check this one... math.stackexchange.com/questions/74347/… :)))

@Phonon Lol :D

Also, that question exemplifies everything that is wrong with math.SE :/

@ACuriousMind my 1st impression as well:)

Well, it'd be wrong for us, but that doesn't necessarily mean wrong, period.

@DavidZ I didn't mean they're "objectively" wrong in allowing such questions. But this is the reason I visit there rarely, and do not contribute.

1 hour later…

11:14 PM

Is space (vacuum) static or does it expand or contract?

5

Q: What is the meaning of the "expansion of space"?

When we say that "the space between galaxies is expanding," what do we really mean? For instance, if I think of space as being a Cartesian grid, then when space expands should I think of it as adding more grid-points or as making the distance between the grid-points larger? Or is this a flawed pi...

general-relativity cosmology space-expansion cosmological-inflation

@user11937 There's much more of this if you search our SE for "space expansion"

@ACuriousMind All the angular momentum vectors tend to point in the same direction because like you said, everything formed out of the same disc, which had an overall angular momentum vector. You can get captured moons, but I think retrograde orbits like that are pretty unstable, especially if there are many other moons to interact with, so they don't tend to stick around.

ho ok

oh ok

(@Dave) There are some exceptions, like Uranus is pretty weird in that its spin angular momentum vector is almost 90 degrees offset from its orbital angular momentum vector

« first day (1447 days earlier) ← previous day next day → last day (3469 days later) »

Transcript for

Oct '1421

The h Bar

General chat for Physics SE (physics.stackexchange.com). For M...

join this room
about this room

00:00

06:00

12:00

18:00

all times are UTC

site design / logo © 2024 Stack Exchange Inc; legal

mobile