« first day (1572 days earlier)      last day (3655 days later) » 

18:00
@KyleKanos So I've got 3 arrays -- p[D], q[D], r[2*D] where D is an integer
@tpg2114 Mostly for fun little projects , nothing in particular . Just want to learn a programming language
r[0], r[1], ..., r[D] = p[0], p[1], ..., p[D]
But then I need to do
@JimdalftheGrey Now try this:
5
Q: Will an object falling into Earth's orbit start spinning?

user2914191Assume an object falls towards Earth (I've drawn a hyperbolic orbit, but this would apply to any orbit). The object starts at $A$, and at this point it is not rotating i.e. an observer on the object would measure no fictitious forces. Will the object rotate as it passes the Earth and moves away? ...

r[D+1] = q[0]-p[0]
@KyleKanos But I don't know anything about it , where should i begin? I was thinking of starting with Learn python the hard way , do you know about it?
18:01
r[D+2] = q[1]-p[1] etc
@GlenTheUdderboat We discussed that question in chat already yesterday
@ACuriousMind And, summary?
@GlenTheUdderboat The answer, thanks to ChrisWhite, is that OP's intuitive picture is wrong because "not turning on a straight line" means that the direction of the object stays the same compared to the four-velocity, not compared to the spatial velocity
@AGoogler Why not try starting with something somewhat straight-forward: Compute & Plot the Planck function.
Ooh looks nice
18:03
@AGoogler Udacity etc. MOOCs also have more "formal" Python classes if you're interested.
I now see that no one posted that answer there, though :P
@KyleKanos So basically , instead of reading tutorials do you think i should do 'projects' like that using google and things?
@ACuriousMind Isn't that at the speed of light??
@AGoogler It's tough for me to give a good reference -- I already knew a whole bunch of languages when I picked up Python, so it was just learning how Python works more than how to program. So I just basically picked a problem I wanted to solve and then tried/searched around until it worked
Which is the approach @KyleKanos is recommending here
@AGoogler Basically yes.
18:04
@ACuriousMind Got it!
@alarge What does MOOC mean?
But I would say that only works if you already know how to program in something else. If it's all brand new to you, you'll want something more structured
@tpg2114 Yeah that's the problem
@AGoogler And after you've used Python for some projects, you really ought to pick up a book and see if you were doing it "right". i.e. to learn best practices. Before this, though, yes, working on small projects is a good idea.
MOOC = Massive Online Open Course
18:05
Well, for someone who doesn't already know the answer, the little "nuclei" in "What is happening is energy is being taken from another source (AKA the nuclei)." isn't going to really be an answer to the question where the energy came from. You should (roughly) explain how/why this energy is stored in the nuclei. Also, the broad statement "Energy conservation holds in quantum mechanics" is misleading, because not all quantum states have a definite energy, so it is not at all clear what it quantumly means to say that energy is conserved. — ACuriousMind 6 mins ago
Like MIT courseware thing.
@AGoogler Well, on the plus side, Python is crazy easy to pick up and make things happen quickly. So it's an awesome language to start with
I know math and bit of logic gates and binary but never learned a language
@ACuriousMind what do you mean with "it is not clear"? Don't we just mean that energy is conserved whenever is defined?
18:06
But I don't really know of anything more formal to actually walk you though how to program regardless of the language
MIT also has a python course
@glance No, we mean that energy conservation holds as an operator law, in particular for all expectation values of energy
For example, in QFT, this means that the classical conservation laws only hold "up to contact terms"
@KyleKanos Um... so this is super embarrassing. But I thought I came up with a method to do what I needed, and grep'd in my code for std::copy cause I know I used it before
The crucial thing is that the former numbers become operators, so you need to define a notion of conservation before you can say they are conserved
And like... seriously, 5 lines above where I was trying to do it
18:07
@AGoogler You mean OCW? Probably. The point of a MOOC is, usually, though that it's more of a course, so you'll do it at the same time as other people (usually there are also deadlines and you get a grade and a certificate at the end).
@ACuriousMind uhm. I see your point of course. Not being in an eigenstate means that there is not a definite energy hence an energy measurement gives a different result at different times
I did the exact same thing before and have it already :(
Hahaha
I hate when that happens
I also figured out the one-liner regex for doing Pn -> P(n) conversion
@alarge Oh never heard about it
And I literally looked at that section of code this morning and went "What the heck is that doing? I don't get it"
18:08
@alarge Where can i enroll?
@ACuriousMind so the only important thing is that the average value of the energy is conserved right?
I forgot to wrap it with parenthesis
@AGoogler I've looked at some courses when I wanted to learn something new, but in my experience the best way is to start with a practical example of some code and start tweaking it. That doesn't make you an expert, but makes you be able to do stuff quickly.
@glance In QM, that's correct, I think
This is what happens when I get right in the middle of refactoring a code and then put it away for 6 months without looking at it :(
18:09
I see
As I said, it gets more difficult in QFT - see "Ward identities"
@AGoogler I gave you links earlier, coursera.org edx.org udacity.com. edX for example is MIT-run. Udacity is more self-paced. I think there's a C# game programming course (for those new to programming) on coursera starting this week if you're interested.
@AGoogler So my best advice, take the course if you want to, but you might be better off with smaller challenges. Like do your next homework assignment in Python instead of Matlab
@ACuriousMind doesn't it always? :)
It probably would also get more difficult if you tried to use the conserved current in QM, but I don't think anyone ever tries that because QM is inherently "zero-dimensional" from a field theoretic viewpoint.
@glance Yes, it does :D
18:11
I don't know what field you are in (or whether you are in school, or working, or whatever) but I'm sure there's all kinds of things you could use it for to get some experience that will also be practical for whatever you do
I think this stuff about conservation laws is my second biggest quantum pet peeve right after people invoking the uncertainty principle for all sorts of things it really isn't meant to mean.
@tpg2114 I think if you can solve your problems elegantly with Matlab, you're only going to learn some syntax if you do it with Python, and none of the good stuff (because NumPy/SciPy cover a lot of the Matlab basics).
@tpg2114 Heh, I don't use matlab yet.
Another suggestion for learning a new language: Try Code Golf & Programming Puzzles, except don't try golfing (shortening) it
I've looked at thaqt before actually
But thanks for reminding me!
It looks cool
18:13
@ACuriousMind Because he keeps deleting his answers, I've actually reviewed crank's "first post" 3 times now
@alarge Which actually makes it a great way to start -- you are learning to speak Python by using what you know (Matlab)
@tpg2114 Well here I'd disagree. For example, I really, really, don't think if you want to learn C++ that you should first learn C.
@alarge That's not what s/he said.
@alarge I'm not saying that at all -- that statement was predicated on already knowing Matlab and using it as a Rosetta-stone type way to translate knowledge into Python
@ACuriousMind do you mean $$ \psi^* \nabla \psi - \psi \nabla \psi^* ?$$ Using it for what?
18:17
Thanks guys , I'll (hopefully) start tommorow
@tpg2114 Fair enough, but I think you would be better served by learning things a different way. For example, Matlab programs are almost entirely devoid of OOP and you won't be accessing some features and syntax of Python present in basically every Python program out there by literal translation.
@ACuriousMind I'd have said it was the notion of quantum fluctuations of the vacuum, and the whole thing of "virtual particles popping in and out of existence"!
I get your argument, though, but I would rather recommend one to learn problems that are more directly suitable for Python than translating line by line from Matlab.
I also didn't advocate direct translations
Just said "Take something you would know how to do in matlab and do it python"
by the way , which projects should i use? i don't think i can cook many ideas myself atleast initially
18:20
@AGoogler Are you in school?
college
i don't know the american equivalent
but i'm curently learning basic calculus and classical mechanics
Start out something simple -- a program to compute your GPA
The beauty of that is you can make it as simple/complex as you want. Maybe you start off by putting all your grades in a list in the script
Where I am , there is no GPA , we use percentages :p
Then you modify your program to read it from a file
@AGoogler It is certainly more motivating if you have a practical issue than just a general wish to learn a language.
18:22
^ What he said
i see
So rather than GPA -- just have something that computes your grade in a course
Or find random physics equations and plot them
So you have to feed it the weighting of all the different assignments (if that exists) and then you pump in your grades and it gives you the average so far
Or compute the quarterback rating
18:24
Or what Kyle said -- depending on how advanced you want to get with your basic mechanics/calculus, try plotting the displacement of a mass-spring-damper with time
Passer rating (also known as quarterback rating, QB rating, or passing efficiency in college football) is a measure of the performance of passers, primarily quarterbacks, in American football and Canadian football. There are two formulae currently in use: one used by both the National Football League (NFL) and Canadian Football League (CFL), and the other used in NCAA football. Passer rating is calculated using a player's passing attempts, completions, yards, touchdowns, and interceptions. Since 1973, passer rating has been the official formula used by the NFL to determine its passing leader. Passer...
The MOOCs often have interesting problems and they are at the right level of difficulty, which is why I'd recommend looking at their material.
Or any undergraduate course that uses Python (if that's the language you want to learn) for that matter.
I've always wanted to know , is undergraduate what comes after school?
Undergraduate is college, comes after high school
@AGoogler "School" is pretty ambiguous -- it really just means "educational institute"
If you want to be really specific, in the US at least
Primary school is little kids
Secondary school is after Primary School until "college"
College is undergraduate (meaning working towards an associates or bachelor degree) and also graduate (meaning working towards a masters or doctorate)
Although some people will call graduate school "post-graduate" also to indicate education after you graduate (the first time)
18:31
how many years long is undergraduate and graduate each?
Undergrad is typically 4
Graduate is typically 2 for a masters, until hell freezes over for a PhD
PhD is usually 4-6 years total depending on where you start. Most PhD programs here don't require a masters first so you can jump right in after undergrad
I've been told it's different in European countries where usually PhD students don't take classes anymore and are assumed to have a masters (or the same education as a masters). There the PhD's are also usually fixed length, like 3 years
At least for the friends I've had in France, Belgium and Spain
but most people don't do PHD and stop at graduate , right?
Most people finish undergraduate and stop
@tpg2114 3 years seems short, even after a masters. Are you sure?
where i live , after school we have 2 years college and then 4 years for engineering degree
18:34
@GlenTheUdderboat It depends on the circumstances
@GlenTheUdderboat The 4 people I've worked with had fixed contracts that funding was cutoff after 3 years
@tpg2114 at least in Italy, we have (usually) a fixed length of 3 years for the PhD, but with some classes/exams to attend for the first year(s)
@AGoogler So a 2 year degree in the US is usually called an Associates degree
so does undergraduate = engineering degree?
@AGoogler If you are in engineering, sure
It's a Bachelor's degree
Which could be in an engineering field
or science, or art, or music
Or literature, or just about anything
18:36
how common are people taking non engineering degress in the US ?
my favourite example is a bachelor's degree in interpretive dance
I don't know the actual numbers, but engineering is usually just one of potentially dozens of departments at a university
@JimdalftheGrey Golf course grass management is my favorite
@AGoogler It sounds like when you are done with an "engineering degree" you have the equivalent of a masters degree. It's hard to say without knowing material covered and stuff for sure
But based on the timing, that's probably what it is equivalent to
@AGoogler I don't know specifically for the US, but I think it would be safe to say a vast majority of degrees in the US are not in engineering
@glance My PhD program, whether you have a masters or not, is 50 hours of classes (about 3 years worth)
@JimdalftheGrey Thats surprising , since in india most students end up as engineers or doctros!
*doctors
18:40
@tpg2114 Wut?
@tpg2114 I took engineering in undergrad. Took 5 years. My bachelor's is not equivalent to a master's :(
So if you look there, you can see total numbers (unfortunately, not relative to other fields)
But if you look at the enrollment in engineering disciplines, over that time frame, is 500,000 or so
Consider that large state universities may be over 60,000 total students on their own
You can imagine that 500,000 is a small fraction of total enrollment
> Our data indicates that an average of 2,200 riders out of approximately 21,000 students and 5,000 employees ride CAT bus to the university, and we’re aware that the cancelation of CAT bus operations precluded a number of students from attending class.
18:42
how many of you are physicist?
that just does not compute
(this is from an email)
@KyleKanos Well technically anything between 0 and 2,200 is "a number"
@AGoogler Again, I don't know for the US, but most degrees are done in non-science fields around here. For instance, more than 50% of the students at my university are studying Liberal Arts. Then there is fine arts, political science, business, humanities, science, and finally engineering
But there's no way that our city's bus system transports double the number of employees as students
@JimdalftheGrey GT is well... known.. for our undergrad taking 5 years. I did mine in 4, but didn't take any summers off
@KyleKanos How do you read that? I see that as "2200 people out of 26000 take the bus to campus"
No breakdown of how many of the 2200 are students or employees
18:44
@KyleKanos That's just reading it wrong.
And 5000 employees ride it
Or is it saying 2200 out of (21000 + 5000)?
@KyleKanos I read that as 2200 out of (21000+5000)
@KyleKanos yeah
18:44
@KyleKanos Duh.
@JimdalftheGrey Georgia Tech
Got it
I'm dumb
Thanks guys
@ACuriousMind I think part of why people get hung up about units is that they don't understand how to set units equal to one. Can you explain this? I never understood that.
18:45
@GlenTheUdderboat Yeah, 50 hours of classes. They like to punish us.
@tpg2114 Spread over three years?
@GlenTheUdderboat Well, spread over how ever long it takes you to graduate
But you have to take 6 classes in the first two years in order to take the PhD qualifying exams
If you don't pass those, you can't get into the PhD program (and if your advisor is nice, you can take the remaining 12 hours to get a masters and leave)
that's it? I had to take 6 classes to get my master's degree
Our masters is 30 hours
6 in the core classes, some research hours and some math hours
6 classes* which is like 18-20 hours
18:47
Hmm , how is it possible to cover big syllabus in only 6 classes?
wait, is this hours thing how long you spend in class for the entire course?
@JimdalftheGrey 1 class = 3 credit hours = 3 hours of class/week for 1 semester
What are you people talking about?? Please distinguish courses, classes, hours, whatever.
Likewise, but 1 class for me is around 33 hours of lecture time total
Uh
33 hours class?
18:49
by class, I mean course
Total lecture time will be... 54 hours of lecture time I think
We have 17-18 weeks in a semester
11 per term, 22 in a semester (not including exam weeks, obviously)
So 6 courses = 18-20 credit hours = 324 hours of lecture
@tpg2114 Oh dumb me , I thought only 30 hours of lecture
@JimdalftheGrey So are you on quarters? Or trimesters?
@AGoogler Nope :) Studying for qualifying exams took a good 6 months and like... a thousand pages of notes
Which I actually boiled down to a handy set of Beamer slides to study
18:50
we have 4 terms, but the summer ones are shortened, so the regular terms are 1/3 of a year but the summer ones are 1/6
It makes little sense
Yeah, okay. That's trimesters.
basically
Summers are optional for us
I'm on ~15 week semester system
But we only have 2 semesters a year. August-December, January-May, May-August (optional)
Our classes are typically 50 minutes, meeting 3 times a week =2.5 hours --> 37.5 total hours per course
And it's usually like 17 or 18 weeks during the fall and spring, and 15 weeks during the summer
18:52
Sounds like all these systems are pretty much the same with minor deviations here and there
What are you guys learning?
I was required 36 total credit hours for my MS program (18 of which were courses, the remaining were research)
Yeah, pretty much. There isn't much variety in the US when it comes to that kind of thing. Basically just when terms start and end
For my PhD, I was required like 18 total credit hours of classes and did that after 1 year
@AGoogler Now? nothing. I'm on an undesired break from school
18:53
@KyleKanos No required electives or minors?
@AGoogler Aerospace engineering (computational)
@tpg2114 Not for the masters, no. It was a mandatory 2 years & they didn't want to waste time with electives
Lucky
@tpg2114 Cool!
It was also a small program. There were 4 students accepted each year
We're forced to get a math certificate for a MS, math minor for a PhD
18:53
@tpg2114 e...lec...tives? What means this word?
And at most two of us worked with the same advisor, so none of our electives would have been aligned
@JimdalftheGrey Depends on your department. They force our electives to either be math, computer science, or physics. And require 2 of our electives to be in aerospace but outside of our domain
I got permission to take some atmospheric sciences classes and have those count
Which I kind of cheated on a bit -- I took intro to atmospheric fluid dynamics. It was all stuff I already knew + Coriolis force + change z-coordinate to pressure
The recently started requiring the new PhD students at Clemson to take a curriculum. Required 4 courses minimum in your subfield
But it did include some forecasting/meteorology which was rpetty awesome
I took the 4 astro courses, but sadly didn't really learn anything in 2 of them because the instructor wasn't really that good
18:56
@KyleKanos We have to pick 2 "subfields" and take 2 classes each in those, and then 1 "outside" subfield and take 2 classes in that -- this gives us our 6 qualifying exam classes
And then for a PhD, we have to take 2 more "outside area" classes
One course, he basically gave us O(4) papers per class to read outside to learn about topic X. Thankfully, there were no tests & only one project: make a model of a galaxy by drawing from a double gaussian distribution of points
So I did aerodynamics and viscous flows for my subfields, thermodynamics/gas dynamics as my "outside" subfield and then had to take structural analysis and FEM as my "outside area" electives
Plus then my math minor which I just loaded up on numerical PDE and numerical methods classes
The other course was GR where he kept drawing manifolds and never even got to the EFE until the last class
Plus the required statistics class
So even though it was 50 credit hours, there's pretty substantial overlap between all my courses so I probably didn't really learn all that much new after the first couple...
Curse you HNQ list!
I keep thinking UX = Unix and not User Experience
And when I click on those, I am always disappointed b/c I don't care about UX
19:07
how do you calculate the self energy of a point charge ? is it finite?
19:30
@DanielSank there is a problem here that I tried but didn't solve. Maybe you can solve. A boy took two metal plates, very big and parallel. One plate is connected to the ground and the other isn't. On the grounded plate he spread a charge B and on the other a charge A. He asks which charges will be on the inner sides of this capacitor. (see continuation)
@DanielSank In my opinion the charge B on the grounded plate is irrelevant because the Earth may supply or absorb whatever charge needed. In my opinion what is relevant is the charge on the free plate, and I assume that all the charge A will go to the inner face of this plate, while the grounded plate will absorb from the Earth and spread on the inner face a charge $-A$. But I am not sure. Can you take the problem?
user54412
@KyleKanos There's some poetic irony in a person who cares about Unix not caring about UX ;)
@ChrisWhite Terminal or Die!
More seriously, it seems to me that the few UX questions I've accidentally gone to are more about people designing a GUI
Since I'm not interested in GUI design, I'm not really interested in going there.
@DanielSank This is the site.
I'm sure there's some more interesting posts there that I might care about, but I've not the time to check it out
19:52
@ACuriousMind That's way too formal. I figured out $(1-x)^{-1}\sim 1+x$ is actually what I was looking for :D
20:12
Bounty expires today, would be a shame for it to go to waste: physics.stackexchange.com/questions/160912/…
20:27
@AGoogler Infinite
In quantum field theory, the statistical mechanics of fields, and the theory of self-similar geometric structures, renormalization is any of a collection of techniques used to treat infinities arising in calculated quantities. Renormalization specifies relationships between parameters in the theory when the parameters describing large distance scales differ from the parameters describing small distances. Physically, the pileup of contributions from an infinity of scales involved in a problem may then result in infinities. When describing space and time as a continuum, certain statistical and quantum...
Renormalization is a hack.
Certainly not
The well known QFT renormalization may be iffy, but I think renormalization occurs also classically in e.g. stat mech
vzn
vzn
21:06
@Danu some (unexpectedly/ surprisingly) strong sentiments about philosophy of science in this room.
Yeah, I think it's important to pay attention to these things, especially when making claims about 'fundamentals of QM'
@Danu I understand that you are a student. In which year? More shortly, do you remember electrodynamics?
@Sofia Yes, I do, my last course on it was slightly less than 2 years ago, or 2 weeks ago if you count the quantum field theory.
vzn
vzn
seems, not so many physicists & other scientists take philosophy of science all that seriously.
Why?
@vzn Which makes them look really ridiculous to those of us who do
vzn
vzn
21:09
so danu did you take class(es) in the area?
Yes
@Danu I tried hard to solve a problem of a user. I didn't succeed, but now I seem to have an idea. And I'd like that someone tell me if it seems logical.
@vzn Only one class, but I wouldn't mind taking more if I had the time
vzn
vzn
philosophy class? what title?
@Sofia Okay...
@vzn Philosophy of science
vzn
vzn
21:11
enjoyed a philosophy (of AI) undergrad class myself many yrs ago.
some brief confrontational fireworks in it (seminar fmt).
@Danu the problem is here. The user has two parallel big metallic plates,
@vzn Heh, I had many a discussion with humanities students who couldn't understand QM
vzn
vzn
scientists have trouble understanding QM also :\
@Danu and on the plate connected to the Earth he pours a charge B, while on the other plate a charge A, opposite.
@Danu he wishes to know what amount of charge one will have on the interior faces of the two plates.
ok
21:21
@Danu then, can we move to a chat room and talk?
Sure, whatever you want
I can join the one that that guy on the question opened
@Danu it's the best
@Danu let's go there.
@Danu can you enter that room?
I'm there: Just start talking ;)
@Danu No, I have the impression that it doesn't work. I don't see you.
I'm in the room that the asker offered to open
Please just copy-paste a link to whichever room you want to talk in
I don't care
21:37
@0celo7 Lol, that's just Taylor expanding, isn't it?
@ACuriousMind Yeah lol
@ACuriousMind You! Do you maybe know of some mathematical monograph on Yang-Mills?
@StanShunpike Well, all these ugly constants like Boltzmann or hbar or whatever are, essentially, just unit conversion factors. If you define all of them to be just 1, then everything becomes expressed by powers of one base unit (commonly taken to be energy, that's why you see masses expressed in GeV and such, that's c=1m for example)
I've been looking around for quite a bit, but can't find anything...
@ACuriousMind And length = 1/GeV, time = 1/GeV....
Does anyone have a recommended text to learn about Yang Mills from? I have been using Quantum Field Theory for the Gifted Amateur to learn QFT from but I realized they don't seem to cover Yang-Mills. I have studied a bit of Riemannian geometry and from Danu I learned it's the same derivative. So clearly I just need a text that helps me make the bridge from Riemannian geometry to the gauge covariant derivative. But just googling around about it hasn't helped much.
@Jiminion Yeah... I don't think she's in there
Dealing with less-digitally-comfortable people is such a hassle sometimes ;)
She's in that one
@ACuriousMind Oh, that's clever! I was just thinking about that yesterday, how Boltzmann's looked like a conversion factor. I learned about statistical mechanics for the first time and I was like, "Boltzmann's constant has pretty simple units. I expected something more complicated." But that makes sense. Cool.
@Danu Ooof. Well, I pieced what I know together from the "review" section of the papers that deal with gauge theories, but they always just present the piece of it they need. I know no place where it's all presented in one piece.
That has also been annoying me
@KyleKanos Thanks man
21:43
@Danu: Are mods restricted in the number of reviews per day?
@KyleKanos Not that I know of, why?
I ran out of close votes and first post reviews
@glance Nah, I meant the conserved currents that are, by Noether's theorem, also associated to symmetries. The Noether charge is, after all, just the spatial integration over the $j^0$-component
@KyleKanos Actually... the normal ones may just apply
And seeing Qmechanic as a recent editor of a post made me curious
21:44
I never really pushed it :P
@KyleKanos Qmechanic edits probably 300 posts a day. What about this edit made you curious?
The fact that I ran out of First Post reviews (where most of my edits are made) made me wonder if mods don't run out
That's all
But edits.... Can you even run out of edits? I don't think so
I doubt he goes through the queue all that often.
Sorry
I don't think you can run out of edits
But I do a lot of editing via the First Post queue
21:47
That is a lot
It's kinda the most important one
First impressions and all
That's why I tend to steer away from it
I treat newbies like everyone else. And I can be pretty nasty to everyon else
@ACuriousMind Told ya I've done it before ;)
@Danu: Whee, you wrote an answer!
@Danu Zeidler's third book is on Yang-Mills

« first day (1572 days earlier)      last day (3655 days later) »