« first day (3052 days earlier)      last day (2265 days later) » 
00:00 - 18:0019:00 - 00:00

00:00
Let me see where googling leads me
Hi chat
Hi @Astyx
What's the difference between the Levi-Civita tensor and the determinant ?
@JoeShmo Do I need to implement get/set methods for all my fields? oO
00:03
@Lozansky not necessarily
This might pose a problem
I'm mixing the graphical and the logical part of my code a bit
do you have it on git where i can take a look?
No, but I can upload it there I guess
get into the habit of doing that anyway
and study more algorithms
and math
It's a bit... unstructured right now though
00:06
there's your first problem
Given that there is an isomorphism $f$ from G to H If $ n \in \mathbb{Z}^{+}$ and $G$ has a n element of order $n$ then $H$ has an element of order $n$.
Hi guys, got a question: "Show that a connected metric space with at least two points contains no isolated points". Can anyone check if this is the easiest way to prove this?
Suppose in connected MS $(X,d)$ there is an isolated point $x \in X$, so that $\exists r > 0$ s.t. $B_r(x) \cap X = \{x\}$.
Then we argue that the set $\{x\}$ and $X \cap \{x\}^c$ is separated:
To prove the first part, i.e. $\overline{A} \cap B = \emptyset$
Proof: Let $x \in G$ so $g^n = e_G$. Consider $f(g^n) = f(e_G) = e_H$. We can write $f(g^n) = (f(g))^n = e_H$ Let $h = f(g)$. Note $h \in H$ so $h^n = e_H$ as desired.
Is this a valid proof?
we see that $\{x\}$ is a single point so it's closed, thus its closure is the same. Then given disjointness the above must be true.
To prove the second part, i.e. $A \cap \overline{B} = \emptyset$, see that $x$ is not an accumulation/limit point(because it is an isolated point). Then there are no sequences in $X \cap \{x\}^c$ that converges to $\{x\}$ (we can't take arbitrary $\epsilon$ because for any $\epsilon$ less than $r$, we cannot provide the definition of convergence to $x$).
Then $x$ is not in the closure of $B$, so the intersection must be empty. Then it contradicts the connectedness of $X$.
@SharathZotis I've seen this misconception quite a lot of times actually
$h^n = e_H$ does not mean that $h$ has order $n$
00:17
why not isn't that the definition of order?
no
$h^n = e_H$ means that the order of $h$ divides $n$
@JoeShmo Tried adding you to a new room, not sure it worked
Oh yes you are right
it is because it is possible that it is not the least positive integer such that we get to the identity of $H$
right
hi @Daminark
Hey!
00:23
Guys, I have a question
for a x in R not Q does there always exist a sequence such for all n in N, x_n is in R/Q?
I meant R not Q?
@Lozansky don't think that it did
such that x_n converges to x
?
I know that it holds for the set of real numbers
because R is dense over R
but what about R not Q?
it should still hold
am I right?
@mathsresearcher are you saying that for all sequences that is elementwise irrational, converges to an irrational number?
no because there are uncountably many irrationals that are $\frac{1}{n}$ close to 0 for any $n \in \mathbb{N}$
Then you can construct a sequence such that $x_n \in \mathbb{R} \cap \mathbb{Q}^c$, such that $|x_n| < 1/n$.
Then by Archimedean property, $\forall r \in \mathbb{R}$, there exists $n \in \mathbb{N}$ such that $\frac{1}{n} < r$
Taking $r$ to be arbitrary, you get definition of convergence to 0, and 0 is a rational number
what you're asking essentially is whether the irrational metric space with the $|\cdot|$ metric is complete
00:31
My professor said that for a function f(x)=x^2 if x in Q and 0 is rational there will be a sequence such that f(x_n)=x^2_n or f(x_n)=0
so i'm not sure how he got both of these
I'm getting really confused by your subscripts, can you use LaTeX?
Choose a sequence of rationals or irrationals. (Your definition of $f$ has a flaw.)
without assuming that there exists an elementwise irrational sequence that converges to an irrational
Who cares about convergence?
Analysts probably do
00:32
hi Demonark
@OneRaynyDay I'm not sure how to use Latex
literally put dollar signs around your expression
\$ x \$
$x$
okay so for a function f(x)=$x^2$ if x is rattional and 0 if x is irrational, there will be a sequence such that $f(x_n)=(x^2)_n$ or f(x_n) =0
thats what the professor said
The professor means: You can pick a sequence consisting of all rational numbers and you can pick a sequence consisting of all irrational numbers.
Are you trying to pick them so they converge to something in particular? Or just random sequences?
to 0
converge to -
0
00:36
OK, easy in either case.
Give me a sequence of rational numbers converging to $0$.
OK, now give me a sequence of irrational numbers converging to $0$.
$sqrt(2)/n$
Brilliant. OK, we're done.
oh so it holds for 0 but in general it is not true that for an irrational number there exists an irrational sequence that converges to it
00:38
Of course.
I'm not sure you know what your question is @mathsresearcher
You can find a sequence of rationals — or a sequence of irrationals — converging to any real number you want.
@TedShifrin That makes sense, how would I prove it for the irrationals?
oh
Didn't we talk about dense sets last week?
00:40
to let the the real number be bounded
by two irrationals
yup yup
i just remembered
but I think we talked about just the rational case
the proof I gave above is meant for the general proof to any real number (just replace zero w/ any real number)
Hi @Ted!
Hi, a @Balarka
Ted, so that would mean the irrationals are dense over the reals
dense in the reals, yes, of course
00:51
thanks so much @TedShifrin and @OneRaynyDay
most welcome, maths
It wasn't ever mentioned in the course that the irrationals are dense over R so thats why I wasn't sure
You should try to prove it.
Yeah, thats a good idea
@TedShifrin I wish I was attending your lectures on analysis, multivariable and linear algebra to be honest
Well, I retired 3 1/2 years ago, so you're safe :) ...
00:59
wipes sweat from forehead
Thank you, Demonark.
Lets say $G$ is a group of order 12
so the possible orders of subgroups of $G$ are 1,2,3,4,6,12
by Lagranges theorem
Of these $G$ must necessarily have subgroups of order $2$ and $3$ by the conclusion of Cauchy's theorem.
Are there any orders for which subgroup $H$ of that order (if it exists) must necessarily be normal in $G$
If, e.g., $|G:H| = 2$, then $H$ is normal.
I know for sure the trivial subgroups 1,12 will be normal. However, I am not sure how to find the other one(s) if they exist. I don't have Sylows theorem available to me
01:07
And you don't have a particular group of order 12, right?
There are very different ones.
Yup its just a general one
Do you know some concrete examples?
I agree @user193319 if the index of $H$ in $G$ is 2 then $H$ is normal
At some point in life I want to teach a class and say "Let's have a concrete example", and have that example involve physical concrete
How do I find such an $H$ though?
01:09
Force against a dam, Demonark. Of course, you don't believe in "useful" calculus problems.
@Sharath: Whether there is one depends on $G$.
So if $|G: H| = 2$ that means the we have cosets $H, G-H$
That's true but for the wrong reasons. I'm not against usefulness. I'm against calculus problems :P
Uh huh.
And in any event I'm willing to change my stance if it means making that pun
I figured as much.
01:12
Is there a theorem where $|G: H| = \frac{|G|}{|H|}$
Sure.
Lagrange
Assuming finite group.
Ah I see so then H could be order 6 to be normal
1,6,12
It could also have order 4 and be normal.
You need examples, @Sharath.
Boo @CaptainAmerica
01:13
@TedShifrin is galois theory every considered in differential geometry?
Ok thank you
Boo to you too :D
Nah, maths.
I have really big news
YOu didn't burn down the house tonight?
01:14
It's already burned from last night so yes
Lol, not tonight.
Demonark is a fan, I can see.
OK, good.
lol @Dami
Three hundred sixty-five degrees
BURNIN' DOWN THE HOUSE
howdy @Fargle
01:15
Hey @Ted
Today after my presentation at the grand opening I met the president of an engineering company. He really liked my fish feeder design and said he wants me to present it to his engineers once I finish the prototype. He gave me his business card and everything!!
great ref
Wow, that's really cool, @CaptainAmerica.
I know, I'm really excited :D
@TedShifrin But have you never studied the absolute Galois action on $(\infty, 2)$-germs on a differential topos?
01:16
math flies out the window :D
smacks a @Balarka
@TedShifrin Ted, i'm curious, is it possible for a professor to teach a class he never took as an undergrad or postgrad?
Sure. I did that several times.
A lot of faculty don't want to work that hard, though.
@Ted That's basically what differential geometry is, a special case of higher Hopf-Galois theory, you know
Sorry, pushing it too far
I am getting annoyed by myself now
It saved me from getting in trouble too. My English teacher called my mom and said I'm getting an F if I don't turn in my overdue essay. My mom didn't even get mad. But I have to stay up tonight and do it.
01:18
References. [1] Byrne, D., "Burning Down The House," Speaking in Tongues, 1984.
I didn't know self-annoyance was a thing, a @Balarka
yes, @CaptainAmerica: Get out of chat and write your essay.
@TedShifrin how long would it take you to prepare for such a course?
I should, just wanted to share the good news. God Bless us, Everyone! ;-;
Ok, goodbye.
@Fargle Remain In Light > Fear of Music > Speaking in Tongues
Either agree or die in fire
Yeah you're 100% right
01:21
puts down flamethrower
Not even by threat, that's just the correct ranking
It was a lot of work to do a good job. I did a year-long applied math course back 30 years ago. It was a great course, but a ton of work. I did probability right before I retired. That took a good deal of thought/preparation. I did some stuff in grad courses I'd never learned/studied, too.
@TedShifrin Thats interesting, whats the most challenging course youve taught? for you and for your students?
I have no idea.
01:27
@TedShifrin Have you ever been interested in number theory?
Nope, not really.
How do I show that there is a hyperplane in $A^n$ which contains $n$ points $p_1, \cdots, p_n$?
There are lots of ways to do that. What are you trying?
Did someone say number theory?
@Daminark Are you interested in number theory?
01:29
It's definitely one of the things I like :D
@TedShifrin Pass to the moduli space of affine hyperplanes in $\Bbb A^n$ :3
@TedShifrin I am not sure how to begin. I just write $p_i=(x_{1,i}, \cdots, x_{n,i}$ and try to make these points satisfy some equation. But this seems complicated. How about considering the case where some of the points are collinear?
I wouldn't write coordinates for everything. But a hyperplane is given by a normal vector and some point on the hyperplane.
Work with vectors.
I love linear algebra to be honest I think its essential to differential geometry but I'm not that interested in polynomial vector spaces
what sort of thing is semidirect product, categorically speaking?
01:35
It's the unique map such that for every two elements $x$ and $y$...
@TedShifrin do you think its best to learn differential geometry while learning GR?
@Daminark what?
Lol I was joking, the point was that the sentence started off sounding all universal property-like and then it womp womp it references elements
I can't answer that, @maths. I don't know GR.
01:57
@TedShifrin Is there some way to solve the question using just linear algebra?
Guys
What is the probability that second card is a king given that the first is a spade
when are these two events independent?
why*
@AlessandroCodenotti You asked me if the Hawaiian earring is an inverse limit a couple weeks ago and I said no. I think I was wrong, which I realized while writing this answer.
I guess if this is right then it is a counterexample to "$\pi_1$ commutes with inverse limits"
02:16
:48013915
Sure. You get a system of $n-1$ linear equations $A\cdot (x_i-x_1) = 0$, for $i=2,\dots,n$ (for the entries of $A\in\Bbb R^n$). So you expect a $1$-dimensional solution space (which gives $A$ up to scalar multiples). If the points are not in general position, the rank is $<n-1$ and you have infinitely many hyperplanes.
Can anyone help me with the probability question I asked?
 
2 hours later…
04:04
So @mathsresearcher you have two cases,
The first case is that the first card is a king of spades,
The second case is that the first card is a spades but not a king
The first cases' probability will simply be $\frac{1}{52} * \frac{3}{51}$
Can you tell me the probability for the second case?
Once you have found that you can add and use the formula for P(A|B) "Probability of A given B" to find the answer
$4/52 * 12/52$
?
i mean 4/51
Good so 12/52*4/51
$P ( A | B ) = \frac { P ( A \cap B ) } { P ( B ) }$
but if these are all the cases, why do I need the formula? why don't I just add them?
So if we let A be the event that a King is drawn as the second card
And B be the event that a Spade is drawn on the first Card
The reason we have to use the formula is because
all we have found is the probability of A and B happening
but we are trying to find the probability that A happens given that B happens
Thus we must divide out the Probability of B happening
Lets say we have groups $\mathbb{Z}_2 \times \mathbb{Z}_2 \times \mathbb{Z}_2 \times \mathbb{Z}_5$,
$\mathbb{Z}_2 \times \mathbb{Z}_4 \times \mathbb{Z}_5$, and
$\mathbb{Z}_8 \times \mathbb{Z}_5$
Which of these have an element of order 20
How do I go about answering this without doing it exhaustively and checking manually?
@MikeMiller You have any idea on this one?
04:41
Hello
What is limit $x\to0 [\frac{\sin x}{x}]$ ?
[.] is step function
Shouldn’t it be 1 because sinx = x as x approach 0
Nvm I think I figured it out I forgot about the theorem which tells us order of elements of direct product is lcm of the order of the individual elements
@Fawad L'Hopital to the rescue
your intuition is correct, btw, but the formality is wrong. $sin x \neq x$ for small $x$ in general. but they are roughly equal
oh step function. how is the step function defined?
do you round up, or down?
@JoeShmo down
well in that case the limit is $0$
But at x tends 0 we don’t say sin x is nearly x. We say sinx = x right?
04:55
NO
we say sinx is nearly x
more formally, for "small" $x$, $|sin(x) - x| < \epsilon$ where $\epsilon$ is also "small"
Ok thanks
do you see why the limit is 0 though
05:37
I would recommend graphing (sin x)/x
Hi, getting ghosted on current problem on MSE: math.stackexchange.com/questions/3036268/…
Anyone fancy to take a look?
06:30
@BalarkaSen that's interesting, I'll read it later!
@MikeMiller that's what I was hoping for! It can be bad but not that bad
3D printed Perko knot
06:57
You mean the Perko knots ;)
10_161 and 10_162, the one and only
Also, the angle they use here is confusing… the rightmost crossing looks like it has odd degree
07:32
Yo. Anyone want to talk about homotopy groups?
I'm calculating the fundamental group of a $g$-holed $k$-punctured torus $T_{g,k}$ where $k,g>0$.
I'm going about it with Van Kampen, with one set being a {$4g$-gon with edges identified without a disk} and the other being a {disk with $k$ holes}.
The $4g$-gon with the disk is a bouquet of $4g$ circles, so it has the fundamental group $F_{4g}=\langle a_1,a_2,\dots, a_{4g}\rangle$ (the free group on $4g$ generators).
You want to take a small neighborhood of the skeleton of the $4g$-gon
But I'm being pedantic
The torus with a disk missing has fundamental group The disk with $k$ holes has fundamental group $F_{k}=\langle b_1,b_2,\dots, b_{k}\rangle$. The overlap $A$ is an annulus with free group $\mathbb{Z}=\langle c\rangle$.
So Van Kampen tells us that

$\pi_1(T_{g,k})=\langle a_1,a_2,\dots,a_{4g}, b_1,b_2,b_k\,:\,a_1\cdot a_2\cdots a_{4g}=b_1\cdot b_2\cdots b_{k}\rangle $.
We can solve the relation for $b_k$ in terms of the other generators, so we can toss out the relation and $b_k$. This leaves us with $\pi_1(T_{g,k})=F_{4g+k-1}$.
@Cbjork No, bouquet of $2g$ circles
07:36
Ah thanks
you are right
Other than that, the computation is correct
Are all the higher homotopy groups trivial?
Here's a trick: Note that surface of genus $g$ minus a point deformation retracts to the 1-skeleton, the bouquet of $2g$ circles. Then surface of genus $g$ minus $k$ points deformation retracts to bouquet of $2g$ circles wedge a bouqet of $k-1$ circles.
So the space is homotopy equivalent to wedge of $2g+k-1$ circles
@Cbjork By this argument, yes
A wedge of circles has trivial higher homotopy group: it's got a contractible universal cover!
(A tree)
thanks @BalarkaSen
07:59
Hi! does the function z=(x^2-a^2)*(y^2-a^2) have a specific name?
 
3 hours later…
10:58
I tied the Perko knot in my earphones hoping for clarity and I have not found it
grammar: If I _ (know) that it _ (make) you sad, then I _ (rather not give) you this present to begin with.
stoke's theorem
@LeakyNun how should i do the proof?
you should cite the theorem
11:07
I know stokes theorem. How should I proceed to proof?
@LeakyNun there should be $\int \phi . \hat{n}$
That should be equal to ${ \nabla \times \phi } \hat{n}$
@AkivaWeinberger would you like to instruct me how I should have said the sentence?
11:28
"If I had known that it would make you sad, then I wouldn't have given you this present to begin with"
(Or "Had I known that it")
@LeakyNun
I’m trying to identify and categorise the singular points of legrendes equation, $(1-x^2)y’’ -2xy’ +l(l+1)y=0$ where l is a constant.
oops
didnt mean to hit send
Or maybe "If I knew it would make you sad I wouldn't've given you this present to begin with" @LeakyNun
i have the points, $x=\pm 1$
now I think only x=1 is regular, yet according to my notes x=-1 is too?
I dunno much differential equations, sorry
Nevermind
i got it
I’m just stupid
11:38
@LeakyNun Whose heart are you breaking?
@JakeRose divide by $(1-x^2)$ throughout and split it into $(1+x)(1-x)$. Here $x=-1$ is a pole of order 1, hence regular.
12:06
Hi chat
12:53
Hey Astyx
Could someone help me find the PS of $y'' + 2y' = x $
Which general form should it have?
I tried $y=Cx$ and $y=Cx^2$
Oh wait, $y=Cx + Dx^2$ ?
yup
13:34
Why graph theory is known as one type of geometry ?
Can anyone tell me ^ ?
I wouldn't call it geometry in the classical or modern sense
@user616128 Can you tell me any reason for not calling it geometry?
Why I wouldn't call it such?
Well if you look at the classical definition of geometry, it doesn't really meet any requirement
There isn't a worry about relative position, or distances, size, shape etc
The modern definition is a little hazy, but I don't see any etale fundamental groups or topoi appearing
13:51
Graph theory is topology, I thought
Surely you mean 1-dimensional CW-complex theory?
Hm maybe not
37
Q: What's the relation between topology and graph theory

FrancescoI read the Wikipedia articles for both topology, graph theory (plus topological graph theory). Does topology encompass also graph theory? Or topology is only about studying shapes while graph theory is about relations and the two meet in topological graph theory?

In any case I wouldn't call it geometry
Maybe call it "discrete math"
By the way,
3 hours ago, by Akiva Weinberger
I tied the Perko knot in my earphones hoping for clarity and I have not found it
I think I found clarity
(And also that projection^ is a much more symmetric projection than any one the ones I found online)
14:15
Frick
No
I messed with the knot a bit and now I can't get it back to looking like the picture
I'm pretty sure I didn't change the actual knot though
Oh I got it
My god these are confusing
14:38
Oh, oh know what, the image I drew above is the same as this one from Wikipedia, except that the top-right strand in Wikipedia's image has been moved to the bottom in mine
Grrr logic is SO HARD
in Logic, 1 min ago, by user21820
@Secret This doesn't make any sense at all. Firstly your indentation is clearly wrong. Secondly at one point you wrote "Q(0)" for no apparent reason at all. Can you please follow the rules strictly?
I am a naturally born rule breaker, no wonder why I found logic so hard to wrap my mind around
it's not a great sign when it's 8:40 am and you feel like taking a nap
Oh, also, Wikipedia's and mine are mirror images
(The Perko knot is chiral)
(I dunno which is "the" Perko knot and which one is the mirror image. The figures from Perko's paper agree with my version.)
I apologize for all the knot theory spam by the way
When I'm bored and I have nothing but my headphones to fidget with, this is what happens
Hm, are both of Wikipedia's images mirror images of mine?
I am that kind of person that will subconsciously break rules unless I fully understood them
15:29
Let $X$ be a topological space, $\{f_i\}$ is collection of continuous functions from $X$ to $\Bbb{R}$. Suppose that $f(x) = \inf_{i \in I} f_i(x)$ exists for each $x \in X$. Does it follow that $f$ is continuous?
I'm kind of the same way, Secret, but it's more conscious curiosity on my part. Like, I am interested in axioms of mathematics, solely because I enjoy the idea of adding axioms specifically to see what breaks.
@user193319 What could the collection of continuous functions look like?
Anything.
Why not pick a few examples?
Just to explore the space a little bit.
There is no space to explore. $X$ is any topological space.
Yes, but they are to $\mathbb{R}$, meaning we can evaluate a little bit of what they look like.
15:36
But we don't really know anything about the functions.
So, you seem to be hinting at the fact that it isn't true. Am I right?
We know they're continuous, and we know what it means to be open in $\mathbb{R}$, right?
So we can gather some info about what it would be like to be open in the preimage.
Wait, is $f(x) = \inf_{n \in \Bbb{N}} x^n$ or some variant a counterexample?
Where $X = \Bbb{R}$.
Or $X = [0,1]$ is better, perhaps.
Is it?
@AlessandroCodenotti Thanks!
15:45
The infimum of $x<-1$ doesn't exist as $n\rightarrow\infty$, right?
Which is why $X = [0,1]$.
Ah, fair enough.
16:14
OK, so this is the last I'm gonna write about the Perko knot pair on this chat, for a while at least, I promise, but I just wanted to share that I wrote a question on it:
0
Q: Perko pair - What's the handedness of these pictures?

Akiva WeinbergerIn 1974, a paper titled On the Classification of Knots appeared showing that the knots $10_{161}$ and $10_{162}$ in Dale Rolfsen's knot table were actually the same knot. He included this picture, showing how to deform one into the other: From that point on, $10_{161}$ and $10_{162}$ became kn...

One thing I never understood in classical logic:
[Given]  Let Q(n) denote "Forall k in N ( k<n implies P(k) )".
[Given]  Then Q is a property on N.
[If sub] If Forall n in N ( Forall k in N ( k<n implies P(k) ) implies P(n) ):
[If sub]         Forall n in N ( Q(n) implies P(n) ).
[Forall elim]    Q(0) implies P(0)
[If sub]         If Q(0):
[If sub]                 Q(0)
[or intro]               Q(0) or not Q(0)
[Given intro]            Given k in N:
[Implies intro]                  If Q(0):
[Implies intro]                         If k < 0:
[Implies intro]                                 If 0 < 0:
Just how on earth I keep going in circles without realising until I get to the end of the line
16:54
I have covered basics of Linear alg. I am not sure which concepts and terms will need to understand Galois theory. Can anyone tell me?
17:51
@guy that responded to me this is not programming, but possibly an infidel has been allowed to edit mathematics Wikipedia articles which would explain why $\sum_{n \operatorname{mod} c}$ has been used, but I don't know what a representative is
00:00 - 18:0019:00 - 00:00

« first day (3052 days earlier)      last day (2265 days later) »