« first day (2727 days earlier)      last day (2285 days later) » 

12:52 AM
$\newcommand{\im}{\operatorname{im}}$Theorem: the dual functor on the category of vector spaces over $K$ is exact.

Proof:
Let $U \overset f \longrightarrow V \overset g \longrightarrow W$ be an exact sequence, i.e. $\im f = \ker g$.
Let $f^* : v^* \mapsto v^* \circ f$ and $g^* : w^* \mapsto w^* \circ g$.
To prove that $W^* \overset {g^*} \longrightarrow V^* \overset {f^*} \longrightarrow U^*$ is exact, i.e. $\im g^* = \ker f^*$.

Let $v^* \in V^*$ such that $v^* \in \im g^*$, i.e. there is $w^* \in W^*$ such that $w^* \circ g = v^*$. Then, for every $v \in \ker g$, $v^*(v) = w^*(g(v)) = 0$
 
1:08 AM
Anyone alive?
 
All I know is that I exist @Faust
 
Exist huh.
 
im having a bit of trouble focusing on hw
 
Why is that?
 
1:13 AM
need to learn some linear algebra
cause i do math 80hrs a week and eventually it gets kind of boring
 
Switch it up. Do 80 hours of boring a week, and it will eventually get math.
 
lol
 
Or drink more coffee
 
im trying to find some video lectures relating to a second course on linear algerbra i think i can manage watching something but i cant seem to find what im looking for
 
Dude I was having the same thing earlier today with my modern physics course. I just couldn't handle the book, so I turned to youtube. But I got frustrated with youtube so I went back to the book :P
 
1:19 AM
lol
 
Can anyone recommend me a good PDE theory playlist online?
 
no, but if you find one ping it to me
 
Anything from OCW looks like it was recorded around 100 BC
 
Do they even have anything for PDE anyway?
 
its wierd alot of math videos are really really terrible quality
 
1:21 AM
true
 
like 320x240 is not a video format
how do u even find a camera that will record in that
 
3B1B has a ten-episode series on linear algebra
 
You're basically watching 4 or 5 pixels draw lines on a chalkboard
 
@AkivaWeinberger is it advanced or intro?
 
Intro, probably not what you're looking for
 
1:23 AM
yeah
 
ted has a good lecture series on intro to linear algerbra
 
yeah, but I need graduate stuff
nvm
 
oh i know a pdes course of video lectures one sec
 
There's this Indian guy with Khan Academy-like video lectures on math, he might have stuff
 
Thank you
 
Doesn't look like he has anything relevant actually
Still a good resource, lots of stuff there
 
@LeakyNun we are not even talking about that and your remark is irrelevant here. No I said your remark that "I am illogical" is rude. You flat out called irrational. It is rude to more or less call someone insane.
 
show me when I said that you are illogical
 
TIL that famous and well-received architects are called "starchitects"
 
2:08 AM
The proof that $\sqrt2$ is irrational at 2:40 in this video is pretty cool
(The rest of the video is shit)
 
@AkivaWeinberger nice
 
So you can repeat that construction infinitely many times and have an infinite descending sequence of squares with integer sides
(He frames it in a "minimal criminal" sort of deal but it's essentially the same)
 
2:31 AM
I'm trying to overkill-define a palindrome number. Is this sufficient?
Let $q$ $\in$ $\mathbb{N}$. Let $(d_{1},d_{2},\cdots,d_{n})$ be the ordered $n$-tuple of integers such that $10^{n}d_{1} + 10^{n-1}d_{2} + \cdots + 10d_{2} + d_{n} = q$.
If $(d_{n},d_{n-1},\cdots,d_{1}) = (d_{1},d_{2},\cdots,d_{n})$, then $q$ is a pallindrome.
 
$q$ is a palindrome of length $n$ iff there exists an $n$-degree polynomial $f$ with coefficients between $0$ and $9$ such that $f(10)=q$ and such that $f(x)=x^nf(1/x)$
 
Nice, I've never seen that before!
 
Yeah, $x^nf(1/x)$ reverses a polynomial
As a consequence, its roots will be the reciprocals of the roots of the original polynomial.
 
Now the question is: Do I use this in homework and risk annoying my professor? :P
 
Maybe put little hand-drawn emojis next to it :P
(Or legit emojis if this is typed)
 
2:38 AM
Perfect idea!
\laughingface
 
Whoa: Apparently the word "emoji" is unrelated to "emotion" etymologically
(Contrast "emoticon", which is)
It comes from the Japanese 絵文字, which is 絵 (e, "picture") +‎ 文字 (moji, "character")
("Character" in the sense of, like, a written symbol)
 
That's an eerie similarity.
 
hi chat
 
It's almost as if whoever "invented" Japanese and English got together to leave easter eggs for for humans down the road.
 
In defining a set with LUB property
say E subset of S
S has LUB property if for each E ,non empty, then E has a SUP in S
the two cases are the SUP is in both E and S , and the other is the SUP is not in E right?
 
2:49 AM
@MatheinBoulomenos
 
hi
 
@CookieToast There's an extinct language once spoken by an Australian Aboriginal tribe called Mbaram, in which the word for "dog" is... dog
 
Haha I've heard that one before. Spooky
 
It came from gudaga
So it's a cognate with guda (also meaning "dog") in a neighboring language
 
hey akiva
 
2:57 AM
Hey
 
are you going to do amc 12?
 
Oh, yeah, I forgot about it
Yeah, I will
When is it?
 
i think like second week of february
feb 7th
i didnt do good at all last year
 
@Jacksoja Yeah, the supremum is in S no matter what (that's what it means when we say that S has the Least Upper Bound property), but it may or may not be in E.
For example, the set of real numbers has the LUB property. [0,1] contains its supremum, while [0,1) doesn't (in both cases the supremum is 1).
(The interval [0,1) is written as [0,1[ in some countries)
 
for a second i thought you mis-placed parentheses
like the issue with smileys in parentheses (this one :) )
 
3:01 AM
It's weird seeing stuff like ${[0,1[}\cup{]2,3]}$. Like, is $\cup$ inside a pair of brackets?
 
I want to ask a simple question in scheme theory
any takers ?
 
It's worse if it's spaced like $[0,1[\cup]2,3]$, which is LaTeX's default spacing (you have to explicitly tell it not to do that with curly braces)
 
@Adeek you should've said "any schemers?"
:P
 
hehe
 
hi amin in the dark
 
3:04 AM
Hi, Idelhaj in the black
(Blidelhajack?)
Said in the present? "Say"?
 
>dark
>black
>present

One of these things is not like the others :P
Anyway, hey guys! How's it going?
 
Yeah, well, the joke wouldn't work otherwise
 
Tru
 
im just doing some integrals problem
 
I'm good, the chat is a bit devoid of math at the moment
 
3:06 AM
WITHOUT THE FUNDAMENTAL THEOREM OF CALC
 
gasps from the audience
Riemann integrals, I assume
 
yep
 
Riemann sums and such
 
supremum of lower sums / infimum of upper sums
not even a single limit in the chapter
 
Stuff like $\sum n^2\approx\frac13n^3$?
 
3:08 AM
yeah i can show you my proof of that
it was actually $x^3$ and it was ugly
 
Yeah
You don't need the smaller terms though
 
it was like
you partition $[0,a]$ into $P_n = {0, a/n, 2a/n, ..., a}$
 
I feel like you just need to find $c_1$ and $c_2$ such that it's $\frac14n^4+c_1n^3\le\sum n^3\le\frac14n^4+c_2n^3$, which may or may not be easier
$\frac14n^4+O(n^3)$ <-- better way to write the above
 
@AkivaWeinberger thanks Akiva
 
Then
$$L(f, P_N) = \left( \frac{a}{n} \cdot 0^3 \right) + \left( \frac{a}{n} \cdot \frac{a}{n}^3 \right) + \left( \frac{a}{n} \cdot \frac{8a}{n}^3 \right) + \dots$$
$$=\sum_{i=1}^n \left[\frac{a}{n} \cdot \left(\frac{a(i-1)}{n}\right)^3 \right]$$
 
3:13 AM
Er, you probably want to shift over your index there
 
$$ = \frac{a}{n} \sum \left[ \frac{a^3(i-1)^3}{n^3} \right]$$
$$= \frac{a^4}{n^4} \sum (i-1)^3$$
then i looked at the sum of cubes nad found a formula i think is right
$$= \frac{a^4}{n^4} \cdot \frac{(n-1)^2n^2}{4}$$
 
Hm yeah sounds right
 
$$ = \frac{a^4}{4} \cdot \frac{(n-1)^2}{n^2}$$
it's the same for $U(f, P_n$, you get $\frac{a^4}{4} \cdot \frac{(n+1)^2}{n^2}$
 
For general exponents you get a complicated expression using Bernoulli numbers
 
of course as $n \to \infty$ the extra term becomes negligible
 
3:15 AM
But we have $\sum n^3=(\sum n)^2$ which is pretty neat
 
between any two elements of Q , there is an element in R right?
the lectures stated the other way around of this
ie between two elements in R, x,y, x<q<y
q in Q
I don't know how usefull that statment is but it sound true to me
 
@Jacksoja Yes, just average them
Between any two reals, there are infinitely many rationals (and infinitely many irrationals)
 
but is the average in R/Q ?
in R but not in Q
 
The average of two rationals is rational
(Write R\Q for that)
 
yes that is my point
okay thanks, i mix those up
 
3:21 AM
Ah, OK. Yes, it is true that between two elements of Q (in fact, between any two reals) you can find elements of R\Q.
 
so the gaps in Q are very dense
 
The relevant terminology here is dense. Both Q and R\Q are dense in the reals.
Er, yes
 
neat
 
Incidentally, something is called co-dense if its complement is dense
so Q is both dense and codense
 
the universal set being R yeah
 
3:23 AM
Yeah
 
nice thanks :) this was helpfull
 
The Midwest is Illinoying
and Vietname is Hanoying
 
And you are annoying?
:P
jk
 
3:51 AM
3
Q: Are ordinal or cardinal infinities theories for real?

Mozibur UllahThere are a number of notions of infinity in mathematics that are respectable. One of the first is 'the point at infinity' to the line or plane; but one can argue that this is a spurious infinity as in another perspective this simply closes up the line to make a circle; and the plane to make the ...

> Imagining further substitutions beyond that, things become less clear. The first uncountable ordinal ( ω1 ) would be at the "limit of the limit of the limit ..." of this process - if that makes any sense.
actually, its much worse than that, For $\omega_1$ you need:
 
Yeah, any countable series of "limits" wouldn't work
prepares self for the inevitable wall of text
 
$$\omega_1 = \prod_{\alpha < \omega_1} \lim \alpha$$
 
you need an uncountably nested limit, I think...
 
3:53 AM
that kinda reflects how only $\aleph_1$ nets can reach $\omega_1$ and not sequences, that single fact making it a source of many topological counterexamples
 
I need to learn about nets at some point
All I know is "they're like sequences but bigger"
 
Meanwhile, compact sets are really convenient in representing infinities on finite medium like a sheet of paper
 
and turn "sequentially compact" into "compact"
 
I don't really know much about nets other than the stuff given in wikipedia. but it is attractive to me since it is not limited to something of countable length, thus will be very useful when I get to explore topologies that are not first or second countable
 
Makes you glad that we've been able to figure out what the 'correct' definition of "compact" should be
(I am inconsistent with my use of quotation marks)
Y'know, the finite subcover thing.
 
3:59 AM
yup, and one can talk about countably compact by including countable subcovers
 
@Akiva so a directed set is a set with a preorder such that finite subsets are bounded above
 
In a sense, compactness is one of the main reasons why the Jordan curve theorem is even provable at all. It sounds really hard to prove (despite being intuitively obvious), but once you realize that Jordan curves are compact, and compact sets tend to be nice, it sounds a bit easier.
 
A net in $X$ is a map $f:A\to X$ where $A$ is a directed set
 
Not that that's the only thing you need—it's still hard, even from there—but it's a start.
@Daminark Preorder?
Oh, is that the thing with the inclusion maps?
 
A preorder is a partial order but we don't require antisymmetry
 
4:02 AM
So it's like a "less-than-or-equivalent-to" sort of thing?
But partial
 
It's exactly that, the only strange thing is that you can have $a\le b$, $b\le a$, but $a\ne b$
 
$\preceq$
$\precsim$
There we go. $a\precsim b$ and $b\precsim a$ should mean $a\sim b$.
Like, that should define an equivalence class.
 
ah that makes more sense
 
("What if we took the $\le$ symbol and made it wavy")
 
And it reduces to a partial order on those equivalence classes, that makes sense
 
4:05 AM
And if we get rid of the "partial" bit, it's like ranking a bunch of people in a competition in which we allow ties
So finite sets are bounded above… by something not in that finite set?
 
But yeah if you have a space $X$ and an open subset $U$, you say a net $(x_{\alpha})_{\alpha\in A}$ is eventually in $U$ if there's some $\alpha$ such that for $\beta \ge \alpha$, $x_{\beta} \in U$
 
Like, everything always has something above it?
 
The bound may or may not be in the set
So, you see that $\mathbb{N}$ is obv a directed set
Since it's totally ordered
Here's a non-trivial example
 
So $\{1,2,3\}$ is also a directed set?
But $\{1,1'\}$ is not if those elements aren't related, since they have no upper bound
 
Under which ordering? Just $1 < 2 < 3$?
Yeah exactly
 
4:07 AM
but $\{1,1',2\}$ would be, if $1'\prec2$
 
But yeah one example is that you take partitions of an interval
 
@Daminark Yeah
 
Is $sin(x)$ analytic on $\mathbb{R}$?
 
@Daminark Ah, I see
 
Ordering here is containment
 
4:08 AM
and then you have common subdivisions
@orbit-stabilizer Yes
 
Exactly
 
It famously has an infinite series representing it @orbit-stabilizer
$x-\frac1{3!}x^3+\frac1{5!}x^5-\frac1{7!}x^7+\dotsb$
(Cosine is similar but starting with $1-\frac1{2!}x^2+\dotsb$)
@Daminark Kinda sounds like lattices
 
Wasn't sure if that was just the maclauren series, i.e. valid only around some small neighbourhood of 0
 
Nah, it's convergent everywhere.
On any bounded set, the partial sums converge uniformly.
 
Lattices are, in fact, directed
 
4:10 AM
partitions like so?
 
@orbit do you know about radius of convergence?
 
where larger members have more common subdivisions
 
Ah, yeah, you can use the ratio test for that, can't you
I guess a quick argument for it converging everywhere is, "the terms eventually get smaller than that of any geometric series"
 
Radius of convergence is $\infty$
 
4:13 AM
@Secret it's tricky for me to interpret this picture (or pictures in general), but if you have $[a,b]$ I'm defining a partition formally to be $\{t_0,t_1,\ldots,t_n\}$ where $a = t_0 < t_1 < \ldots < t_n = b$
So now it's just set containment, though I do think you have the right idea, it's that the upper bound is a common refinement
 
Unexpectedly, the radius of convergence is $\dfrac1{e^x+1}$ is $\pi$.
 
Right, that makes sense
 
@Akiva it becomes less unexpected if you think of complex analysis
 
Swapping $x$ with $z$, you see there's a singularity at $i\pi$
 
4:14 AM
Yup.
And power series, in the complex plane, always converge in circle-shaped regions
 
Yee
 
I sometimes wonder whether the quaternions contains the complex numbers, but I guess not because commutativity breaks down
 
They contain infinitely many copies of the complex numbers.
 
Commutativity breaks because $i,j,k$ don't play nicely with each other
 
I'm just trying to look at sin(x) in different ways
 
4:16 AM
For the imaginary unit you can take any quaternion of the form $bi+cj+dk$ where $b^2+c^2+d^2$.
 
But the idea behind the quaternions is this
 
gotta prove that $\int_{0}^{\infty} sin(t^2)dt$ is not absolutely convergent.
 
So, complex numbers were thought of because in trying to solve the cubic, you needed to take square roots of negative numbers (otherwise no one would've really cared to do so)
 
(If you square $bi+cj+dk$, the imaginary parts cancel because of the anticommutativity thingy and you're left with $-b^2-c^2-d^2$)
 
Even if cubics had real solutions you needed to do this
 
4:17 AM
TFW typing up your homework takes longer than actually doing it
 
So turns out complex numbers are nice and do a lot of fantastic things, but also they're really just a way to put multiplication on $\mathbb{R}^2$
So someone tried to do it on $\mathbb{R}^3$, failed, and eventually did it in $\mathbb{R}^4$
That's the quaternions
 
Hence, the quaternions contain infinitely many (continuum-many) square roots of $-1$. And the Fundamental Theorem of Algebra fails spectacularly.
 
But then if you just restrict to the appropriate plane it's just $\mathbb{C}$. And yeah what Akiva said
 
@Daminark And the guy realized it while walking on a bridge, and then did the incredibly nerdy act of carving his equations (without context) into the bridge with a knife
 
Is it possible to have some complex function $f(c), c\in \Bbb{C}$ such that the radius of convergence is different from one will expect to get if considering onyl complex values. One thing that wonders me is I can see the radius of convergence is sensitive to what happens in the complex plan for real functions, but it seems we never ran into the problem of the radius of convergence being sensitive to the auternions?
 
4:20 AM
@Secret by the standard definition of radius of convergence, it's always a real number
 
@Daminark And (by coincidence, almost?) they turn out to be useful in describing rotations in 3D
 
@Daminark sorry I made a typo, let me elaborate:
7 mins ago, by Akiva Weinberger
Unexpectedly, the radius of convergence is $\dfrac1{e^x+1}$ is $\pi$.
 
'cause $SO(3)$ is $S^3$ with opposite points identified, and you know what else is $S^3$? The unit quaternions!
 
@AkivaWeinberger Get yir geometry outta here
 
This is true because the singularity of this function is at $i\pi$ which is not in the real line
 
4:21 AM
:P
 
But I am wondering whether there exists a function complex (or even real) where the radius of convergence can only be correctly calculated because its singularity is at the quaternions
 
Shouldn't be a thing
A function $f:\mathbb{C}\to\mathbb{C}$ is analytic iff it's locally expressible as a power series
 
'Cause $q\bar q$ ends up equaling $1$ for those guys, and if you represent a 3D coordinate by $P=xi+yj+zk$ then rotations get represented by $qP\bar q$
and you'll notice that $q$ and $-q$ represent the same rotation, which fits in with the topology of the set of rotations
 
So if you don't have a complex singularity, you can keep pushing on. You may have heard of something called analytic continuation, where you take a function that's defined in some open set $U$ a priori but extend its domain of definition
 
@Secret It's really hard to define "analytic" and "holomorphic" in the quaternions anyway
and there's a third word as well, what is it
I dunno
But the lack of commutativity messes it up.
There's no quaternion analysis like there is complex analysis.
(Or, I suppose you could try to study it, but it's ugly and not useful)
(I'm sure you can find PDFs on it)
 
4:26 AM
You may appreciate this Secret, but it's provably the case that the only real division algebras are the real numbers, complex numbers, quaternions, and if we allow non-associativity, the octonions
 
@Secret But yeah, you only need to look for complex singularities to find the radius of convergence.
 
@AkivaWeinberger there is but like you said it seems to suck
 
@Daminark Finite-dimensional, anyway. I think otherwise the rational functions ($\Bbb R(x)$) count
 
Tru
 
@AkivaWeinberger Hmm.. I see, so the lack of commutativity resulting in what is "holomorphic" being unclear basically means if a function does not return any quaternions for complex or real outputs, then the radius of convergence will be fully determined by its behaviour in the complex plane
 
4:33 AM
Some other footnote: One reason I like to study objects starting from the weakest is because it is usually easier for me to specialise (make stronger versions of) than to generalise (where there are many possibilities, and it will not be easy if you are too used to the stronger cases)
 
Is $f(x)=ixi$ holomorphic? What would its derivative be, if so? Is it analytic?
(This maps $a+bi+cj+dk$ to $-a-bi+cj+dk$)
@Daminark You need "unitary" as well, otherwise the 2D algebra that's essentially the complex numbers with the "multiplication" rule $z*w=\overline{zw}$ counts as well.
Also, the reason the 16-dimensional sedenions don't count is because they have zero divisors.
 
Well, wikipedia said all complex conjugates are holomorphic in the quarternions, which is in contrast to the case in complex. But again, I have not actually read into the topic yet, so technically I don't understand it yet

And also, it seems that most complex analysis notions will break down and needed to be redefined from the ground up, along with using a different cauchy riemannian like equation: http://www.zipcon.net/~swhite/docs/math/quaternions/analysis.html
 
@Akiva I think that's gonna still be isomorphic to the complex numbers
This is all up to isomorphism
 
No it's not, it doesn't have a unit @Daminark
 
Wait that makes no sense
Division algebra requires a unit
 
4:38 AM
4 mins ago, by Akiva Weinberger
(This maps $a+bi+cj+dk$ to $-a-bi+cj+dk$)
O wait, that only conjugates i, not a full quternion conjugate
I misread the question
 
> The quaternion function $q^2$ is not quaternion-differentiable in any sense known to the author.
Eww
 
A division algebra is one where you have $1\ne 0$ and for any $x$ you can find some $y$ such that $xy = yx = 1$, basically a non-commutative field
 
No
> We call $D$ a division algebra if for any element $a$ in $D$ and any non-zero element $b$ in $D$ there exists precisely one element $x$ in $D$ with $a = bx$ and precisely one element $y$ in $D$ such that $a = yb$.
 
Daminark's case is for the associative subset of division algebra
 
Oh you're allowing non-associative business?
 
4:42 AM
Sure, yeah
 
Well, when we quote the definition, we of course quote the general one, and then specialise to the context of the problem
 
If you want it to be associative then yeah you need a unit
 
which means in our context only associative division algebra are important since anything at or before quarternions are associative
 
Eh, he had the octonions in there
 
ok sorry I misread again, then use the general definition
(I must be not having enough sleep)
 
4:44 AM
I ask for associative, I mentioned that some folk allow non-associative in which case you can throw in the octonions for sure
 
There's an easy proof, incidentally, that there's no 3D normed division algebra
 
I did not expect that there would be more in that case though, however usually I'm in the "if it's not associative we need to burn it to the ground"
 
which is that you can find two numbers that are the sum of three squares whose product isn't the sum of three squares
I can't remember what the numbers were, in the end
 
Oh what I had in mind was something to the effect of, if you have a normed division algebra structure on $\mathbb{R}^n$, then I think $S^{n-1}$ would need to be a Lie group
 
Ah, $(1^2+1^2+1^2)(0^2+2^2+3^2)$ seems to be the smallest example
($3\times13=39$ is not the sum of three squares)
 
4:48 AM
But if you have a Lie group $G$, say of dimension $k$, then $TG \cong G \times \mathbb{R}^k$
And this isn't true of $S^2$
 
(Cont'd) (That's for normed thingies though)
@Daminark What's $TG$?
Tangent plane?
 
Tangent bundle
 
tangent fundle ?
 
And it does work for $S^1$ and $S^3$?
 
Yeah $TS^1 \cong S^1\times \mathbb{R}$ and $TS^3 \cong S^3\times \mathbb{R}^3$
 
4:51 AM
Ah, right
and $TS^2$ gets caught up in the whole "You can't comb a hairy ball" thing
so you can't orient all the planes in a sensible way
but you can comb $S^n$ for odd $n$
 
Daminark: I usually discuss things in the associative because that's where most people are comfortable with, but I am often aware of the nonassociative generalizations and will bring it up when relevant.

I am not afraid of nonassociativity (though I still don't understood it much beyond lie algebras in physics) because 2 years ago I found that there is strong evidence you need to blow up nonassociativity to have not very trivial "division by zero algebra" (warning, non mainstream yet) such that the multiplication structure is not $\Bbb{Z}/n$ (which when equipped with a distributive law, wi
 
(Does $TS^0$ make sense as a concept?)
 
$S^0$ is a point, how are we going to differentiate a single point...?
 
Two points
And I guess no, then
 
I mean it's technically a smooth manifold
Or.. I guess it depends
It's a bit of a degenerate case... :thonk:
 
4:56 AM
'Cause $\Bbb R$ is a one-dimensional normed division algebra over $\Bbb R$, so if your equation works then we should have $TS^0\simeq S^0\times\Bbb R$
> There are exactly four normed division algebras: the real numbers ($\Bbb R$), complex numbers ($\Bbb C$), quaternions ($\Bbb H$), and octonions ($\Bbb O$). The real numbers are the dependable breadwinner of the family, the complete ordered field we all rely on. The complex numbers are a slightly flashier but still respectable younger brother: not ordered, but algebraically complete.
 
yeah, and that's why physics uses a lot of algebraic stuctures more than maths do as claimed by h bar
 
> The quaternions, being noncommutative, are the eccentric cousin who is shunned at important family gatherings. But the octonions are the crazy old uncle nobody lets out of the attic: they are nonassociative.
 
@Akvia not quite
 
Say my name
 
You're looking at $TS^{n-1} \cong S^{n-1}\times \mathbb{R}^{n-1}$
 
4:59 AM
:P
@Daminark Oh, OK
 

« first day (2727 days earlier)      last day (2285 days later) »