« first day (2190 days earlier)      last day (2731 days later) » 

12:03 AM
@obe So once again you're not sticking with a decision...
 
12:21 AM
@0cleo7 is this sound: $\int \int_E f(x,y)~dx~dy \to \int \int_L T(u(x,y),v(x,y))~\frac{\partial(u,v)}{\partial(x,y)}~dx~dy$?
 
user218912
@0celo7 fml
 
defines the volume of a transformed function
@obe or can you confirm thats correct
 
user218912
looks ok to me
 
@Obliv What on earth is that supposed to mean? When you write down a formula, always explain your notation! What are $E,L$, what is $T$, where did $f$ go?
 
@ACuriousMind thank you
 
12:26 AM
@acuriousmind oh hey acm. E, L are regions and they change because f transforms into a function T(u(x,y),v(x,y))
 
Also, what kind of Frankenstein derivative is $\frac{\partial(u,v)}{\partial(x,y)}$?
 
oh I meant to do \mid on both sides to signify determinant whoops
 
@ACuriousMind Hey, can you explain $$\int_{B_y(3\rho_j)\cap B_p(\rho)}|\nabla f|^2\phi^2\,dV\ge C_1^2C^{-8}_{\mathcal{VD},\phi}C_{\mathcal P,\phi}\rho^{-2\epsilon}|f_{B_0}-f(y)|^2\rho_j^{2\epsilon-2}V_\phi(B_y(\rho_j))?‌​$$
standard notation
 
@Obliv So, what you're trying to say is that $E,L\subset\mathbb{R}^2$ and you have a function $t : \mathbb{R}^2\to\mathbb{R}^2, (x,y)\mapsto (u(x,y),v(x,y))$ such that $t(E) = L$? I don't know what "f transforms into a function T(u(x,y),v(x,y))" is supposed to mean, though.
 
@ACuriousMind Standard notation of Jacobian
 
user218912
12:30 AM
why is what Obliv said wrong? other than forgetting the determinant which some books don't even put
 
what is $T$ even supposed to be?
 
@obe It's not wrong, it's completely meaningless unless he explains the notation.
 
@ACuriousMind How much pure complex analysis do you know, anyway?
 
And yes, I might be nitpicky about E,L and such, but I genuinely have no idea what $T$ is.
 
@acuriousmind how do you get mad at me not explaining notation when @0celo7 literally posts blocks of latex like he just did above
 
12:31 AM
@Obliv ...he posted that to make fun of you :P
 
user218912
lol
 
oh well.. i need more coffee
 
@ACuriousMind I mean...if you actually understand measured Poincare-Neumann-Sobolev inequalities in geodesic balls, I'd appreciate an explanation lol
 
@0celo7 Just the basics
 
@ACuriousMind Well I figured out why grad level complex analysis (papa Rudin) is a prerequisite for the functional analysis sequence
 
12:34 AM
@ACuriousMind the function $T$ is defined such that the domain is still a subset of $\mathbb{R}^2$ but the image is transformed by $(x,y) \to (u(x,y),v(x,y))$ as you stated above. I think you knew what i was talking about but you are teaching me to be specific anyway, right?
 
Our former department head taught FA and operator theory using his own books (GTM something, GSM something), and they're full of complex analysis!
I don't really care for complex analysis though, so I might be screwed
He also wrote two books on complex analysis
I might get one this summer
 
@Obliv I think you're trying to ask about what's called the Transformationsformel in German, but I still don't understand what your T is, sorry
 
@ACuriousMind Literally, the change of variables formula in English.
 
@0celo7 wat
@0celo7 Heh, so simple sometimes...
 
@ACuriousMind John Conway (not the algebraist)
He used to be the math department head at my school
His functional analysis book says that functional analysis is best understood applied to complex analysis
Consequently, the course catalog says that complex analysis is a prerequisite for functional analysis
I guess no one changed it since he left
Now the prof who teaches it just does C* algebras
 
12:37 AM
The relevant question is whether they still teach from his books or not, right?
 
@ACuriousMind My advisor taught FA one semester and just did Sobolev spaces, you would have loved it
 
I don't think there's a standard curriculum, it's what the grad students want
+ what the prof wants to teach
 
@acuriousmind how would I phrase it correctly? $\int \int_E f(x,y) dA$ where $E$ is the domain of $f(x,y): \mathbb{R}^2 \to \mathbb{R}$ becomes $\int \int_L T(u(x,y),v(x,y))\mid J \mid ~dv~du$ where $T(u,v): \mathbb{R}^2 \to \mathbb{R}$ and $E,L \subset \mathbb{R^2}$ and $\mid J \mid = \frac{\partial(u,v)}{\partial(x,y)}$
no mention of the word transformation anywhere. Strict definitions. Beat that
 
You still haven't defined $T$.
 
12:40 AM
@ACuriousMind What's your guess for the highest number of GTMs written by one person?
Not including multiple editions
 
@0celo7 not the slightest idea
 
The six editions of Classical Fourier Analysis would win
what the hell does that mean?
 
oh i'm stupid
 
@ACuriousMind My guess would be 3
Lee and Conway are tied probably
 
wait actually I have no idea what you're asking for @acuriousmind how can I define $T$ while keeping it generalized?
 
12:43 AM
WHAT IS T SUPPOSED TO BE
 
@0celo7 A MAP OF R^2 TO R ON A DIFFERENT DOMAIN DEFINED BY THE FUNCTIONS U(X,Y) AND V(X,Y)
I LIKE TALKING IN CAPS TOO.
oh crap i think it should just be $f(u(x,y),v(x,y))$ no need for $T$
 
Who even knows how to prove change of variables for the Riemann integral
@ACuriousMind Did you do it back in the day?
 
@Obliv I have no idea what $T$ is supposed to be! I would write the change of variables formula as $\int_L f(u,v)\mathrm{d}u\mathrm{d}v = \int_E f(u(x,y),v(x,y)) \det(D \phi) \mathrm{d}x\mathrm{d}y$ where $\det(D\phi)$ your Jacobian determinant.
@0celo7 Only for the Lebesgue integral, and it was not elegant.
 
that is correct @acuriousmind it was simply a test of your knowledge, in the end.
 
@Obliv Try: Let $U,V\subset\Bbb R^n$ be open, $\phi:U\to V$ a diffeomorphism. A function $f:V\to\bar{\Bbb R}$ is integrable precisely when $(f\circ\phi)|\det D\phi|$ is integrable over $U$, and in this case one has $$\int_{\phi(U)}f=\int_U (f\circ\phi)|\det D\phi|.$$
@ACuriousMind It's apparently worse for the Riemann integral.
In fact, when Fleming was writing his analysis book, he just wanted to do Riemann because it's an undergrad book. But he found proving change of variables for Riemann was too hard, so he decided to just do Lebesgue integration
The extra effort for that was less than coming up with the necessary epsilons and deltas for the Riemannian proof
 
12:53 AM
brb going home
 
@ACuriousMind Umm, my QM homework wants me to prove that the spin-1 matrices form an irrep
Isn't that PhD level algebra?
 
@0celo7 No. Take the three spin-1 matrices $J_1,J_2,J_3$ and show that for any vector $v$, the span of $v,J_1v,J_2v,J_3v$ is three-dimensional, i.e. the entire space. Therefore, this representation doesn't have invariant proper subspaces.
 
What?
I didn't understand a word of that.
@ACuriousMind How do I show that such a span is 3-dimensional?
 
1. A representation of an algebra $\mathfrak{g}$ is irreducible iff it doesn't contain proper invariant subspaces. 2. If $(V,\rho)$ contains an invariant subspace $W$, then for $v\in W$, $\rho(\mathfrak{g})v$ is also an invariant subspace of $W$. 3. If you can show that $\rho(\mathfrak{g})v = V$ for all $v\in V$, then there are no invariant proper subspaces.
@0celo7 Write it in terms of the basis vectors upon which you know the action of the $J_i$ (or linear combinations of them) and construct three linearly independent vectors of the form $v + \alpha J_1 v + \beta J_2 v + \gamma J_3 v$.
 
1. Yes, I know the definition. 2. Sure, $\rho(\mathfrak g)\rho(\mathfrak g)v=\rho(\mathfrak g)v$. 3. How do I do that?
@ACuriousMind I don't know how to show such things are linearly independent...
How do I even find these things?
 
1:06 AM
In this case, you could simply try to show that the three standard basis vectors lie in that span.
Play with the raising/lowering operators, that sort of thing
 
There's literally infinite things I could try
How am I supposed to do that?
 
@0celo7 That's the case for every exercise :P
 
@ACuriousMind you must think I'm joking
but I really have no clue how to do this
@ACuriousMind I'm having a hard time understanding why capacitors in series have the same charge
my prof used some weird argument with Gaussian surfaces, but it went over my head
so if we put a Gaussian surface around the inner parts of the capacitors...so what?
why does the enclosed charge have to be zero?
 
1:29 AM
@ACuriousMind Apparently there's something called an electrometer that measures voltage without passing a current.
HOW
 
2:08 AM
@0celo7 A leaf electroscope does that. There is actually a tiny current in the device, but it is motion of the conduction electrons in the metal of the device.
You can also build one out of a FET with it's base floating.
 
They're pretty old it seems
 
I think the reference voltage in that case is the negative terminal of the battery.
 
user218912
qft is getting hard
 
@BernardMeurer this phone is pretty cool
 
@0celo7 Good
 
2:15 AM
can it run Linux?
 
Spec-wise? Yes
There are no drivers for the hardware on it though
Although, theoretically, once a jailbreak is possible you could port something like Bochs to it's architecture (AARCH64 I'm guessing?) And then emulate Linux on Bochs
That will be slow though
Does AARCH64 imply Hypervisor support? If so you could minimize the iOS kernel and run Linux on Ring -1 as a VM
That will be very hard though
 
get me Linux on this phone
 
Specially because I'm not certain on how minimal you can make the iOS kernel, i.e. how many processes you can kill before it goes crazy
You can kill the SpringBoard though, that's for sure and that must be where most of the RAM goes to
 
The home button isn't a button
 
@0celo7 Get me a phone to test it with lol
 
2:18 AM
it's a touch pad that like vibrates the phone and simulates a button press
it's freaky
 
It's like the new macbook's trackpad
 
I think Bob is getting one
although the dog is sick and they had to spend an outrageous amount of money to fix him
 
I would really like to have that sort of money
 
and he's still not drinking anything :/
 
The dog or Bob?
 
2:19 AM
the dog
Bob has plenty of whiskey
 
Oh, that sucks man :/
I still need to have some Whiskey with him one day
 
dog randomly started pooping blood
doctors dunno what was wrong
 
Damn, how old is he?
 
8?
damn, he's an old dog
 
That's not that old
 
2:22 AM
no, but he's my baby :(
 
Isn't your cat your baby?
Einstein
 
the cat is my One True Heir
 
the dog is eating now, but he won't drink anything
 
That's impossible though
 
2:24 AM
so they're gonna take him back to the vet for an IV if he doesn't drink tomorrow
@BernardMeurer the dog is pretty stupid apparently
 
Ah, yeah that's probably a wise decision
Nuit was a pretty smart dog
She would bark at me, knowing I was a bad influence on the house
 
Nuit loves me
I'm her third favorite person
little pig dog
 
Ron told her to kill me but she didn't even move, so I guess that is a form of love?
 
she's blind and didn't feel comfortable attacking
back in the day you'd be toast
 
Lol, now I'm remembering the day I met Ron, the man was not in his best shape
Had an accident playing football
 
2:26 AM
he's like 50, should not be playing football
*46
 
He like, drove in the, what do you call the patch of road in front of the house/garage?
driveway?
 
sure
 
He arrived as me and Michelle were opening the door, drove on the driveway, then backed up, then drove up, then backed again, then drove up and left the car
And then when we asked "dude, what was that?"
He was like "What?"
It was pretty funny
Well bearded man I must say though
 
he looks like a wild man
 
Man, I really, really want Civ VI
 
2:31 AM
how expensive is it?
 
It's like 50 bucks right now, I'm waiting for it to get cheaper. I don't get why they charge so much for this crap
Christimas sales will be my time of action
 
because it's a good game
so do you need analysis help or not?
 
Explain me derivatives
 
do you want Banach spaces?
 
Not now :p
 
2:34 AM
what is unclear?
 
Everything. All the chub n' tuck did was write a bunch of definitions and theorems on the board and I was like wat
wat is dat
That's definitely problematic
Lol, the fear is real huh mate?
Don't ever dump that girl dude
 
She's pretty good
 
She's super nice, you'll have a very hard time finding a better one
 
Ok, derivatives
Do you understand limits?
 
She wasn't even mad when she saw I was better at bottle flute than she ever will be
@0celo7 I think I do, a limit is where $f(A_n)$ converges to when $A_n\rightarrow x$
 
2:40 AM
what is that notation...
 
I'm saying if I have a sequence $A_n$ that converges to $x\in\overline{\mathbb R}$
 
use $x_n$
$x_n\to x$
that makes some sense
unlike what you're doing
 
$\overline{\mathbb R} = \mathbb R \cup \{-\infty, +\infty\}$
Okay, if when $x_n\rightarrow x$, $f(x_n)\rightarrow y$ then y is a limit
 
@BernardMeurer god, not this again
 
2:44 AM
@SwapnilDas How does one pronounce your name?
 
@Mew your site looks a lot better now
 
@SwapnilDas link?
 
Sure
 
@BernardMeurer for all such sequences, this must be true
 
Nil is the same as English nil. Swap could be pronounced like swup.
 
2:46 AM
@0celo7 So $\forall x_n \rightarrow x; f(x_n)\rightarrow y$, only then $y$ is a limit?
 
@BernardMeurer Ok, now prove it is equivalent to the other definition using epsilon delta
no notes
do it
 
@0celo7 "Other definition?"
 
$\lim_{x\to c}f(x)=L\Leftrightarrow (\forall \epsilon>0)(\exists\delta>0)(\forall x\in A)(|x-c|<\delta):|f(x)-L|<\epsilon$.
$f:A\to\Bbb R$
 
*$\overline{\mathbb R}$
 
What?
 
2:49 AM
It's valid for $\overline{\mathbb R}$ too
I have no idea how to even begin to prove this
 
Explain to me what $f(x)-\infty$ is.
 
It's $-\infty$
 
How is $|-\infty|$ ever $<\epsilon$?
 
unless $f(x) = +\infty$
because $+\infty-\infty$ breaks maths
@0celo7 Okay, let's just use $\mathbb R$
 
@0celo7 Are you a physics, math or engineering graduate?
 
2:51 AM
I have no clue how to prove that
 
@SwapnilDas Last two
 
@SwapnilDas He's graduating in visual arts
 
@BernardMeurer Did he not talk about this in class?
 
Don't think so
 
OK, thanks.
 
2:51 AM
I am actually a porno producer
@BernardMeurer o.O
Have you not done anything with epsilons and deltas?
 
@0celo7 Oh, you meant that definition? Yes we did that, but no proving as far as I remember
yes, the epsilons and deltas are a meme among the students already
 
@BernardMeurer Nonsense
 
"What time is it?" "For every delta greater than 0 there is an epsilon that says it's 2:30 PM"
 
You must prove it
 
Does anyone agree with the fact: rigor cleans the window through which intuition shines.
 
2:54 AM
@BernardMeurer You pick epsilon first, then delta...
@SwapnilDas Nope
 
@0celo7 Yeah, yeah, you get the joke
@SwapnilDas No, cleans the window
 
I think there's a translation error there
 
@0celo7 No. Idea. How. To. Prove. That
 
Sorry.
 
@BernardMeurer Ok. Think about $\Leftarrow$ first.
That direction is the "easy" one
The other one is tricky
 
2:56 AM
$$\lim_{x\to c}f(x)=L\Leftarrow (\forall \epsilon>0)(\exists\delta>0)(\forall x\in A)(|x-c|<\delta):|f(x)-L|<\epsilon$$
Okay, let me think
For all eps greater than 0
There will be a delta greater than 0
such that for all x in A
What's A?
 
$f:A\to\Bbb R$
 
Ah
What's that called in english?
 
domain
 
and $\mathbb R$ is the counterdomain?
 
codomain
 
2:58 AM
Got it
Okay, so such that for all x in the domain (A)
$(|x-c|<\delta)$ implies $|f(x)-L|<\epsilon$
 

« first day (2190 days earlier)      last day (2731 days later) »