« first day (2274 days earlier)      last day (3042 days later) » 

13:14
@Danu Great talk
Hello
Any ideas how to solve this thing?
https://goo.gl/J6kQ7X
@SuperMan you know, you can write it here
I am not sure how to open |3^x - 1|
Alright, I'm back.
@SuperMan you don't have to
you just need to separate into cases
13:18
hmm
little hint?
well
case one is $3^x-1 < 0$
case two is $3^x-1 \ge 0$
Hi @Balarka :)
Do you know how to show that $U(n)\simeq Sp(2n,\Bbb R)$?
I was thinking of trying to write down a homotopy that explicitly cancels the commutator with $J_0$
13:29
Is there a short hand easy way to do modular inverse?
like $2^-1 mod 5$ for example?
@DemCodeLines try from 1, adding 5 at a time, until it is divisible by 2
1 + 5 = 6, which is divisible by 2 (3).
nice
@Danu I am a bit uncomfortable with this group, but U(n) sits inside Sp(2n, R) in a natural way, doesn't it? Look at the special orthogonal matrices in Sp(2n, R), eg
@DemCodeLines but this algorithm is faster:
In arithmetic and computer programming, the extended Euclidean algorithm is an extension to the Euclidean algorithm, which computes, besides the greatest common divisor of integers a and b, the coefficients of Bézout's identity, that is integers x and y such that a x + b y = gcd ( a , b ) . {\displaystyle ax+by=\gcd(a,b).} This is a certifying algorithm, because the gcd is the only number that can simultaneously satisfy this equation and divide the inputs. It...
13:36
@Danu See page 225 of Arnold.
I think it's $\mathrm{U}(n)=\mathrm{O}(2n)\cap\mathrm{Sp}(2n)$
Huh, ok.
If that's so then a deformation retraction should be given by Gram-Schmidt
Oh, he means a def ret.
He means homotopy equivalence.
So the options are...Morse theory or writing down the thing explicitly, right?
Or is there some notion that intersecting by $\mathrm{O}(2n)$ does not change homotopy type?
I just said Gram-Schmidt - that's probably how you'd do it. I don't know what Morse theory means.
13:41
You could maybe show they're homotopy equivalent to isomorphic CW complexes.
"isomorphic"?
Equal, whatever.
I don't know how you plan to show that.
I don't plan to show it.
Then why say it?
13:42
Morse theory has proven useful in the study of classical groups.
13:53
@BalarkaSen Good point.
Nothing like that necessary @0celo7.
By the way it's $U(n)=O(2n)\cap Sp(2n)=GL(n,\Bbb C)\cap Sp(2n)=O(2n)\cap GL(n,\Bbb C)$
What I was doing @Balarka was trying to go to the intersection with $GL(n,\Bbb C)$. But $O$ is much better.
Ah
Even then, GL(n, C) deformation retracts to U(n)
Oh, you were trying to retract stuff in Sp(2n) to GL(n, C) \cap Sp(2n). Yeah, that may require a little more effort to write down
14:22
@Danu Yeah I can read the book.
14:33
@Danu do you already know that Sp cap GL(n,C) is Sp cap O?
14:47
@Danu In any case, $\text{Sp}(2n)$ acts transitively on the space of positive-definite symmetric matrices by $g \mapsto A^{-1}gA$. The stabilizer of the identity matrix is $\text{Sp}(2n) \cap O(2n) = U(n)$, so we have a fiber sequence $U(n) \to \text{Sp}(2n) \to \Bbb R^{n(n+1)/2}$, since the space of positive-definite symmetric matrices is convex.
Requiring ring theory genius:
0
Q: *bump* How do I construct this finite non-commutative ring? A formal language string ring.

Enjoys MathLet $S$ be a formal language string over $\Sigma$, and $S_{\leqslant}$ the substring lattice. If $t$ is a substring of $S$, we write $t \leqslant S$. Define a substring ring of a string $S$ to be a unital ring $R_S$ together with a surjective map $\phi : R_S \to S_{\leqslant}$ such that: For a...

Then a deformation retraction of the last space to the identity matrix gives a deformation retraction of $\text{Sp}(2n)$ to $U(n)$.
I posted last night, with not the greatest help
It's intesresting to formal language grammar people
should be
@MikeMiller Ah, thanks.
@MikeMiller I'm asking in essentially the context you outline
"Met" is the notation the lecturer uses for pos. def. symm. (scalar products)
good morning
how is everyone
14:54
Hungry
im sick :{
home from school atm
@Danu Mm?
@user3502615 mathematics shall heal you XD
@MikeMiller This story about transitive actions and stuff
14:56
It's exactly what the symp. geom. lecture was about today
So it's nice that you answered in that context.
dihydrogen monoxide?
@user3502615 yes
i need to get a dihydrogen monoxide detector, to protect my kids
But it seems like that's exactly what's written there, so I'm not sure what I did for you.
14:57
that is, if i had kids
@MikeMiller I was trying to understand why that diagram is right
what are some recommended prerequisites for a first course in homological algebra?
Regular algebra? :D
is that is?
There is a nice set of notes online, by Ash, called "graduate algebra, a one year course" or something like that
It contains all the prep material, I think
15:01
my algebra book covered group theory, isomorphisms, and a bit of galois theory
I think that's it, yeah. Except for the usual mathematical maturity
is that enough?
I don't know.
btw, the nth roots of unity form a group under multiplication, right?
@user3502615 yes
the identity element is $1$
15:06
welp
now its time to prove some isomorphisms, baby
@user3502615 which isomorphisms?
I think I'd probably find homological algebra largely unmotivated unless I encountered the various constructions in homological algebra before in topology.
im interested in taking algebraic topology
I wouldn't study it for its own sake either... But I know some people who don't like the topology and love homological algebra.
should i just take general topology first?
15:07
Don't yell at me, but what does homological algebra do if not directed at algebraic topology?
@0celo7 It's largely useful in algebraic geometry.
@DHMO just some exercises in my book
i have to prove that the $n$-th roots of unity are isomorphic to $\mathbb{Z}_n$
One can probably think of many constructions in homological algebra as analogues of the constructions in topology for affine schemes.
@Mahmoud wow!
that's neat
@BalarkaSen What is an affine scheme?
@0celo7 A picture (topological space, if you like, but a little more than that) associated to a commutative ring.
15:12
So like a sheaf, but with a c. ring instead of an abelian group?
Well, shit. My PDF viewer has crashed 3 times now
@user3502615 Well, the function would just be $f_n(r) = e^{2ir\pi/n}$
@0celo7 I can't really parse that.
Parse what?
Apparently this one PDF is 1.5GB, no wonder it crashes my reader.
isomorphism is an equivalence relation, right?
@DHMO i know :P
hey @0celo7
15:18
@Balarka A lot of the constructions in homological algebra turn out to be useful but are hard to have imagined first in topology. At least, I think so.
@BalarkaSen A (pre)sheaf is a functor from the open sets to Ab, so is a scheme a functor from open sets to cRng?
@MikeMiller True, I suppose.
@0celo7 No.
i am gonna take homological algebra next year
@Adeek hi
are most of the people in mathoverflow pursuing a phd?
or do you need to have one
15:22
most of the people in mathoverflow already have a phd, I believe
@BalarkaSen I want to verify something with you because my brain thinks they are the same thing. So suppose we have groups $G_1, ..., G_n$ and normal subgroups $H_1 ,...,H_n$. Elements of $(G_1 \times ... \times G_n) / (H_1 \times ... \times H_n)$ are of the form $(g_1,...,g_n)(h_1,...,h_n)$ right ? While elements of $(G_1/H_1) \times ... \times (G_n/H_n)$ are of the form $(g_1h_1,...,g_nh_n)$ ?
Huh?
Elements of $G/H$ are of the form $g \pmod H$
nvm. Elements of $(G_1 \times .. \times G_n)/(H_1 \times ... \times H_n)$ are of the form $(g_1,...,g_n)(H_1,...,H_n)$
...why $\times$ on one, and $x$ on the other?
yeah I can see the distinction now. It is amazing how silly mistake I make when I am tired.
@0celo7 yeah I edited it.
15:28
@Adeek By $(H_1, \cdots, H_n)$ you really mean $H_1 \times \cdots \times H_n$.
Mind reader Balarka
yeah
-1
A: What's the point of Pauli's Exclusion Principle if time and space are continuous?

user363984if they want people can get really mad for what i'm about to say so i will SHOUT IT LOGIC AND MATH/MATH FRAMEWORKS ARE JUST CONCEPTUAL AND SIMBOLIC LIMITS TO OUR MIND!! THE UNEXPLAINED AND THE CEASE OF OXTILITIES AGAINST UNCOMMON AND "ABNORMAL THINGS" IS THE ONLY WAY FOR EXISTENTIAL FREEDOM!! A ...

what notation is used to denote the set of nth roots of unity?
$\phi_n$
15:32
well, thats weird
because im using $\phi$ to denote the isomorphism :{
I don't think there is one.
can i just say "Let $\Upsilon_n$ denote the set of n-th roots of unity"?
any stats people recognise this distrubution? i.imgur.com/WNeqPRh.png
@user3502615 no one uses upsilon for anything, so sure.
15:37
alright :P
the only worse off letter is omicron
thanks @DHMO even if you are a bot :)
lol
@DHMO doesn't get to be a person
@Lembik what's the difference between a bot and a person?
energy?
15:47
@0celo7 ow
@DHMO Poisson, maybe?
@Semiclassical but they start at 0
eh. then flip it.
A hash table has space for 75 records, then the probability of collision before the table is 6%full?

http://math.stackexchange.com/questions/1984343/a-hash-table-has-space-for-75-records-then-the-probability-of-collision-befor
Help me solve this:
https://www.wolframalpha.com/input/?i=log2(x)%3D3-x
i mean, if you plotted that with the horizontal axis flipped, it'd look pretty poisson I think
i.e. $-X$ is Poisson-distributed, not $X$ itself
I forget which way a Poisson distribution tilts, though
15:51
Hi @iwriteonbananas
What the fu*k is up, Balarka
Dude, language
Still doing hydrostatics, balarka?
wow
15:53
@Semiclassical Not anymore.
We did thermodynamics the day before.
did you try that problem I suggested?
@Balarka Is the second law of thermodynamics valid?
Yes.
Full stop.
15:54
@Semiclassical We'll discuss Archimedes the next day, so not yet.
ah, okay
who asked you? ;D
Has anyone verified the second law in full quantum gravity yet?
Haven't encountered the 2nd law yet.
15:54
Has anyone verified anything in full quantum gravity yet?
only limiting cases lol
Yeah
on the other hand, I know that one of the places where people have had some results (however limited) is re: black hole thermodynamics
And they do talk about the entropy of a black hole, so presumably there's a version of the second law there
@Semiclassical All we know there is that string theory agrees with semiclassical QG
But ST might be wrong
Sounds right.
so it's all garbage
3
15:56
lol
Well, it's physics.
So that comment is redundant.
so, what's entropy? (I know as far as the 1st law)
@Semiclassical :)
To the extent that it's all a grand thought experiment for now, it's worthwhile as an exercise in imagination and reasoning
to the extent that people forget that first part, though, it probably is garbage
when it comes to the 2nd law my sympathies tend to be with Eddington when he said
16:00
I hoped someone would teach me about entropy when I made that comment, if it's unclear.
"The law that entropy always increases holds, I think, the supreme position among the laws of Nature. If someone points out to you that your pet theory of the universe is in disagreement with Maxwell's equations — then so much the worse for Maxwell's equations.
If it is found to be contradicted by observation — well, these experimentalists do bungle things sometimes. But if your theory is found to be against the second law of thermodynamics I can give you no hope; there is nothing for it but to collapse in deepest humiliation.""
@Balarka We know
@Semiclassical ...why do we assume stat mech to hold in full QG?
ask people doing QG
We can have weird things like monopoles that are forbidden classically but we expect them quantumly
16:03
@MikeMiller "But we don't care"? :P
So why is the second law sacred?
I don't have any understanding of which principles people doing QG actually consider sacrosanct versus naught
did I enter the physics chatroom by mistake?
Yes.

 The h Bar

General chat for Physics SE (physics.stackexchange.com). For M...
^ math chat
ah, thanks :P
16:06
lol
the conversations in h Bar was mostly either about 0celo-math, election or JD-physics the last time I talked there.
JD has been banned for a year
and we're not allowed to talk about him
(and the first and third are vastly different than math and physics, to anyone who's wondering)
Not sure why 0celo-math is not math, but ok.
@Semiclassical 1500 word lab report! Still working on the second to last section
It's math mixed in with refutation of classical logic, complaints about proofs not being rigorous and "it's in Lee, page [blah]".
16:11
I saw that there's a new paper by Taubes out today @Mike
Well I don't understand how in analysis I must give careful $<\epsilon$ arguments but in alg. top I'm supposed to "picture" gluings and shit.
Me too. I looked at it and then moved on.
Also it's not my fault Lee has everything.
Makes me feel so good about myself ;D
@Danu Is this paper seriously not in TeX?
@Danu Ok, I don't have the intuition. What do you want me to do?
16:15
Stop complaining about books that assume you have the intuition
It's a problem only with algebraic topology.
So how does one get the intuition?
For me? Take a course.
(is the easiest)
Hatcher assumes the intuition already and supposedly it's an easy book.
It can get a lot harder than that, sure! But it's not "easy" in the sense of "first year undergrad should have no problem"
Balarka is 15 and he has no problem.
16:18
@0celo7 did you try topological manifolds by john lee ?
I think that would be nice book to read before hatcher
Yes, it's much better
But I don't have time to do all of this
haha
When I read Lee it all makes sense, but Hatcher/Bredon still seems like nonsense.
did you solve the problems ?
No, I don't have hours and hours.
16:21
Take a course
what are the applications of commutative algebra?
I don't have hours and hours.
Eyo @TedShifrin
that is, what courses would it be useful to know it for
Heyo @Danu
16:23
balarka is 15?
@user3502615 Algebraic geometry and algebraic number theory, mostly.
Hi @Ted
Greetings @Balarka
Learning connections right now.
Hi chat, I have a differential topology curiosity, although I admit I don't fully understand the question itself:
Is the deRahm isomorphism on relative cohomology multiplicative?
16:27
@BalarkaSen is that true?
@s.harp Can you write that out in more detail?
Yes, and be careful even with the definition of deRham relative cohomology.
hi @Ted
Hi @Alessandro
@user3502615 Nah.
16:28
It used to be true.
Yeah... what is relative cohomology even, in de Rham context?
@BalarkaSen good, you were making me seem self-concious for a second
@Danu There are definitions. Usually you want to work relative to a submanifold.
If $A\subset X$, it's important to decide if you look at $k$-forms on $X$ that vanish (as forms on $X$) along $A$ or that vanish when restricted (pulled back) to $A$.
16:29
@DAnu, @Ted (this is relevant to the second space being a boundary) as far as I heard you take the differential forms that are zero on the boundary. This is apparently the same as if you took the ones that are zero on the boundary with strong regularity conditions for derivatives being zero also
What do you mean by "zero on the boundary"? See what I just wrote.
@Mike I don't know if I can formulate it more concretely, it is just a curiosity that I dont really understand, however the formulation apparently was:
$\rho_{rel}: \Omega^*(X,L)\to C^*(C,L)$ induces a multiplicative map on cohomology
It should just be restriction for the SES of complexes to work.
What's the difference between $\Omega^*$ and $C^*$?
@Ted I'm afraid I don't know which of the two it was
That's why I bring it up. :)
16:32
$C^*$ is a cochain complex from a CW structure for example
Hello
@s.harp There is no such homomorphism at the level of chains.
I need hint to solve this equation
I would be really grateful if you help me with it
The de Rham chain complex is special. It has the structure of a commutative differential graded algebra. This is NOT true of any of the other standard cochain complees.
16:33
@Mike then I have seriously misunderstood the talk I just heard, oops
Who talked?
a friend who is about to finish his master thesis
LOL ...
@SuperMan: What's your issue with it?
yeah I don't know so much about differential topology^
I do not know how to find the 'x'
@TedShifrin
16:35
@SuperMan evaluate the expression on the right first
log3(81) = 4
subtract 1 from each side
sure
x log3(x) = 3
Right ... Exactly what I was going to say. But don't do it for him, @user3502615.
solve from there :]
@TedShifrin don't intend to.
You cannot solve it algebraically, however. You just have to make an educated guess. I don't like problems like this.
16:37
I can't subtract 1 form left side
Wait. @user3502615: How did subtracting $1$ turn into subtracting $\log_3(x)$?
why is that?
Right @Superman
I recall relative de Rham being in Bott & Tu.
sorry about that
thought that said (x+1)+log3(x)
anyways
16:38
yes
@TedShifrin Take a great circle on a sphere, delete one hemisphere separated by the curve. Then the normal field of the other hemisphere along the great circle is parallel because all of them are parallel in R^3, yes? (So derivative is 0).
My comment still stands @SuperMan. There is no algebraic manipulation to do. You have to make an intelligent guess.
so you have (x+1) log3(x) = 4
yes @user3502615
@Balarka: It works only because the constant vector field is everywhere tangent to the sphere along the curve. Yes.
16:39
Yes, for sure.
And you can deduce that the vector field tangent to the great circle is likewise parallel because of the constant angle. On the other hand, you can see it directly (which is what I like to emphasize geometrically) from the "physics" of constant speed circular motion.
@0celo7 They do a generalized version, relative along a map. But yes.
I don't know a good way to "feel" that it's parallel. Of course, the directional derivative is normal to the curve, but...
BTW, @Balarka, I hope you enjoy my discussion of the Foucault pendulum :)
NOT normal to the curve, BUT normal to the sphere. Hence an ant living in the sphere sees no change.
Normal to the sphere, sorry.
You're right.
It's all about what a resident of the surface sees.
16:43
Right, because the component of the change of that vector field on the tangent space vanishes, the ant wouldn't see anything.
Right.
That's a fair intuition.
Thanks
You'll see discussion later about why you rotate uphill or downhill when you parallel translate along a general line of latitude.
I don't know how to multiply matrices.
What the hell is wrong with me
Go learn, then.
16:45
Too late now, I failed my QM test
rip :[ study more?
I doubt one messed up matrix multiplication causes a fail, but who knows.
Maybe he messed it up in every calculation?
what is the better alternative to LaTeX's "eqnarray"
environment
What does better alternative mean?
I use align a lot ...
16:46
eqnarray keeps fucking up for me
Guess you did not honor the LaTeX gods appropriately.
$e^x$ is a surjection, correct?
I just messed up the whole exam.
i mean,
@user3502615 Misses 0
16:53
@0celo7 how was the exam
hmm
does $\mathbb{Z}_n$ include $n$?
@Adeek I did literally nothing correctly.

« first day (2274 days earlier)      last day (3042 days later) »