12:00 AM
Ahh well that's better. I thought they expected a number and I'm not even sure of what the job entails at the moment

indeed...

It's good to get an idea of what other people are getting paid for similar work in the same geographic area so you have a range in mind. That way you at least know if the company is offering to pay much less than normal (or more than normal)

also, you should internally know what you want

I mean internally I want all of their money lol

they don't need to know until you negotiate
lol

12:05 AM
I don't understand the salary negotiation drama (probably due to my lack of experience). Why don't the companies make it transparent according to the position the developer will be appointed to?
Does it usually cause losses on their end? And/or are they actively looking forward to offering as low salaries as they can?

Some do list the range for a certain position. Others not as much and I just don't want to get in the position of saying a number that's way too low
Looking at that workplace SE kind of makes me feel better about my job hunt

2 hours later…
2:24 AM
@Blue Pretty sure that happens a lot, the company wants to offer as low of a salary as they can get away with.
This may be of interest: askamanager.org/2018/10/…

2 hours later…
4:37 AM
Well, that's a new one

5:06 AM
Whenever we derive lorenz transformation equations we take the frames moving in one axis but sometimes the observer can also move diagonally or has velocity in three dimensions. Then shouldnt the relation between y and z coordinates of the observers change?

@EmilioPisanty You make some good points :)
For what it's worth, I often edit clickbait-y titles of HNQ questions on Physics to replace them with more straightforward titles that actually show the question being asked. It would be nice if there were a list or something of hot questions so we could do that more proactively.

5:38 AM
Chat's been pretty slow lately

5:58 AM
@DavidZ I mean, we've been asking for that for years

Yeah, I'm just lamenting out loud

2 hours later…
8:13 AM
0

Recently I've been getting thoughts on why a β$^-$ particle couldn't be used in Rutherford's experiment. What significance would small mass of electron hold to make the experiment inconclusive?

8:54 AM
Lovely. I can't thank SomethingBadHappened enough; I've been waiting for the opportunity to unlock my vote on that one too. I've decided that the twitter-based-development needs downvoting.

9:04 AM
how is the area changing here?

@starunique2016 As the small inner coil rotates the flux though it is given by $BA\sin\omega t$
i.e. it's the orientation of the small coil relative to the flux lines that matters

where from does sin omega t comes?
so its 1 option/.

@starunique2016 The flux though the coil is given by $BA\sin\theta$ where $\theta$ is the angle the coil makes with the magnetic field lines. Yes?

oh....yess...i was thinking cos theta....confusiong with electric fiel.
field*
omega t is the anglular distance covered.

Actually looking at the question they are considering the angle between the normal to the coil and the field lines, so it would be $\cos\theta. Then differentiating to give you the rate of flux change will give you the$\sin$. 9:19 AM ok... after that what to do withe the area. oh...sorry got it. i already go the area///its simply pi r^2 So the flux through the small coil is$B\pi r^2 cos\omega t$and$dB/dt = -B\omega\pi r^2\sin\omega t$9:36 AM @JohnRennie Would you consider removing the answer portion of your comment here? (i.e. delete and repost without it) (or I could edit it out for you) @DavidZ done! thanks 3 hours later… 12:09 PM Aha, a rep cap. Nice :-) Though I can't honestly say any of today's answers were great ones. 12:50 PM If I have a rotational symmetric bound current density and no free current, does that mean$\mathbf{H}$is$0$? Let's say I have a sphere of radius$a$Its magnitization is$\mathbf{M} = M_0 r/a \sin \theta \hat{\phi}$The bound volume current density is$\mathbf{J_b} = \nabla \times \mathbf{M} = 2 M_0/a \hat{z}$The bound surface current density is$\mathbf{K_b} = \mathbf{M}_{\text{surface}} \times \hat{r} = M_0 \sin \theta \hat{\theta}$Claim: The "upwards" going current inside is redirected along the direction of the polar angle on the outside I think the situation is rotational symmetric Then the magnetic field is circumferential Draw an Amperian loop with radius$r \sin \theta$inside the sphere Ampere's law gives$\oint B_{\phi} dl = \mu_0 I_{\text{enc}} \Leftrightarrow B 2 \pi R = \mu_0 \pi R^2 J_b = \mu_0 \pi R^2 2 M_0/a \Rightarrow B = \mu_0 R M_0/a = \mu_0 r \sin \theta M_0/a$Is my reasoning correct? 3 hours later… 3:52 PM wooowee saw a job description for a 6 month contract that requires 6-8 years experience and PhD preferred...lawl... good luck with that buddy lol no thanks So evidently spark is normally on top of hadoop instead of separate. It's just built to handle data in memory rather than on disk I believe Still seems like big data is something made up by marketing. Cloudera has a vm with tutorials that I did last night...it was less than helpful hmmm So Hadoop manages a cluster of Harddrives and Spark manages a cluster of RAM? Well Hadoop has HDFS, which is a distributed filesystem. Then there are a number of computation options on top of that, with two of them being MapReduce and Spark. My understanding is that MapReduce does a bunch of (Read from disk -> Calculate) while spark seems to Read everything into RAM first then a bunch of (Read from RAM -> Calculate) which should typically be faster if you have enough RAM I'm not entirely sure on that, but it's what it sounds like at the moment Also MapReduce is fairly high latency as far as I understand, so it's less suited to working on streaming data But everything is also kind of a zoo in my head at the moment Then you have a bunch of applications to pull data in, query it, or make it look like SQL 4:22 PM everything into RAM... how much RAM u gotta have...I can never manage that You never have enough RAM. You just haven't figured out a use for it yet I have 64gb of ram and it ain't enough I tried to load a 256 sized batch of training data and ran out of ram lol I always seem to either have way more than I need or not nearly enough so I'm pretty much stuck with 96 and 128 as my max batch size oh wait my bad I ran out of vram not ram lol More GPUs! 4:28 PM I ran out of RAM on a different issue unfortunately that project is kinda dead cus my programmer ghosted me :( Speaking of RAM, we had signs all over our lab that said "No RAM allowed here" and I was always confused It took me way too long to realize that it meant radioactive materials lol You should bug your department for a DGX machine what's that? aka "How much can we make a box cost?" 4:32 PM Ah...I can ask IT They have a budget for "deep learning" I dunno if it's enough to buy that tho lol I have no idea how much they cost, but based on this I'm going to go with way too much > FOR A LIMITED TIME ONLY, SAVE$20K OFF

hmmm

can I use it for gaming though?
Get 20,000FPS on COD Black Ops?

I mean they do say it has 4K output!

4:35 PM
It better be give 20,000 FPS on all ultra high settings

Where does the last step come from?

lol their leasing options...\$499 the first 6 months then \$2733 a month
uh oh

This is for a rotational symmetric magnetic field and a circular current loop with velocity $v$ in the $z$-axis
The magnetic field is also static
$a$ is the radius of the loop... I think that's all you need to know

the question is how much support they give? Probably very minimal?

It does say support is covered through the lease period. I have no idea how involved that support is though
"Sorry you'll need to hire one of our consultants for that question"

4:40 PM
probably...
ugh, my python script is stalling at the save to SQL part
I wonder if they reset the servers when it was trying to save it
I guess I should save a pandas dataframe as a back up first

Gotta save at every step just in case everything else breaks

meh, wasted 24 hours of computing
gonna restart this

Data Scientists are really "Waiting Scientists"

"my model is learning" is the new "my code's compiling"

CONTROL-C EVENT
my model is processing data, it ain't learning right now :P

4:47 PM
When I was working a lot with C/C++, I'd always do a make clean just to make sure it had a bit to compile
And because there would occasionally magically be something that broke and was fixed by a clean

@ACuriousMind lolz

5:03 PM
alright guys, i need to come up with 1 sentence definitions for AI, ML, and NLP
lol
wikipedia to the rescue lol

AI is unfortunately not a well-defined term. It may be useful to distinguish between weak AI and strong AI

We use math to predict things

Oh, this one is good:
4

Machine learning in layman terms is an algorithm that allows machines to identify patterns in data and then develop a model which can be used to predict unseen data. Artificial Intelligence is the ability of machines to make intelligent decisions which are equal to or better than their human cou...

5:25 PM
Yep, thanks :D

1 hour later…
6:35 PM
Is this statement a separate thing from being part of a smaller solid angle for an object that's further away?
> The cosmological redshift also slows down the rate of arrival of photons compared to the rate at which they were emitted (called cosmological time-dilation).
Nevermind it explains more right after. Oops

7:31 PM
Hmm Hmmm, I hardly visit IPSE, but this is gold :)
Need some practice in that

Neutral, boring response is my default anyway!

@Blue I've met at least one argumentative person that takes non-committal responses as indications they have not been heard, and so they will just repeat their opinion in a louder voice :P

Yeah, such people do exist too :P
I, on the other hand, have more cases of getting caught up in extended online arguments (well, debates turn in arguments beyond a certain point). I need to snap out of that habit
@ACuriousMind I wonder whether those traits are actually correlated with narcissism or insecurity
Some people just don't rest until they get approval (for their opinions) from someone

I think some people kind of know they're wrong, but can't accept it

7:48 PM
Well, it is difficult

And with the internet, you can pretty much always find someone that agrees with you no matter how ridiculous the thought is

@Blue For online arguments, just walk away.

:)
@ACuriousMind Yeah, I need to practise that :P

Be reasonable and accept their alternative facts /s

good evening
im puzzled by a rather elemental question about quantum mechanics. Lets say we have a particle in the ground state of an harmonic oscillator. at some time t0 we quench the potential suddenly to a different size, which will cause the state to be no longer an eigenstate of the hamiltonian
it will be a superposition of the oscillator states of the new potential. and in fact this superposition will invole even arbitrary high energy states of the new hamiltonian
which means we can find the particle with an extremly high energy.
the probability for that is very small but still it is there so where does the energy come from when we do the measurement?
of i guess it just comes from the source of the potential right?
the reason im asking is because im studying feshbach resonance, which involves the coupling of a scattering state between two atoms to a molecule state in a different spin configuration which has an energy that is larger than the energy of the scattering state
and the coupling causes there to be a probability of finding the atoms in the molecule state even though it has a higher energy then the scattering state
but in this case the energy cant come "from outside" like in the example of the quench

8:12 PM
3300 rep boissss
10% rep increase in 2 days

noicey
HNQ?

@curiosity If the resonant state would violate energy conservation if you could detect it, then you can't detect it, i.e. it is a "virtual state" that only exists internally in the formalism. The outcomes of the scattering, which are what you can detect, will not violate energy conservation.

yep
and now my motivation to post answers has subsided again
lol
openai is asking me to do a coding challenge...

@ACuriousMind but in the hamiltonian of the feshbach resonance there are off diagonal elements which couple the scattering states to the molecule. and if there are these matrix elements, doesnt that mean that after the state is exposed to that hamiltonian it will be in a superposition of the different states which also involves the molecule state

1 hour long math puzzle to solve using python apparently

8:20 PM
like in the case with the quenched oscillator
and if it is in this superposition, it should be possible to collapse it into that state

I might fail tho tbh

@enumaris Sounds like fun, success or no

i mean, in the case of the oscillator you wouldnt say that the particle is in a superposition of only virtual states right? but none of these has the same energy as the ground state of the first oscillator.

@curiosity In the case of the oscillator whose potential changes, the Hamiltonian is not time-independent, therefore the system is not time-translation invariant and there is no reason to expect conservation of energy to hold
I don't know anything about Feshbach resonance, but either the Hamiltonian is time-independent and therefore energy is conserved, or it is time-dependent and energy is not conserved (the system is "open" in some sense in that in interacts with something external).

Maybe the prompt will be like: "In one hour, please write a script that produces a chat bot capable of passing the Turing test."
:O

8:29 PM
Easy, just write obfuscated code that actually just forwards all messages to you
(assuming you pass the Turning test)

You guys can judge whether I pass the Turing test or now :D

hmmm

Unless you guys are robots as well... strokes beard

we are actually treating it time independent. i try to find something on the internet

@enumaris Aha! Inconsistency detected, yesterday you said you can't grow a beard

8:30 PM
maybe im overlooking something but we are solving the stationary SEQ

@ACuriousMind uh oh...
strokes very sparse beard
I wonder what packages I'll be able to use
like if they ask me stuff about strings, can I use the regex package...
hmmm

@enumaris "or now" seems it's not ready yet

lol
joking about a typo is dangerous cus I almost didn't know what you were talking about

I'd imagine the language libraries are available unless they say something

@enumaris Plot twist: They ask you to reimplement the regex package from scratch :P

8:37 PM
(arxiv.org/pdf/0801.2500.pdf) on page 125 it sais $\langle m|V|k\rangle=g_k$ and treats the problem time independently

I think I'd cry. My current assumption is that regex engines are black magic

where $|m\rangle$ is the molecule state and the other the scattering state

this one would be nice to have regex on

the bound state is in the closed channel meaning it has a higher energy then the scattering state which is in the so called open channel

@ACuriousMind lol...if they ask that, I just won't get the job then I guess

8:40 PM
but this collision occurs whithin a ultra dilute gas so maybe the energy would come from the gas?

@curiosity Ah, I was just about to ask if there's something in the environment. Yes, any discrepancy in energy will end up in the environment.

@enumaris Is solving those type of problems part of the job application process?

You should only be concerned about energy conservation if this could also happen in a total vacuum.

@Blue yeah

@ACuriousMind even if it is ultra dilute?

8:41 PM
@curiosity If the ultra-dilute gas is necessary for this to happen, then yes.

Looks reasonable if you have to solve it remotely. On the spot, I'd get some bugs probably

I guess if you have a string like '5???5???5' then it should return False...
@Blue you have 1 hour time limit
It's like a test lol

@danielunderwood On the bright side, regexes are easier to parse than most programming languages.

but i think we are at no point assuming that there have to be other particles around. we are applying the model to the collsions in a gas but the treatment of the collision is only concerned with the two particles

@enumaris I see. That particular one looks more or less simple for 1 hour (just string operations). Maybe in the actual application, they'd give a problem from the Tough or Medium category

8:46 PM
that one is just a random coding puzzle I found online
it's not the one openai will give
I have no idea what kind of puzzles openai will give, I assume it's similar

Good luck anyway! :)

@curiosity If that is the case, then we're back to square one: The transient bound state - the resonance - is a virtual state that you cannot detect.

I think for that particular one it's the "every pair of numbers that add up to 10"

@ACuriousMind i think thats the right answer, but i dont see why. because if the hamiltonian has these matrix elements then the state will go into a superposition that involves that state right?
so why cant we detect that state from the superposition

Scattering theory is a bit counter-intuitive in that the possible existence of states that are not kinematically allowed nevertheless influence the scattering
That molecule state is at no point truly "physically realized" during the scattering, much like a during a scattering in QFT, none of the individual Feynman diagrams depicts what "really happens" during the scattering.

8:51 PM
unfortunately i have not much training in qft
but if we have a particle ( or two particles) in a superposition of a bunch of states, every state is a possible outcome of a measurement isnt it?

@curiosity Yes. Are you in fact concerned that a measurement of any non-stationary state could violate energy conservation?

i think i am yea

If so, see e.g. physics.stackexchange.com/a/358412/50583 - measurement is a process that disturbs the system, and so energy is not conserved because you interacted with the system
Depending on the interpretation you prefer, the explanation for "how" this happens is different

ill take a look at it

Copenhagen would probably say that energy is only conserved statistically (i.e. the expectation values fo energy are conserved, but individual measurements can violate it), interpretations with a universal wavefunction will entangle the measured system with the detector, "storing" the "missing" energy in the detector
Either way, you can't expect the outcome of a single measurement of a system to adhere to energy conservation in the classical sense

9:04 PM
this post has helped but im not sure about my question still. in my case the particle has a definite energy before the collision. then it comes in the vicinity of the other particle and interacts with the hamiltonian so that it is now in a superposition of eigenstates of the free hamiltonian
but ok, if i were to measure the energy after the interaction, it is no longer a stationary state so energy fluctuations are to be expected
but then also we are saying that the particle after the scattering is again in a definite momentum state. that all the collisions are elastic
im sorry if its not clear what im asking now. im simply confused
" The molecular state
causes an effective interaction $V_{eff}$ that corresponds to two atoms colliding and forming a
molecule (matrix element $g_k$), spending some small amount of time (of order $\hbar /(E − \delta)$)

in the molecular state (propagator $\frac{1}{E-\delta +i \eta}$
and exiting again as two unbound atoms " this just doesnt go into my head.

@curiosity ..why would we claim it is in a definite momentum state after the scattering?

i have no idea

Again, I don't know the details of what you're doing, but likely you're just computing a scattering amplitude, i.e. the probability to detect the particle in a certain momentum state after scattering

yes

That is not the same as claiming that it is in a momentum eigenstate without trying to detect it.

9:10 PM
thats why im saying that its in a definite momentum sate afterwards. its elastic scattering
oh i see

Quantum mechanics without measurement is deterministic - the outcome of the scattering before detection is always the same state.

i think im getting it. the whole feshbach resonance business in the end just comes down to give us one value: the scattering lenght "a" of the interaction. if two particles undergo a collision this will cause the state to always become $\psi=e^{i k x}-\frac{a}{1+i ka}e^{ikr}/r$ . this state is actually a superposition of many free eigenstates, also the molecule.
but because we only want to know the scattering amplitude we are asking about the probability of this superposition to collapse into a free state, which would invole our measuring apparatus delivering energy to the system
does that sound right?

Think I solved it
:D
but there might be some fringe cases...
like cases where the string contains a unicode digit that's not 0-9...like a 7 with a circle around it or something like that
and it's only 9 lines long, not too bad in terms of that...
dunno about efficiency and runtime tho lol

but is there any truth in saying that the particle stays for a small amout of time in the molecule state and then leaves it again as a scattering state? the arxiv article claims that.
it even estimates how long this state exists

9:25 PM
"For the input "5??aaaaaaaaaaaaaaaaaaa?5?a??5" your output was incorrect. The correct answer is true."
wtf?
For the input "9???1???9???1???9" your output was incorrect. The correct answer is true.
Also should be False............

@enumaris ...why would it be false?

they say "every pair of numbers"
and "exactly 3 question marks"
Have the function QuestionsMarks(str) take the str string parameter, which will contain single digit numbers, letters, and question marks, and check if there are exactly 3 question marks between every pair of two numbers that add up to 10.
they did not specify the pairs are adjacent
for the first string, obviously the first 5 and the last 5 also add up to 10, and in between there are 6 question marks
similarly for the second string

...that would mean that the answer is always false as soon as either three 5s occur or one pair is repeated.

yes
that was my understanding
That's what the question's english means right..."every pair of 2 numbers that add up to 10"

An artificial intelligence coding puzzle with an ambiguous spec? Impossible.

9:33 PM
"every pair" means....every pair...no? Like every possible pairing of 2 numbers...
unless by pair they only mean when the numbers are not interrupted by other numbers...
booo
I disagree with this question

If we look at the the examples they gave in the original question they mean something else: coderbyte.com/editor/guest:Questions%20Marks:Python
> For example: if str is "arrb6???4xxbl5???eee5" then your program should return true because there are exactly 3 question marks between 6 and 4, and 3 question marks between 5 and 5 at the end of the string.
@enumaris Yes, apparently :P

Well in that example, there isn't any other pairs that add up to 10
cus 6+5 != 10 and 4+5!=10 either

@enumaris True true

there's only the 2 pairs that they specified

Dunno. Maybe the test cases they have are erroneous
Happens sometimes

9:36 PM
So they needed to actually say something like "5???5???5" should return True
This is also a reason I don't like these "online programming quizzes"

@ACuriousMind last question... is it just jabbering if the authors claim that this energetically forbidden state forms for a short time? or might that actually be the case maybe if we were to treat the problem time dependently?

@enumaris .....same :P

In any real use case scenario, I would talk to the person asking for this and spec it out accordingly
but on an online quiz, it's just "you're wrong"
and that's that
meh, w/e
my solution is correct for my specs
it ain't my fault they wrote the question without knowing what the question really means

I guess in the very popular competitive coding platforms the number of erroneous questions is lesser. Usually, several 1000s try them. Topcoder is a good one I heard

9:39 PM
I never heard of Codebyte before

but I dunno how
lol

There's a "Contact" button
But is that meant for reporting questions? :?

meh
w/e
I also don't like the time limit lol
I like to ponder
are they inputing an array or a set of strings...
wait...what is the output format they want...I got everything wrong, but I put the inputs manually - run it - and it's right
T_T
" so for this example your program should return 1-0. "
the...the string "1-0"???
cus I returned a tuple (1,0) and apparently it's wrong
they wanted the string "1-0'
omg I'm gonna fail this test lol
life is pain

10:16 PM
I guess I should just do a ton of these and get used to it

10:34 PM
wow this maze question is marked easy but I'm having a surprisingly hard time with it D:
maybe I shouldn't spend so much time trying to think of some clever way and just brute force it lol

11:03 PM
hey @ACuriousMind do you know if the formula $E=h\nu$ for the energy of a photon is derived/explained anywhere in an introductory book for quantum mechanics? In Griffith's introduction to electrodynamics, $E=pc$ for a photon is properly explained by Einstein's relativity, and if we consider a photon as an electromagnetic wave, I'm fine with all the stuff about Poynting vectors and what not.
However.. when they mention the quantum mechanical point of view, they never really bother to explain how they get their formula for the energy. I've seen other users on stack 'complain' about the same thing, but I haven't found a satisfactory/comprehensible (for me) explanation in the answers.
Would you mind explaining this, or referring me to a book or some other source?

I believe Planck was originally responsible for saying that light came in discrete bits of energy $h \nu$, but thought it was more of a mathematical description than actually being physical. The photoelectric effect suggested that it was physically the case

holy crap, I must be brain dead or something, I can't seem to figure this one out lol

yea, I just found this answer: https://physics.stackexchange.com/questions/410273/how-did-planck-arrive-at-the-idea-that-energy-is-quantized?noredirect=1&lq=1
and I guess there isn't really a 'purely quantum mechanical' derivation that I was looking for (I guess what makes it 'quantum like', is exactly that the energy is quantized, but I'm guessing that doesn't mean that quantum mechanics was used in its derivation)

11:27 PM
giving up lol

I'm not terribly familiar with the history, but I think he was working on blackbody radiation. It was more of a cause of QM than something derived from QM as far as I know. You can look at Planck's/Einstein's papers from that period, but they may only be in German
If you have access, this may have something interesting link.springer.com/article/10.1007%2FBF00327765
@enumaris isn't it just iterate through all elements and check the elements to right, below, and diagonal?
(along with a bounds check)

are you talking about the box of vowels one?
or the path through a maze one?
the box of vowels one was exactly that and it was super simple
the maze one I totally brain died trying to solve it lol

oh yeah I was talking about the vowels. I didn't see the maze one

11:56 PM
my brain ded