« first day (2030 days earlier)      last day (2893 days later) » 

6:00 PM
thermodynamics, classical Mechanics, optics, Waves, electricity and magnetism
 
@ramsay I think Optics is cool, Electromagnetism too
You can do some cool stuff with thermodynamics + electromagnetism writing about the thermoelectric effect (Aka. Peltier effect)
 
how will i start, do i have to use laboratories, i have lots of confusion (i have never done assignment like this)
 
Um...shouldn't you ask your teacher how you're supposed to do your assignment?
 
@ramsay Step one: figure out what you want to do
 
Guys
Do you think that a function $\cos_e \theta$, to span all possible conic sections
Would be continuous at $e = 1$
Can you go continuously from ellipses to parabolas
I think that won't do
That would be like...
$\cos \theta \rightarrow \theta^2$
 
6:08 PM
@Slereah No clue
Ask ACM
 
Well he's right there
ACM isn't a gypsy fortune telling machine
You don't just put coins in him and ask questions
 
@BernardMeurer yes after that
 
But my guess is that you can't :(
And here lies the rub with my coordinate system
 
@ACuriousMind our teacher is quite rude with us
 
@ramsay Step two: Figure out what do you need to do it. A lab? Some material? Buying something? The help of a physicist? You gotta figure that out
@Slereah I disagree, @ACuriousMind is exactly a gypsy fortune telling machine, except that he's FREE
@ACuriousMind What will I have for dinner tonight?
 
6:11 PM
@BernardMeurer i see, what next
 
@ramsay Than you do it :)
 
Is that you @ACuriousMind
 
@Slereah LOL
 
Hm
If I can't use that coordinate system
What to do
 
@Slereah I have a bigger crystal ball, but otherwise, quite accurate
 
6:13 PM
Is there a systematic method to do coordinate systems for direct sums of manifolds
 
@ACuriousMind So, what about my dinner?
 
@BernardMeurer beer and chocolate pudding
 
@ACuriousMind That will be my mission. I'll go buy beer and pudding now :)
 
The ACM never lies
 
Hmm... to beer or not to beer?
 
6:20 PM
@Danu That's not a valid question. The answer is always yes
 
Hey... a highly cited physicist on SE: mathoverflow.net/questions/239857/…
(Not on PSE...)
 
@Danu Dayum, that's cool
 
@BernardMeurer quest.
 
@3075 No changes
I'm dying
 
it could take a few days.
 
6:39 PM
@lucas sure
 
@AccidentalFourierTransform Excuse me. What sure?:-)
 
what's eating you?
you wanted to ask something
i think
right?
 
@AccidentalFourierTransform Oh yes. Thank you because of your attention. I asked it from MAFIA and he/she answered. Thank you again:-)
 
all right then
cheers
 
So yeah
 
6:44 PM
@ACuriousMind What will my PhD be about?
 
i have a question:
if i roll a ball horizontally with and in some part of its track there is a well, so will the ball come back after falling inside (*collisions are elastic, no kind of resistant force*)
 
I think that function wouldn't be continuous
So it's not continuous or injective
I'm starting to think
Probably not a good coordinate chart
 
I'm a bit confused about distinguishing between matrices and tensors, matrix multiplication and inner products
 
Matrices are just arrays of numbers
Tensors are multilinear maps on vector spaces
 
@Carlos IIRC there is a post on PSE about that
 
6:54 PM
For example, the angular momentum tensor as seen in a different frame of reference with a Lorentz transformation matrix Lambda is given by Lambda M and not inner(Lambda, M) = Lambda^T g M
 
like, almost the same title
 
Is that correct?
Oh, which post?
 
Ill try to find it
 
In my notation juxtaposition, e.g. Lambda M, denotes the matrix product not inner product
 
there are many, actually
 
6:55 PM
Thanks!
 
lifehack: write the key words of your question into google, and add site:http://physics.stackexchange.com/ at the end
works like a charm
 
I saw some of those earlier but I can't find one that relates to the Lorentz transformation specifically
I assume that X' = Λ X in en.wikipedia.org/wiki/… denotes matrix multiplication, the article seems to indicate so
 
whats your overall level at tensor calculus?
just a beginner?
 
So the Λ which denote Lorentz transformations already "know" the metric structure (Minkowski in this case)
I've done some tensor calculus, Christoffel symbols and all that
e.g. I explicitly derived the geodesic equations for a sphere in spherical coordinates
Same for hyperbolic space in the Poincare disc/ball model
It's just a conceptual subtlety/distinction that's often not presented clearly
 
nvm then, its just that some time ago I wrote a semi detailed answer introducing lorentz transformations for beginners
not for you then ;-)
 
7:01 PM
@Carlos note that you can use $\text{math}$ here
 
Sometimes I find that I have to go back to simpler concepts and really understand subtle points that I missed
@Danu Oh thanks
 
to enable mathjax see here
17
A: Any chance of MathJax in chat?

Ilmari KaronenAs a workaround while this request is pending, there exist several client-side workarounds that can be used to enable LaTeX rendering in chat, including: ChatJax, a set of bookmarklets by robjohn to enable dynamic MathJax support in chat. Commonly used in the Mathematics chat room. An altern...

 
Like the distinction between points and vectors (which becomes much more evident in curved geometry, but many people equate the 2 because they're working in Cartesian coordinates in R^n)
 
anyway, you could try to ask a question here
if someone knows the answer they may try to help you
but you should use mathjax cause otherwise its kind of hard to understand what you mean
 
OK straight question: In en.wikipedia.org/wiki/…, the Λ matrix representing a Lorentz transformation can be applied not just to a four-vector but also a rank-2 four-tensor, right? Like the en.wikipedia.org/wiki/Relativistic_angular_momentum
 
7:04 PM
Yeah you can only equate them in a vector space
Yes
Although
 
for higher rank tensors you need more factors of $\Lambda$
 
I think it's a bit of a notation abuse
 
Ah ok
 
Technically I don't think the two are the same object
 
What does the transformation for higher-rank tensors look like?
 
7:05 PM
as in $T\to \Lambda\Lambda T$ for second rank tensors
just let yourself be guided by the notation
"the obvious answer" is usually the right one
when it comes to tensor notation
 
We just say it's the same object because the matrix rep is the same :p
 
$T^{\mu\nu}\to \Lambda^\mu{}_\sigma\Lambda^\nu{}_\rho T^{\sigma\rho}$
by "obvious" I mean that there is only one combination of indices that makes sense
 
That looks a lot more elegant!
So in general it's n Lambdas for a rank-n tensor?
 
yes
Or, alternatively
$2n$ lambdas of the spinor rep
 
7:11 PM
e.g. for a rank 3 tensor the transformation is T^{\mu\nu\pi}\to \Lambda^\mu{}_\sigma \Lambda^\nu{}_\rho \Lambda^\pi{}_\tau T^{\sigma\rho\tau}
I get it now, thanks a lot!
 
You mayn't
 
@Carlos you forgot the $'s
 
Oops
$T^{\mu\nu\pi}\to \Lambda^\mu{}_\sigma \Lambda^\nu{}_\rho \Lambda^\pi{}_\tau T^{\sigma\rho\tau}$
 
@Carlos there you go
 
The index $\mu$ appears twice upstairs
Shouldn't it be $\sigma$ instead?
 
7:22 PM
@dmckee I had posted a wrong answer and then I deleted it but before deletion it had been upvoted. Can you delete it?
 
Seems to be a typo yes
 
@Carlos yep
good catch
 
@lucas It is not the vote that is preventing deletion but the acceptance. Unfortunately we (the mods) are strong encouraged not to delete accepted answers.
Your best bet is probably to improve the post to the point that it is correct. Or to add a caveat to it pointing out the limitation of answer.
 
@dmckee No, mine isn't accepted. OP accepted another answer.
 
Uh ... link?
 
7:27 PM
3
A: What are the accelerations of blocks?

lucas $$l=x_A+(x_A-k_1)+k_2+k_3+2\pi R$$ $$k_3=x_B-2R-(x_A-k_1)$$ $$l=x_A+x_B+constant$$ so, we have: $$0=v_A+v_B$$ and$$a_A=-a_B$$

@dmckee I posted the correct answer after.
That time I wasn't familiar with the site.
@dmckee My wrong answer exists yet but it is red.
 
@lucas red means deleted
only you can see it (and high rep users presumably)
 
@AccidentalFourierTransform Why is it removed?
 
@lucas maybe you deleted it
 
@lucas Deleted content isn't removed from the database, it's just not show to most users. The shading indicates that it is deleted. You can see it. Mods and some Stack Overflow staff (Community Mods) can see it. Users with at least 10k rep on the site can see it.
@lucas In my view it says "deleted by owner Apr 22 at 20:02"
 
or it was flagged as "not an answer" and reviewers voted to delete it
 
7:32 PM
@dmckee But it is obviously wrong. Isn't it better to remove it?
 
@lucas It is removed. Most users cannot see it, and the rest sees a clear mark that it is deleted.
 
just dont worry about it
it happens to the best of us
 
@lucas It's deleted. That's as removed as things get without special intervention by the Stack Overflow programmers.
 
move on with your life :P
 
I've got 6 or ten posts that I deleted when I realized they were incurably wrong. Don't fret it.
 
7:36 PM
OK. Thanks to all.
 
7:51 PM
hey guys, real life everyday question
what would you guys estimate is the total amount of light hitting us every second
(i.e.: including reflections and refractions from the environment, across the spectrum)
 
@user507974 Depends, in Brazil I can guarantee it's a hell lot
Some guy just made a post on the 3D printing forum I'm part of asking if we know where to get HP LaserJet Toners
 
@Danu I'm having a stupid problem. It's well-known that the derivatives of the Chern-Simons forms $\omega_{2p-1}$ give $\mathrm{tr}(F^p)$, but, uh...they don't. For instance, $\omega_3 = \mathrm{tr}(\mathrm{d}A\wedge A +\frac{2}{3}A\wedge A\wedge A)$ gives $\mathrm{d}\omega_3 = \mathrm{tr}(\mathrm{d}A\wedge\mathrm{d}A + 2\mathrm{d}A\wedge A \wedge A)$. Where's the $A\wedge A\wedge A\wedge A$ from $F\wedge F$? I must be missing something obvious.
It doesn't help that none of the dozen sources I sought out ever does this computation explicitly. That's why I think I must be missing something rather obvious
 
Ah, yes!
My TA actually said that, too. NOBODY DOES THIS
because it sucks
I have a computation in coordinates, only.
Will that do?
 
I would prefer one in forms, but I'll take the coordinate-laden one nonetheless ;)
 
8:11 PM
Sometimes I forget what my physics project even are
I am not very focused
 
8:22 PM
anyone know if by $\sigma$ is an $n-$cycle we mean its cycle decomposition in total is of length $n$ or if it's a single cycle of length $n$?
in the context of the group of symmetry $S_n$
 
I don't know
I don't know anything
What is even a bird
I don't know
 
that's very helpful slereah :)
are you practicing your response for when you get questioned by a law enforcer/detective ?
 
Nah
 
@Obliv Can you email me the specs of the code you want? Also, which language? C, C++, Python?
 
That would be "I ain't talkin' copper"
 
8:24 PM
meurerbernardo@gmail.com
 
What's a moider?
 
anything that works. it can be in assembly for all I care. Thanks man I'll email it later
 
Does anyone know if there's like
A nice trick
 
@Obliv Lol, I'll do it in C++ than, because it's faster than Python but less of a pain in the butt than C
 
To get a decent spacetime from a connected sum
Like if I have two manifolds
I make a connected sum of them
What do the coordinates and Einstein equation look like
 
8:26 PM
the way you talk about high level math is so casual @slereah and yet it makes absolutely no sense to me :D
 
I do use the word "magic" and "shenanigans" a lot when talking about physics
Then again
Connected sum is just basically stitching two manifolds along a disk
Coordinates probably aren't that hard to derive
Though I guess they will be a bit awkward
But
What is the best way to do it for a torus
What's a good coordinate patch for a torus minus a disk
Wait, isn't a torus minus a disk also a plane with a handle
Aaaaah
I have recursive problems
 
a taurus without a disk is not the same thing as an airplane with a handle.. @slereah nice try though
 
I mean, what's the idea coordinate patches for both the plane and the torus such that they are elegant and close to "canonical" coordinates
I guess for the plane it would be polar coordinates - a disk
Easy to do
But what about the torus
If I remove a disk from the torus, I can't use the standard toroidal coordinates easily
Unless the disk I remove is like
a quarter of the torus?
Like $\varphi, \theta \in [0,\pi] \times [0,\pi]$
@ACuriousMind, mighty fortune telling machine, does that sound reasonable
 
@Slereah Uhhhh...not sure if it's "reasonable" but it sounds as if it could work
 
8:41 PM
I'm a bit worried about identifying those edges
Although
I guess it can still lead to a smooth metric, I dunno
ALTHOUGH
Perhaps a less shitty method to use would be
Take a cylinder
Remove two disks from the plane
Identify each ends instead
That way it's easy to find a spacetime where the mouthes are separated by a well known distance
 
8:57 PM
@BernardMeurer oddly enough my experience has been the opposite of both those things :-P
 
@DavidZ Really? It's hard for a compiled language to be slower than a compiled on in general
 
In terms of execution speed, sure, but I find the coding speed usually more than makes up for it
 
@DavidZ Depends on the scope and objective of the project no? I mean sure you'll code faster in Python, but if you'll be doing something computationally intense it will probably pay off
also, C++ isn't that hard to write
 
Well, in my experience relatively few projects are computationally intensive enough to make C++ beneficial.
and honestly, the amount of time I've spent tracking down memory errors in C++ is often enough to write a hundred Python programs to do the same thing :-P
 
Hm let's see
Cylinder will be two coordinate patch
Plus one for the plane
That's... 3 transition maps to do
 
9:08 PM
@DavidZ I see you're point, but still, I need opportunities to learn C++
as much as I love python
and I'd marry Python if I could
 
That's fair
 
@Danu lol. I would not have thought to look in "Principles of Algebraic Geometry" for an explanation of differential geometry
 
That crazy Reed Simon functional analysis book is full of physicist intuition and proofs, amazing!
 
@bolbteppa Shittiest book I know :D
 
I've been ignoring this book for ages because it looked so bad, but when I cracked it open I kept finding at the beginning of many proofs a short summary of what you're actually doing, very cool!
 
9:58 PM
Hey @ACuriousMind when defining the Tor functors; does one need to consider Tor_0 separately?
 
@Danu Unless I'm misunderstanding your notation, isn't "Tor_0" just the tensor product?
 
What I'm asking is: Does this follow from $\operatorname{Tor}_n(A,B)=H_n(F_*\otimes B)$ or not?
Because if so, it's not so clear to me.
Or is it just a separate definition?
 
vzn
@DavidZ, slereah says he can be guest speaker at 16:00 UTC Tue Jun14th mtg, ok with you? if so can you put up link?
 
@ACuriousMind I can't see how the kernel of $F_0$ (over the image of $F_1$) should give $A\otimes B$ in any way
 
@Danu Just recompile the kernel with the appropriate flags
 
10:09 PM
@BernardMeurer hurr durr computers
 
@Danu All I can do to help :p
 
10:22 PM
@Danu Sorry, the dual statement is easier, but I think I have it: The resolution $F_\bullet\to A\to 0$ is exact, so $F_\bullet\otimes B\to A\otimes B\to 0$ is still exact. So $\operatorname{im}(F_1\otimes B\to F_0\otimes B) = \ker(F_0\otimes B\to A\otimes B)$and you have $H_0 = F_0\otimes B/\ker(F_0\otimes B\to A\otimes B)$. Moreover, $F_0\otimes B\to A\otimes B$ is surjective, and some isomorphism theorem gives you this quotient is just $A\otimes B$.
 
Sorry but $H_0=\ker/\operatorname{im}=\ker/\ker =0$? (exactness)
 
What? Okay, let's go slower: $H_0 = \ker(F_0\otimes B\to 0)/\mathrm{im}(F_1\otimes B\to F_0\otimes B)$, right?
Now, $\ker(F_0\otimes B\to 0) = F_0\otimes B$, and we need to find the image of $F_1\otimes B\to F_0\otimes B)$.
 
Ah...
 
This we do by considering that $F_1\otimes B\to F_0\otimes B\to A\otimes B$ is exact, so by exactness, $\mathrm{im}(F_1\otimes B\to F_0\otimes B) = \ker(F_0\otimes B\to A\otimes B)$
 
The point is it cuts out the $A$ step
I forgot about that.
 
10:27 PM
Yes, the complex that computes homology cuts out the $A$ (else it would be exact, and the homology zero).
 
We're considering the homology of $F_* $, not $F_* $ with $A$ patched on (but the latter is the free resolution)
@ACuriousMind This was exactly (hur dur) my problem
 
To finish my statement, the first isomorphism theorem says that $F_0\otimes B/\ker(\phi) = \mathrm{im}(\phi)$, and $F_0\otimes B\to A\otimes B$ is surjective, so we get that $H_0$ is the original functor.
 
Yes, of course, that's all clear
I just forgot that I'm not supposed to consdier the homology of the free resolution
 
10:39 PM
Can somone explain the theory behind SDS-PAGE to me?
 
11:13 PM
@BernardMeurer Numba or PyPy?
 
@alarge I need an excuse to use C++
I know a lot of Python already :p
 

« first day (2030 days earlier)      last day (2893 days later) »