« first day (1720 days earlier)      last day (3505 days later) » 

00:22
@FenderLesPaul wow
Come up with a solution in your sleep?
Btw have you done exercise 10.1 in Wald?
That's another one in the "hopelessly stuck" column
There seems to be an abundance of questions involving $\delta T/\delta g_{\mu\nu}$ the last day or two :/
@KyleKanos I think my comment on the latest one sums it up
unless is there another one?
shieeeet
He deleted the question in which you commented about it being 0
Then posted this one:
0
Q: What is the value of the stress energy tensor?

MrDiIf we are living in a portion of space-time where the metric is very close to flat space and we know that the stress energy tensor is negligible at this portion of space-time is it ok to assume that $\frac{\delta T_{\mu\nu}}{\delta g^{\mu\nu}}$ is also negligable?

@KyleKanos cf. chat.stackexchange.com/transcript/message/22904417#22904417 for my realization there's a new one
00:28
I still don't completely understand what "X tensor components are small" even means, because they can be made "large" by a suitable coordinate trafo.
Well, you can comment that to him
It's a problem I have with all GR treatments
Maybe you can define a topology on the space of tensor components or something
then define a notion of "close"
then define the notion of "close to 0"
It's also funny that he spelled it negligible and negligable in that post.
404'd
@KyleKanos I know Straumann, which is ultra pedantic, mentions the problem of what "small tensor" even means, but is pretty much like ¯_(ツ)_/¯ "can't do GR without it"
He deleted that new post to edit the old post
00:36
@KyleKanos I obtained that paper from Rindler et. other guy.
The 1971 one.
Karlov
I'm pretty sure it has what I need.
I need to review the Fermi derivative.
@KyleKanos yeah
Still had it open in my PDF viewer
You could have just sent it to me
Saved me 10 minutes in 95 degree DC weather
It was paywalled
00:39
I know
I got it for free at the LoC
They let me make a copy
That was nice of them
@KyleKanos Did the paper make sense to you? You've probably never dealt with Fermi transport.
I'm gonna play Skyrim for a bit. Is anyone gonna bother watching? No use streaming if not
@0celo7 sadly that's the only exercise in the chapter I haven't attempted but I can try if need be; any of the other ones I can readily comment on though
and no I haven't solved it yet sorry! busy eating my horribly unhealthy dinner
4
Q: Application of Cauchy-Schwarz with Sobolev norms

0celo7I'm working through the problems in the initial value formulation chapter in Wald's General Relativity. A short summary of the problem. I have to show that $$\sup_{x\in A}|f(x)|\le C||f||_{A,k}$$ where $A\subset \mathbb{R}^n$ satisfies the uniform interior cone condition, $C$ is a constant, $k>...

^^As far as I've gotten.
@0celo7 Bits of it made sense, but not really the whole picture
00:46
@KyleKanos The big thing is defining what "constant acceleration" even means.
As in is it constant w.r.t. time/proper time/space?
it's constant wrt. a fermi transported inertial frame
stream is live btw
@KyleKanos I take that back. Constant direction is what is needed to be defined.
oh I'd be down to watch
where's the link again?
00:53
shit my laptop is really slow
second monitor
dope thanks
all mods tonight
until ~9:45
THE PUPPY
Hey guys. The following question might seem a little idiotic or even pointless.
I thought about posting it... but its really short so it thought to ask you guys
01:08
Feel free
suppose we have a point moving in space. So then can we say that

v= (1/Δs)(xv_x +yv_y +zv_z) ?
delta s is?
s is sqrt{x^2 +y^2 +z^2)
I am reading the Feynman Lecture and it has some pretty basic stuff. I know the velocity is the derivative of position, so when i differentiate this pops up (i think).
v = d **x**/dt
So i wonder if its right and it's indeed that same to sqrt{v_x^2 + v_y^2 +v_z^2}
So i can do it only to its axis?
01:16
It's easier with latex: $\mathbf{v}=d\mathbf{x}/dt$
So if $\mathbf{x}=x\hat{x}+y\hat{y}+z\hat{z}$, then the time-derivative of this is just the time-derivative of the components times the unit vectors: $d\mathbf{x}/dt=dx/dt\hat{x}+dy/dt\hat{y}+dz/dt\hat{z}$
Ohh thank you. I got it
:D
Which would be $\mathbf{v}=v_x\hat{x}+v_y\hat{y}+v_z\hat{z}$.
Dotting this with itself (to give you magnitude) would be $v=\sqrt{v_x^2+v_y^2+v_z^2}$, as you said
01:52
I fixed it.
Well that was a god run except for the end there.
Night guys!
@FenderLesPaul Did you see my MSE post above? I'd appreciate you looking at it
@0celo7 sure thang dude
and I'll try to figure out whatever Straumann is trying to do
cool beans
obe
obe
02:05
@0celo7 I'm not, I didn't move from the quantum chapter yet.
I told you, starcraft II.
is that the blizzard game about the aliens?
obe
obe
Idc about the story.
 
1 hour later…
03:27
(see version before I edited it)
made title good
@KyleKanos You might be amused by some of my edit messages.
What should I have written?
Oh poop, forgot the "lol" bit :(
That was funny
I do things like that alllll the time
@KyleKanos It's all I know how to do to keep myself from depression when editing titles on this site.
Question on energy?
And you'd be proud: my son has taken an interest in cephalopods.
03:32
Electricity: how does it work
@KyleKanos YES!
I showed him the camouflaging one and he was amazed
@KyleKanos I found this book quite interesting.
@KyleKanos Heh, there are so many camouflaging ones! Show him some videos of cuttlefish too. Those things are amazing.
@DanielSank 200 pages is a bit long for a 5 year old
@KyleKanos Of course. I just figured that as parents tend to learn a lot about their childrens' interests that this might be up your alley :-)
Hmm. Does it say anything about how to cook them in the book?
03:37
jerk
I'm sad now.
Imagine someone eating your family dog.
Hmm
Well I don't have a dog...
they're like kids except they never grow up, always depend on you, never get a job and never leave you alone
^ Truth
They do not borrow your car, ask for money, and hate your for ~ten years, however.
I don't forsee any problems with them borrowing the car or asking for money. I'm working to avoid them disliking me
I've also been giving my son an interest in magic
We watch a lot of Penn & Teller Fool Us
03:42
That's a really great show.
Do you have a deck of cards on hand?
Slightly out of arms reach
Want to learn a trick?
Ok, grab the deck and separate out the jacks.
03:45
Ok, take the four jacks in a little four card pile facing you, and put a fifth card at the bottom.
In practice you have to do this real sly so nobody knows there's a fifth card.
Face up or down?
All cards facing you.
You just have a little five card pile, which everyone else thinks is a four card pile of jacks.
Okay...so what's the next step?
Ok, so you have this little pile. Hold it with your fingers on the long edges, palm toward you, and with your other hand peel off one card at a time. In your head, you are counting the cards for the audience.
So, you hold the pile and show the top card, that's 1.
Peel off a jack and show the next, that's 2.
Peel another showing the next jack, that's 3.
Peel the third revealing the fourth (with the fifth secret card behind it), that's four.
At this point you have three jacks in your left hand and two cards in your right - a jack with a secret card behind it.
Ok?
Got it
I guess the stack should be J J J 3 J (assuming I picked a 3)?
03:57
Exactly!
You put those peeled off jacks behind the fourth jack and the 3.
Now place that little stack on top of the deck, face down with the rest of the cards.
Now pick up the deck.
These are circus jacks which do amazing tricks.
Deal out the top four cards face down in a row.
The jacks are lining up for their performance.
You now have, face down in a row, J J J 3 and the deck has a J on top.
ok?
Got it
Ok.
"Jack #1 is the one who can climb instantly from the bottom to the top of the deck"
Put the deck down on J #1, flourish, and reveal that he pops off the top of the deck!
Pick up the deck making sure not to show the bottom card.
"Jack #2 starts on the top and dives down into the little pool of water on the bottom"
Put Jack #2 on top, flourish and riff the cards or whatever, and with a snap flip the deck over. The jack appears on the bottom!
I see where this is going. The 3 will go in the middle?
Yes. Stick #3 face down in the middle while the rest of the deck is still face up.
(This one is the trapeze artist who flips from the middle to the top coming off the top in the other direction)
Then the bogus card is shoved into the middle and you spread the deck showing that he (the contortionist) has inverted himself!
This always surprises people, it takes 2 minutes, and almost no setup. It's the most effective trick I know for a party.
You can precede this with whatever trick you want which gets someone else to "find" the jacks for you.
or whatever
That's clever
I'll have to practice that some
04:03
Yeah, it's awesome.
Don't repeat it if people ask you to. They know what to look for the second time :-)
Your 5 year old on the other hand...
He's sharp though. He caught on with the 2 cards & 1 coin trick
On the 3rd try too
@KyleKanos Don't know that one.
It's simple: take any two cards & a larger coin (quarter, half-dollar, sakagewa dollar, etc)
Layer as ||o (|=card, o=coin)
pinch with the fingers, reveal top card & place back on top of 2nd card
Reveal 2nd card, place back beneath the coin: |o|
Then just slide the coin out between the cards
Works better if you show the cards with different hands (less obvious that way)
In ||o what's on the bottom?
top -> bottom, so coin's hidden at the bottom
So no one sees the coin & it just appears between the two cards
04:08
So I pick up top card. Audience sees back of bottom card.
Right-o
Then I replace top card.
Then with other hand slide second card out, and replace.
But replace under coin.
How to reveal coin?
Yes..You've gotta transfer the pinched coin from one hand to the other
Lift up top card again?
Oh, wait a sec...
I do the reveal as letting it slide between the two
04:10
I thought this stack was sitting on a table.
Do a quick flick on the two or wiggle them or something, then have a helper hold their hand out
It's in my hands?
In your hands
I'm trying to figure out how to make this surprising.
How to hold stack?
Pinched between a thumb & index (maybe middle too) fingers, face down
Longways is along your line-of-sight
04:12
Ah.
Now I see.
Have to set up subtly!
Yes you do
In high school I met a guy who was good.
He would do tricks in the hall, etc.
Holy crap...it's after midnight!
No wonder I'm tired...
One time he did a trick surrounded on all sides by spectators, at the end, he asked the victim for the time.
I gotta get to bed
04:14
Victim looks at watch, find giant coin under wrist band.
Good night.
Later pal.
vzn
vzn
05:05
10
Q: Statistical analysis of data in Physics

Comp_WarriorAnalysis of data is integral in bridging the gap between theory and experiment. How much do the results of the analysis depend upon the choice between Bayesian and frequentist methods? For instance, consider experiments in particle physics. This paper by Louis Lyons states that particle physicis...

 
1 hour later…
06:15
@Phonon Decent, almost through exam period now
06:25
Did you know that a signatur is the same as a call number @Danu because a curios mind didn't :P
06:39
o.0
07:10
@vzn Wow! In my opinion, both the question, the linked paper and the existing answer are excellent. It is tempting to write an answer myself, but I fear it would draw down the excellent quality.
07:24
-3
A: Why does connecting a battery's positive terminal to the negative terminal of another battery not create a short circuit?

TrollMy name is jeffdkdo o I o k k kid,, and it the first place time to get my name on my phone is a great way to the gym and with a few lot to me I have was love you too so much for the rest next week is and what I was a great day and the other day of my friends are like the that the two men and a gr...

A textbook use case for the very low quality flag, and yet nobody flagged it as VLQ. (edit: maybe not so obvious. My bad.)
Not that I'm complaining too much, because it did get flagged, but I guess people are still unclear about what the different kinds of flags are for, maybe?
user54412
I remember seeing it already in the queue. What was it flagged as then?
user54412
Or maybe I flagged it as spam at that point too. It's not directly advertising a product, but it's clearly a Markov chain spambot and should be banned for that.
NAA, spam, and a custom flag
@ChrisWhite sure, but still, not advertising a product => not flaggable as spam
Whenever we delete VLQ posts we check whether the account needs to be nuked
@DavidZ what are some of the most commom mistakes people make when flagging?
Probably the most common ones are (1) using custom flags for cases that are covered by the standard flags, (2) using spam or offensive flags for things that are not spam or offensive, (3) flagging answers that are legitimate but simply wrong as NAA
You might find this interesting if you haven't already seen it:
5
Q: Which flag do I use for an inappropriate post?

David ZOther than off topic flags, there are several different types of flags that one can cast on a question, answer, or comment. What are each of these flags for? Question and answer flags: spam offensive/abusive/hate speech not an answer very low quality other Comment flags: rude or offensive ...

Though actually now that I look at that, I guess the VLQ post I linked earlier maybe wasn't actually a "textbook" example of VLQ. At least the words were actual words.
user54412
07:34
@DavidZ So it seems I'm not alone in thinking "spam" includes unsolicited bot noise. But then expanding to this definition was shot down.
user54412
The problem is that first post said we should flag gibberish under "abusive" since it abuses the system, but then in other posts and based on the wording of the "abusive" flag, it become clear that this would also be declined since it's not hate speech against minorities or something.
@DavidZ awesome! I hadnt seen that. Excellent. Yeah, my first couple flags I tried custom and then I realized I almost never need to use it unless there is some rare issue. I try never for obvious reasons.
@ChrisWhite Yeah, I haven't seen those specific posts but the underlying discussion has been going on for a while. The thing is, SE has a whole structure of automated spam prevention tools working behind the scenes. They take input from spam flags and accounts which are deleted for spamming. The tools are meant to work on advertisements, not just random irrelevant posts.
@ChrisWhite I am somewhat surprised to see that quote from Shog. I wouldn't think such posts qualify as offensive. Maybe I'll bug them about updating the flag description.
user54412
@DavidZ That makes sense. But it's a bit disappointing their tools basically can't handle broad data inputs. After all, the gibberish posts are training to make better spambots to sell to advertisers, and the more bots we slaughter in infancy the better the internet becomes.
I guess so. But I suppose it's a difficult problem.
I don't know how the tools work, but I know they have often warned us mods about how spam flags should only be used for actual spam
 
2 hours later…
09:18
@0celo7 if I have a one form $v$ how am I suppose to know if it represents a covariant or contravariant vector
in other words how do I know if it would have components with upper or lower indices?
user54412
@gonenc one-forms are covariant (lower-indexed) vectors
@ChrisWhite is that always the case?
user54412
yes
@ChrisWhite what if I want to represent a contravariant vector without indices, how do I do that?
user54412
I like Carroll's $\vec{v}$ for contravariant and $\tilde{v}$ for the associated covariant.
user54412
09:23
But I know plenty of index-free types who insist no decoration is needed as it will always be clear from context.
@ChrisWhite "it will always be clear from context" is a dangerous thing to say imo.
why do covariant vectors get special treatment then ie why do we call covariant vectors eg 1-form but don't give contravariant vectors some other name too?
user54412
well, forms of all sizes are special: 1-forms, 2-forms, 3-forms, etc. (these are just totally antisymmetric, all lower-indexed tensors)
user54412
I suppose we could have reversed everything and had a tower of 1-vectors, 2-vectors, etc., and only ever had 1-covariant-forms. It would amount to the same, and we just chose one paradigm.
@ChrisWhite I see
user54412
Another way of explaining away the asymmetry is to note single (contravariant) vectors are sort of natural -- they are just the basis vectors for the tangent space, and tangent spaces are easier to visualize than cotangent spaces. Once you have a tangent space, you can get the cotangent space for free (or vice versa).
user54412
09:31
Adding more levels to vectors doesn't do much (try visualizing the space of 2-vectors -- I can't really). But adding more forms is sensible. 1-forms are single-input maps from vectors to scalars, 2-forms take two vectors as input, etc.
@ChrisWhite if I understand correctly $T^{\mu \nu} \leftrightarrow \mathbf T$, where $\mathbf T$ is a 2-form and this generalizes for arbitrary number of indices
user54412
Is T stress-energy or an arbitrary tensor? Note stress-energy is symmetric and so not a proper form even with indices in the right place.
@ChrisWhite it was an arbitrary tensor the T was for tensor :)
which is of course antisymmetric
@ChrisWhite ::recoils in horror::
@0celo7 why is that?
09:45
@gonenc Context.
@gonenc Not every author uses heresy (indices).
@gonenc No one's stopping you from being wrong.
@0celo7 being wrong don't make me feel too good.
@0celo7 btw the book I'm using uses eg $dxdy$ for 2 forms
what is the rigorous notation for that?
user54412
For pedantry: Let's say $\mathbf{T}$ is covariant, a (0,2)-tensor. Given a basis $\vec{e}_{(\mu)}$ for the tangent space, we have the associated basis $\tilde{e}^{(\mu)}$, where $\tilde{e}^{(\mu)}(\vec{e}_{(\nu)}) = \delta^\mu_\nu$. Then $\mathbf{T} = T_{\mu\nu} \tilde{e}^{(\mu)} \otimes \tilde{e}^{(\nu)}$ and $T_{\mu\nu} = \mathbf{T}(\vec{e}^{(\mu)},\vec{e}^{(\nu)})$.
user54412
(or something like that -- too tired to read my own latex right now)
10:05
Yeah, that's fine @Chris
@ChrisWhite This is correct.
@ChrisWhite Why were you up at 5 AM?
@gonenc Note that because Chris was dying of lack of sleep, he messed up the indices on the vector basis in the last line. They should be downstairs.
Not covariant, those are labels for the elements of the basis, not a tensor index.
user54412
@0celo7 I find it amusing how long it took everyone to notice :P
user54412
I could also just argue that as mere labels they can be anywhere I please. Screw confused readers.
user54412
Here's something I haven't really seen before: Should we geoengineer larger ice caps? It's a proposal to intentionally change the long-term climate on Earth.
11:48
@ChrisWhite Didn't take me long to notice it took me until I had time to tell @gonenc I noticed.
@gonenc And which book is that?
It's fine as long as you understand there is a (suppressed) wedge there.
12:07
@0celo7 some book on diff forms. so the proper notation would be $dx \wedge dy$
which the book doesn't even mention :/
btw how do you write the killing equation ie $\nabla_{(\mu} K_{\nu)}$ in diff form notation
@gonenc uhh
I have to think about that one.
@0celo7 the symmetrization is the problem, if it was $\nabla_{[\mu}K_{\nu]}$ it would simply be $dK$
Yeah, I'm thinking whatever you do, it will be ad hoc and ugly.
@0celo7 would it make a good math.se question I can ask it there?
I think the question is unenlightening.
Wait for ACM to come by, he might know.
12:22
@0celo7 we should better ping @ACuriousMind then ;)
@gonenc The notation used in my course on diffgeo would be $(\nabla K)^\text{sym}(X,Y)$ where $K\in \Gamma(T^*M)\cong\Omega^1(TM)$
Like I said, ad hoc and unenlightening.
@0celo7 Not really ad hoc, but definitely unenlightening
@0celo7 yeah not so elegant indeed
Ad hoc in the sense that you had to define a symmetrizer.
Btw the real Killing equation is just $L_Kg=0$
12:32
@0celo7 haven't read that chapter yet
that one is a lie derivative right?
@gonenc I mean "real" in the sense that that is the enlightening definition
@gonenc yes
@0celo7 what about the Riemann curvature tensor? is there a neat diff form notation for it too?
@gonenc Yes
In differential geometry, the curvature form describes curvature of a connection on a principal bundle. It can be considered as an alternative to or generalization of the curvature tensor in Riemannian geometry. == Definition == Let G be a Lie group with Lie algebra , and P → B be a principal G-bundle. Let ω be an Ehresmann connection on P (which is a -valued one-form on P). Then the curvature form is the -valued 2-form on P defined by Here stands for exterior derivative, is defined in the article "Lie algebra-valued form" and D denotes the exterior covariant derivative. In other terms, where...
12:48
@gonenc See Wald p. 49 for the derivation.
@0celo7 I have a feeling that wald extensively uses diff forms
@gonenc No one's stopping you from being wrong.
You should get that checked out
Aw hell my iPad is going to run out of battery
@0celo7 I didn't get it am I wrong or what?
Dead wrong
Aw hell iPads can't charge though computers
@0celo7 lol
@0celo7 well my feelings are wrong then :D
12:53
Oh lol
I have my physical copy of wald in my bag :D
@0celo7 well I aint got no copy of wald, yet
Wald is pretty good for a second GR book
Nothing crazy like fiber bundles, but a healthy dose of topology
@0celo7 dunno, methinks I'll switch back to QM a bit and learn path formalism
QM is a lie though
God doesn't play dice
@0celo7 well it is closely related to some lie algebra, so you'd except that
@0celo7 if there is one...
12:58
@gonenc You're the only one here who doubts the existence of dice
@0celo7 aint nobody doubting about the existence of a die!
2 mins ago, by gonenc
@0celo7 if there is one...
Sure seems like it
@0celo7 btw do we have the same copy of wald p.49 starts with: "This is a useful formula since $\Gamma$"

« first day (1720 days earlier)      last day (3505 days later) »