data.stackexchange.com/stackoverflow/query/174101/…

Site by site maybe.

I got captcha'd though.

@TheDarkSide there's ways to do all-sites queries

look on MSE or on my favorite queries

I wonder if I am actually a robot.

@EmilioPisanty There you are. :P

@TheDarkSide happens to all non-logged-in users. Login to SEDE is separate to SE.

@EmilioPisanty Yes. I know. But I am referring to the fact that I failed that captcha. :(

Hey guys, I think my quantum question is way too small to be a PSE. Perhaps I should ask here first:

Consider a Sagnac interferometer as shown:

The new layout is .. ehh.

This one's better.

data.stackexchange.com/physics/query/762269/shortest-answers

hmmm

@TheDarkSide You got the right "Leo". Sorry I was wrong. I haven't had my copy for long enough that i may be conflating stuff we did in class with stuff actually in the book.

where $D$ is a detector, $S$ is an electron beam source and M is a retractable half silver mirror which can be set to retract and pop up at a frequency of $f$

@dmckee Ah. I see. Thanks nevertheless. But wait, did you teach Bethe-Bloch at any time in some class?

@Blue quite possibly.

The interferometer is structured so that when $M$ is up, destructive interference occurs at the detector and thus there will be no click, and otherwise there will be a click when $M$ is down

run it against SO first.

@TheDarkSide No. I haven't looked at the derivation since grad school. Sorry.

if it runs on SO then the likelihood that it doesn't work network-wide is small.

basically, SO is something like thirty to fifty percent of the database, or something like that. (I don't have hard numbers, obviously.)

The electron beam source also fires single electrons into the interferometer which if it is treated as a classical source, then the electron travels with some velocity $v << c$

@dmckee Ah. Alright. So, it was as a student. But there was a derivation, still. Good to hear.

I am trying to model whether there will be a delay in the disappearance of the clicks when $f$ decreases beyond a threshold, but I am not sure how to incorporate that into the state description of the electron

@Blue We have a winner on Physics.SE:

huehuehue

@TheDarkSide There is some discussion of the added terms and meaning of the equation in the Particle Data book chapter on the passage of particles through matter, but they also don't have a derivation.

It seems to be a hard thing to find.

@dmckee That's really interesting !!

@TheDarkSide oh wow, that has an absolute stinker of a deleted answer.

still

it's maybe not a great idea to go rooting around in there

that query is just going to turn up a bunch of low-quality cruft which is probably best left alone.

@EmilioPisanty deleted? Oh, no ...

What I have started?

Sigh.

@TheDarkSide a separate answer. deleted as spam or offensive Jun 18 '17 at 2:30 by Community♦

@EmilioPisanty Ah. :: Starts breathing again ::

@TheDarkSide still. I wouldn't be surprised if the answer you linked to also went the way of the dodo.

my package from USPS was delayed :(

I was looking forward to getting it yesterday

whomp, I know what's going on with the query

not sure how to fix...

They've decided to delay it until you fix your query

It's the condition WHERE X=Y

when Y is Null it returns nothing

ever

@EmilioPisanty :: washes blood stains off hand ::

I will drop this issue right here.

so when I did WHERE X=Y it only returns 11,000 rows, but WHERE X != Y 2 million rows get removed where Y was NULL

how fun

Ahh those silly null values

yeah since X isn't equal to a NULL

but nor is it not equal to a NULL

...

Anyways, that great mystery is solved

I actually think NULL isn't even equal to NULL in SQL

I believe you have to do an IS NULL

yeah

you have to handle NULLs separately

using IS or IS NOT

Ok cunning nature, so your resolution is to blur that out into a probability

Fine, I concede defeat and just assume the wavefunction updates instantly

I knew that NULLs acted this way, but it just didn't click yesterday that that was what was happening since the Query was a bit more convoluted

So...someone who joined my team like a month ago is already gone...that was some super fast turn over...

And here I am worrying that I may want to leave a job a year after I get it lol

There seem to be a fair number of people willing to pay people to do their assignments...I wonder what they plan to do once they get an actual job

dunno lol

@enumaris were they doing data science? wasnt laid off?

@Secret some larger scale experiments with mechanical vibrations have been showing up recently heres some summaries phys.org/news/2018-04-entanglement-near-macroscopic.html gizmodo.com/…

Yeah I suppose that's true too. I wouldn't have had food if I paid someone to do my work

@vzn no that's the wrong finding. What I m looking for is not entanglement related, but something that can probe how fast the wavefunction "travels" whenthe initial condition changes

@vzn not data science, it was a PM

@Secret this also turned up recently & seems similar to your question phys.org/news/…

Yeah I have an account on codementor and thought it may be a source of side income, but it seems not really worth my time just to sort through all the requests for help

@enumaris startups are sometimes unstable. or sometimes new depts. how big is your company?

It probably doesn't really help that I refuse to do anyone's assignment for them

lol oops! secret found that link :P

@vzn not a startup heh

About a Billion in annual revenue

I think

2 Billion in annual revenue apparently

I think the PM wasn't fitting in well or something like that

@Blue Well codementor is supposed to be coders helping other coders overcome roadblocks (at least I think). But like 90% of posts are homework problems. It is kind of funny that some of them are "do my assignment for me" and others act like it's some professional problem when it's obviously an exercise for a class

It's a scary world when you start diving into mailing lists trying to find a solution to your problem

It's fairly rare that I have to do that, but it does happen from time to time. It seems to happen more if I'm doing stuff with microcontrollers

tfw you're the only person in the world who can solve the particular issue you're facing

But back then you also didn't have a dozen choices for every part of your project. You had to either write it yourself or use one of a couple options. These days I'm seeing a handful of projects that were good but have either went out of business or been abandoned

heh

Usually when I have that feeling, the problem is me being stupid

Though a lot of other people do indeed work on problems that only they can solve

I'm the only one who can solve my hunger

I dunno. Sometimes I need help with that

Maybe if I go to the fridge enough, there will be something new in there

Only I know what I want for lunch tho

so I'm the only one who can solve it

Knowing what you want for lunch takes some serious decision-making skills

I'm good at it :)

I would go with ice tea

I'm good at making decisions

@TheDarkSide ¯\ _(ツ)_/¯

it's just that if you deliberately go digging for answers selected using a criterion that correlates with low quality, and post them on a public site chatroom...

@enumaris tfw you're the only person in the world who can solve your previous issue, but the problem that comes up out of that solution completely stumps you

also: how come wtf, tfw and ftw get meanings, but wft, twf and fwt don't?

@EmilioPisanty twf = two-weapon fighting.

see Sir Cumference's post for daily experience --->

@ACuriousMind that's nowhere near the caliber of first three

@EmilioPisanty (Not kidding, it's a common acronym in RPG discussion forums :P)

what the fck, the fck what, fck the what, what fck the, the what fck, fck what the

fwt = fighting with tweezers

wft = whales for toddlers

@ACuriousMind that's... a worrying indication of your choices in life.

shots fired?

I am Ron Burgandy?

I am Ron Burgandy?

(Reading the answers) hmm... so counterfactual communication is a lot less intuitive than I initially thought due to my misconception on how wavefunctions behave. Will restudy the basics again in more detail before trying to comprehend it. For now, the answer have addressed the main questions, will accept that first and also to edit the question to address jacob's concerns (and his guess is correct as we will see)

@ACuriousMind I'll race you to the rep-cap

I think it is safe to say from now on I will stop thinking about wavefunctions as some stuff that occupies space, the answer blows this misconception up completely

@EmilioPisanty Do you ever say that unless you're only one vote away from hitting it? :P

@ACuriousMind I do

I might not have said it yet in other circumstances though

@Secret as is ever the case, there's a Bohmian perspective on this: arxiv.org/pdf/1509.00767.pdf

(the interesting thing about that article being how it points to the Bohmian perspective being more complicated than you'd expect, not less)

@Semiclassical Who'd ever expect the Bohmian perspective to be less complicated? :P

hah

@Semiclassical I will definitely read about this tomorrow, but for now I need to get ready to go to bed

well, you might at least expect the Bohmian perspective to be simpler when it comes to position measurements, given that it takes position values as being primitive/noncontextual in a way that spin measurements for instance aren't

@EmilioPisanty do you own a copy of Optics by Born and Wolf?

Well, I did somewhat influenced by (an incorrect understanding of) Bohmian thinking that changing the experimental setup will suddenly update the pilot wave throughout the whole experimental setup and hence causing any electrons that should be in one of the pure state to be "detoured" into another pure state

amusingly, in discussion at the end they bring up the possibility of testing the speed at which 'quantum nonlocal correlations' propagate. in bohmian mechanics as such, the correleations propagate instantaneously

and they've actually tested that, with the conclusion being: "The results put lower bounds on this hypothetical faster-than-light-but-finite speed influence, something like 10’000 to 100’000 times the speed of light."

Soooooo yeah

Not a lot of reason to think that that's a workable approach

I think for me, the hardest classical intuition for me to get rid of is not nonlocality, not contextuality (because postmodernism has similar features to this, except applied to social contexts), not nonlocal correlations, not superposition, not (at least the basic concept of) counterfactuality, but the following:

> The notion that there has to be something (even if just a field) that occupies space

Because this underlies the very foundation of a visual learner

This is why months ago I spent so much time trying to comprehend spacelessness, because my thinking is limited by the tendency for me to think there is always stuff (even if there is nothingness) that occupies space

I think at least for non Bohmian interpretations, I need to give up the assumption that the wavefunction actually occupies space

Because that's how I initially interpret what counterfactual communication is: That the channel is filled by the wavefunction in a destructive interference, hence no particles can show up

The tricky thing for me is that I genuinely believe, within the non-relativistic domain, that the Bohmian account is internally and empirically consistent

So within that domain, the picture you get is trustworthy

But is that a picture you should try hold onto when you move beyond the non-relativistic domain?

I'm a lot less sanguine on that.

Well, the issue is when you go fully relativistic and fully QFT, not only space and time get meshed into spacetime, but that you cannot really say "there's a quantum field here, there's a quantum field there, and how much", but can only say. "In this given region, the probability of something X happens is Y"

can you even say "in this given region" :P

(I also have in mind multi-particle phenomena. A Bohmian account of a single Dirac electron is actually doable without much trouble. But the moment you have more than one, you have the possibility for nonlocal correlations.)

@enumaris well, I suppose one can always say "somewhere in this universe, X happens with a probability of Y"

because by definition, everything is inside the universe

:D

But yeah, I think being able to wrap one's mind around "spacelessness" is crucial to understand all things quantum fields

which is very challenging for a visual learner since we organise everything in some abstract mental space

@user2236 Fasting for a Jewish holiday :/ Currently been 20 hours without food/water

though only got 5 more to go

@Semiclassical well, that is also true for non bohmian interpretations as well, e.g. bose einstein condensates does a lot of "collective things"

oh yeah, the human body can only go 3 days without water

going 1 day is hard ;-;

thanks tho

you're a bad jew i guess? lol

@Secret sure. my point is that a Bohmian account of a single relativistic particle isn't so hard to do

right

it's when you have more than one particle that a relativistic Bohmian account becomes problematic

@Blue i've already gone 4/5 the way there anyway

the biggest problem is that i'm not supposed to be working, which is worrying since i'd like to be studying for physics

@coniferous_smellerULPBG-W8ZgjR I don't.

damn it i feel bad coz i bought it and it was expensive ^^

but if an optic physicist doesnt own it i feel like cheated

well yeah but i had heard it was like the jackson of such books

do you know any grad stud that didnt study from kjackson

i dont know my guess is that he solved that book

@Blue I've indeed never read a page of Jackson :P

ah hmm wow

but you had sommerfeld or something

I really haven't read many textbooks - my courses usually didn't follow a single book and had excellent notes.

my em course has its own notes, but they were hard to decipher.

much harder than jackson

the dude mixed up the metric here and there, the English was far from average

and he assumed we knew SR and differential geometry, for an em course

so it was all tensor notation

my qm courses also followed some special notes. this time it was like advanced abstract algebra

and my mathematical methods for physics was extremely far from physics, it was also a math course to death

tensors in EM are a bit funky lol

like we saw the spectral theorem, the jordan canonical form (where the heck do you use that in physics? come on)

cus generally people don't study Differential geometry until GR

so then when tensors are introduced in EM it's like "tensors are objects which transform this way"

nah my prof followed a bit of wald I think

but he made his own notes

I feel like people should study tensors in EM since they can work from Maxwell's equations without them. With GR you're just sort of out there with tensors and nothing else

So your prof taught you Diff Geo to do EM?

he was speaking about tangent to vectors and stuff i dont remember now

yes

E and B were 4 vectors

heh, that's kind of rare I think..

E and B are not 4-vectors

I had a course where we did SR without really using tensors then jumped into GR and most of the people didn't have a clue what was going on

E and B are components of a 2-form

F

@Blue well i had it for my mathematical methods in Physics . the jordan canonical form, which is useful for operators that are not diagonalizable.... but in QM they are all diagonalizable in some basis

so Boas book was 100% useless

E is the time-like portion of that and B is the space-like portion

yeah we saw the maxwell tensor and electromagnetic tensor Fmunu

we saw all this stuff in EM

it was split on 2 courses, covering a full year

But if your prof left you with an impression that E and B are 4-vectors...I kind of question...the effectiveness of his teaching...

my prof said everything in EM was easy as pie if we understood the light cone

no offense

well... he was extremely bad at teaching, but he really tried hard

i mean we passed from purcell or resnick halliday to his tensor course

Speaking of F, is there a way to classify whether a certain term in the action will lead to linear or nonlinear dynamics? I find it odd to see an action involving $F_{\mu \nu} F^{\mu \nu}$ coming out with linear equations

there was a freaking gap. within 2 weeks we went from 50 to 15

hmm

Like 50 to 15 people in the class?

roughly, yes

maybe 20

wow

cant remember. but at first it was hard to find a place to sit. after 2 weeks it felt empty

I had a handful of classes that dropped a bit, but not that much

@danielunderwood yes, there is a way to tell if a Lagrangian will lead to linear E.O.M. or not, but I kind of forgot exactly the details.

I think it has to do with whether the terms are at most quadratic or not

@danielunderwood $F_{\mu \nu} F^{\mu \nu} = \partial_{\mu} A_{\nu} F^{\mu \nu} - \dots = - A_{\nu} \partial_{\mu} F^{\mu \nu} - \dots$ so linearity will definitely follow

Under $A \in SL(2,C)$ we have $\Psi_{a_1 \dots a_m ; \dot{a}_1 \dots \dot{a}_n}' = A^{b_1}_{~a_1} \dots A^{*\dot{b}_1}_{~\dot{a}_1} \dots \Psi_{b_1 \dots ; \dot{b}_1 \dots }$ being an irrep if $\Psi$ is symmetric in the $a$'s and $\dot{a}$'s, the $D^{(\frac{m}{2},\frac{n}{2})}$ rep,

but under $A \in SU(2) \subset SL(2,C)$ if there is at least one undotted and dotted index then $\Psi$ is not irreducible, and we can now use the product of angular momenta rule - so for a vector $\Psi_{a_1 ; \dot{a}_1}$ we have $D^{(\frac{1}{2},\frac{1}{2})} = D^{1} + D^0$

So we need to use gauge invariance, and two components have negative norm?

Ahhh I didn't think about separating out the potential

That second part wasn't to me, right?

haha if you know it then yes

Basically, what the hell is going on here

I never studied representation theory that well lol

so dunno

Under $SU(2)$ we have that the rep breaks up into a 3 vector and a scalar, okay, but this somehow means we have to use gauge invariance and also something about 2 states with negative norm?

I understood as far as $SL(2, C)$ lol

A 2D linear group over the complex numbers with determinant 1?

Ahh it looks like it is. Yay math for naming some things reasonably

@SirCumference hang in there pal.

@Mithrandir24601 glad to hear it :)

is SL(2,C) 2 dimensional though...

seems like it should be 7 dimensional...? 8 independent components with only 1 condition right?

well @Blue I'd rather learn some Physics than math, for a QM course

like, we didn't cover the H atom at all but we went to some obscure spherical tensors

0 physical intuition, little physics, ton of abstract math. I would have prefered the other way around for sure

$SL(n,C)$ should be dimension $2n^2-1$ or no?

but it turns out it was just the profs. had I studied a few years earlier or later, I would have got other profs

or maybe it's $2n^2-2$...

Basically: if you want reps of the Lorentz group - is not even connected so they are nuts, it has 4 components (+- time component, det +- 1 of a LT), if you consider the positive time component connected to the identity - the restricted Lorentz group - it is not simply connected, so representations of the Lorentz group will act crazy, this is why spinors arise, so we embed the restricted Lorentz group into it's simply connected cover to work with it's representations, this is $SL(2,C)$.

cus the one constraint is actually 2 constraints...?

Complex conjugates form another representation distinct from this original one, i.e. $A^*$ can be used too... We now consider tensors which transform under elements of $SL(2,C)$, and denote indices which transform under $A$ without dots and under $A^*$ with dots to find

$\Psi_{a_1 \dots a_m ; \dot{a}_1 \dots \dot{a}_n}' = A^{b_1}_{~a_1} \dots A^{*\dot{b}_1}_{~~~~\dot{a}_1} \dots \Psi_{b_1 \dots ; \dot{b}_1 \dots }$. As with any group, we can find an invariant tensor, this is $\varepsilon^{\alpha \beta}$, thus $\Psi$ is an irrep when it is completely symmetric in the $a$'s and $\dot{a}$'s separately...

I feel like the group theory taught for particle physics is very lacking...at least at my school

We denote this irrep by $D^{(\frac{m}{2},\frac{n}{2})}$... But if we now take this $A$ to be an $A \in SU(2) \subset SL(2,C)$ it becomes reducible because we can use the old product of angular momenta rule to end up with a direct sum of reps of different spins. This apparently means we have to use gauge invariant vector fields in quantum theory, maybe to eliminate one of the four components of a 4-vector since the rep breaks up into a 3 vector plus a scalar?

Something something about two negative norm states as well

I do know $SL(n,\mathbb{R})$ has dimension $n^2-1$

it's the complex part that's messing me up right now -.-

@enumaris The group theory taught to undergraduates was non-existent!

lol

I'm sure undergrad math has some group theory

(component connected to the identity has det + 1, also SL(2,C) has 6 components matching the 6 of the Lorentz group)

but yeah, no undergrad group theory for physics...

This is hand-wave-group-theory

so it should be dimension $2n^2-2$ then not $2n^2-1$

I do think you're right about the size though. I was thinking $n \times n$ matrix, but each one would have 2 parts due to complex

@enumaris It's of complex dimension $n^2 - 1$, which for $n = 2$ means that it has real dimension 6, which is precisely enough to accomodate the three rotations and three boosts of the Lorentz group.

SL(2,C) has $8 = 2(2)^2 - 2 = 6$ components because 8 parameters but $\det(A) = 1 + 0 i$ gives 6

Yeah I'm sure math students did group theory. We didn't even talk much about using complex numbers for rotations

yeah

@enumaris Not only at yours. If you want to learn proper group theory (or rather, representation theory), you have to learn it from the mathematicians

I forgot the condition $det(A)=0$ has 2 parts as well

And then you can't understand what the physicists call "group theory" anymore because it's a mess :P

Would Zee's book be what physicists call group theory?

I learned group theory mostly by myself

from books

my particle physics class group theory was...super...uh...

not good?

This is from the first few pages of the PCT book

it didn't help that the prof for that class sucked as well

reading up on how to structure a data science/analytics team...some kinda interesting stuff in here...tho mostly it seems like common sense

But then you realize you got into an infinite loop and ran out of money

I want advancement man...don't wanna just be a data monkey forever u know

:P

I would gladly move to be a data monkey

I am currently in the process of reworking applications to better handle "we want something a bit different than we wanted a week ago"

Within a span of a week, both chrome and PSE changed their layouts. Why??

they both hate you that's why

I'm confused.

Easy fix. Just use firefox

@danielunderwood I'd like to be director of data science, and then chief data officer eventually :P

I stopped using firefox a long time ago.

But currently those titles do not exist at my company lol

@enumaris yeah me too :D

I used to use firefox before .. yeah.

@danielunderwood what about chief of reworking applications to better handle "we want something a bit different than we wanted a week ago"?