« first day (2588 days earlier)      last day (2731 days later) » 

14:00
and they never denoted it by $\mathbb T$
My book writes that as $U$ though
mine tooa
too*
as $U_1$ but whatever
@Semiclassical nice
14:00
okay thanks
niiiiice
Semi you are officially a hero :P
alright I'm going to think about those T/F exercises now :P I think I actually have a theorem on that.
I have no earthly idea why it's called T, though
"The notation T for the circle group stems from the fact that, with the standard topology (see below), the circle group is a 1-torus. More generally Tn (the direct product of T with itself n times) is geometrically an n-torus."
Oh, derp. It's in that article
14:02
lolll
I was about to say
Does it make sense to call points in projective space $RP^n$ projectively independent if the corresponding linear subspaces are transverse, i.e. their sum is $R^{n-1}$?
read your own damn sources :P
oh yay torus
14:02
pfffft
hahahah
I read the first paragraph, isn't that enough?!? /s
$[G,G]=\{\langle[a,b]\rangle |a,b\in G\}$
$[G,G]\subseteq N \implies \{g\langle[a,b]\rangle g^{-1} |a,b\in N,g \in G\} \subseteq g^{-1}Ng$
$G/N=\{gN,g\in G\}=\{gg^{-1}Ng,a,b \in N, g,s\in G\} = \{Ng,a,b \in N, g,s\in G\}$
bleh, I give up for now
@abenthy Seems like it
Error: Insufficient group theoric knowledge
14:04
I thought you study chemistry
@abenthy I don't know if that's the right word for it, but the concept certainly looks like it makes sense
Representation theory of groups != group theory
@Secret What are you trying to do?
Yup, chemistry don't need to deal with normal subgroup
Showing that dividing by the commutator group is abelian?
14:05
45 mins ago, by Leaky Nun
@Secret prove that $G/N$ is abelian iff $[G,G] \subseteq N$.
I then give the following weirdness...
19 mins ago, by Secret
$[G,G]=\{\langle[a,b]\rangle |a,b\in G\}$
$[G,G]\subseteq N \implies \{g\langle[a,b]\rangle g^{-1} |a,b\in N,g \in G\}$
$G/N=\{gN,g\in G\}=\{gs\langle [a,b]\rangle,a,b \in N, g,s\in G\} = \{gg^{-1}s\langle [a,b]\rangle g,a,b \in N, g,s\in G\}$
$= \{s\langle [a,b]\rangle g,a,b \in N, g,s\in G\} = \{Ng,g\in G\}$
@AkivaWeinberger Thanks. In a paper I read the statement "the $m$ points in projective space $RP^{m-1}$ span an $(m-1)$ simplex in $RP^{m-1}$". So I guess I can translate this as "the corresponding linear subspaces in $R^m$ are transverse".
which makes no sense as the workings will mean G/N is always abelian (since I can always summond a conjugate in the workings to cancel out a left coset and convert it to a right coset)
In other words, you want to prove that $G/N$ is abelian iff for all $a,b\in G$, we have $aba^{-1}a^{-1}\in N$
14:08
$ab=ba(a^{-1}b^{-1}ab).$
@ShaVuklia how is progress
I'm still thinking
@Secret I don't understand. Yes, $gN=Ng$ due to $N$ being a normal subgroup, but how does that imply it being abelian?
@Secret tell me what it means for G/N to be abelian.
@CowperKettle long time no see.
@Leaky i donno. i'm kind of stuck
14:11
@ShaVuklia do you think it is true or false?
there's no point in guessing:P
alright
my intuition tells me that 2 is true, but I have no arguments for that.
@LeakyNun $[c,d]=e |c,d \in G/N$
what did you learn from Ted?
14:12
no f that
@ShaVuklia come on
What are you trying to show (Sha)
Aug 30 at 3:27, by Leaky Nun
A student is asked to show taht if $H$ is a normal subgroup of an abelian group $G$, then $G/H$ is abelian. The student's proof starts as follows:
We must show that $G/H$ is abelian. Let $a$ and $b$ be two elements of $G/H$.
a. Why does the instructor reading this proof expect to find nonsense from here on in the student's paper?
@Secret answer this question first ^
31 mins ago, by Sha Vuklia
Guys, any idea what $\mathbb T$ stands for in this context: $\mathbb C^*\cong\mathbb R^*\times\mathbb T\cong\mathbb C/\mathbb Z$
This? ^
@AkivaWeinberger ??
oh
don't give her the answer (or any hints for that matter):
21 mins ago, by Leaky Nun
@ShaVuklia T/F: If $H < G$ and $K$ is a group such that $H \times K \cong G$, then $G/H \cong K$.
14:13
actually, $\mathbb C^*\cong\mathbb R^*\times \mathbb T$ makes a lot of sense @Leaky
I can prove that
19 mins ago, by Leaky Nun
@ShaVuklia T/F: If $H<G$ and $K$ is a group such that $G/H \cong K$, then $H \times K \cong G$.
@ShaVuklia I already said $re^{i\theta}$
dude:P
I'll first work that out
and then I'll do the other isomorphism
alright
@Semiclassical I need you explanation for that if you don't mind.
1 hour ago, by Abcd
Please help me understand the bold part of this question: Find the value of a for which $(ax^2+3x-4)/(a+3x-4x^2)$ takes all values for all real values of x. I don't want the solution.
wait but @Leaky
shouldn't it be $\mathbb R_{>0}^*\times\mathbb T$ then
or just
$\mathbb R_{>0}\times \mathbb T$
14:17
1 hour ago, by Abcd
Leaky argues, for all y there's a root for x, therefore discriminant is greater than zero.
@LeakyNun isn't abelian means ab=ba?
18 mins ago, by Leaky Nun
$\Bbb R^*$ would have to be the positive integers though
1 hour ago, by Abcd
I understood neither the bold part of the question nor Leaky's argument.
positive reals
ok
@Secret right, but that doesn't answer the question.
14:18
If anyone else has a simple explanation ...
@Secret There's something wrong in the last sentence.
@Abcd It means that the range of the function is $(-\infty,\infty)$
for some $a$
@AkivaWeinberger Yes, it does. Leaky confused me with strange wordings. I understood this part exactly as you stated then I became unsure after Leaky's explanation.
@Abcd after explaining for like 2 hours one's wordings must become strange
@AkivaWeinberger Alright, now what about this part: for all y there's a root for x, therefore discriminant is greater than zero
14:21
Hm wait
@LeakyNun Why didn't you say what @Akiva said initially? We were basically beating about the bush.
So when we say the range is $(-\infty,\infty)$, we mean $f(x)=y$ has a solution for every $y$, right?
@Abcd because spoonfeeding you with everything isn't my style
@AkivaWeinberger yes
@LeakyNun but beating about the bush won't help either
@Abcd you were relying on me to give you every detail.
Come up with something yourself.
14:22
@LeakyNun Starting from $a,b \in G/H$, I don't think one can even work its way towards $ab=ba$
@Secret what is $G/H$?
So $\dfrac{ax^2+3x-4}{a+3x-4x^2}=y$ always has a solution. This isn't quite a quadratic, so I can't take the discriminant of it yet, but I think we could manipulate it into a quadratic or something?
$G/H =\{gH,g\in G\}$
@Secret exactly. Then how can $a$ be an element of $G/H$?
Multiply both sides by the denominator?
14:23
@AkivaWeinberger yes
you will get:
O and btw, I realise one of my mistakes: $[a,b]=e \not \implies ab=ba$
@Secret that isn't a mistake.
@AkivaWeinberger I suggest you read everything from here:
in Room for Leaky Nun and Abcd, 18 hours ago, by Abcd
Find the value of a for which $\dfrac{ax^2+3x-4}{a+3x-4x^2}$ takes all values for all real values of x
$(a+4y)x^2+(3-3y)x+(-4-ay)=0$, yeah @Abcd
@LeakyNun Well, G/H has left cosets as elements, I guess the issue is that letting $a$ be one of these cosets is an abuse of notation?
@AkivaWeinberger yes
14:25
@Secret you can say so
@Abcd Right, so the point is, for whatever value of $y$ you give me, there's a value of $x$ that makes that equation true.
In other words, for whatever value of $y$ you give me, that quadratic has a root.
Hey Kasmir
@KasmirKhaan do you want group theory exercise?
@LeakyNun But what if we need to find some symbol to refer to one of the elements of $G/H$, will capitals be a good choice?
@Secret $aH$.
Does that make sense? @Abcd
14:28
ok then
@AkivaWeinberger does value of x= root? Root is only when f(x)=0
@LeakyNun Yes ofc =p , still need time with the last one tho =p
@AkivaWeinberger but root is only when f(x)= 0.
@Leaky so I guess I will now show that $\mathbb C^*\cong\mathbb C/\mathbb Z$. I don't need to answer your previous question for that, right?
14:29
@ShaVuklia my questions are separate
37 mins ago, by Leaky Nun
@ShaVuklia T/F: If $H < G$ and $K$ is a group such that $H \times K \cong G$, then $G/H \cong K$.
35 mins ago, by Leaky Nun
@ShaVuklia T/F: If $H<G$ and $K$ is a group such that $G/H \cong K$, then $H \times K \cong G$.
@KasmirKhaan ^
@LeakyNun Thank you ! :D
Seems like here $y\ne f(x)$ @AkivaWeinberger
@lea
@LeakyNun are those seperate excercices?
14:31
@KasmirKhaan yes
two T/F questions
@LeakyNun whats T/F?
true or false
Oh okay =p
$[G,G] \subseteq N \implies \langle [a,b]\rangle \in N$
$G/N \text{ abelian} \implies gNhN=hNgN$?
Well adding these 2 to that one
14:32
@Secret yes
@Leaky if I can find a surjective homomorphism $f\colon\mathbb C\to\mathbb C^*$ such that $\ker f=\mathbb Z$, then I'm done
i should give you an answer in 2-3 hours =p
I have to be afk for 30 mins
@Abcd Oh sorry, got confused with the variable $f$
you should probably put $\equiv$ or $:=$ or $\iff$ @Secret
14:33
@ShaVuklia what theorem is that?
@Abcd There are two functions here. There's the original function $\dfrac{ax^2+3x-4}{a+3x-4x^2}$
and there's the new function $(a+4y)x^2+(3-3y)x+(-4-ay)$
lol I got 245 rep today
The "new" function needs to equal $0$
and it's the thing that has a root
that's 45 above the rep cap
If $f\colon G_1\to G_2$ is a surjective homomorphism, then $G_1/\ker f\cong G_2$ @Leaky
14:34
We need to find an $a$ such that, for any $y$, that quadratic has a root.
@AkivaWeinberger How can we be sure that it has a root?
@AkivaWeinberger Understood this part.
2
Q: Minimize the power function with given constraint.

kamran jamilLet $x_1+x_2+\cdots+x_n=m$ then minimize the function $f(x_1,x_2,\cdots,x_n)=\sum_{i=1}^n (x_i)^{\alpha}$ where $x_i,m,n$ are positive integers and $\alpha>1$. My attempt: I applied the Lagrange's multiplier and found that minimum is obtained when all $x_i's$ are equal, but I am unable to pro...

anyhow, I guess I'll ask on the main site, because I'm clueless at this point
Do we get rep for chatting too @LeakyNun?
@Abcd no.
@ShaVuklia alright
260 rep now lol
14:36
@Abcd Because $\dfrac{ax^2+3x-4}{a+3x-4x^2}=y$ is the same as $(a+4y)x^2+(3-3y)x+(-4-ay)=0$
We want to find an $a$ such that, for any $y$, there's an $x$ that makes the first equation true.
But the two equations are equivalent.
So it's the same as wanting to find an $a$ such that, for any $y$, there's an $x$ that makes the second equation true.
If $x$ makes $(a+4y)x^2+(3-3y)x+(-4-ay)=0$ true, it's called a root of the quadratic $(a+4y)x^2+(3-3y)x+(-4-ay)$.
@AkivaWeinberger which implies that x makes the original function also true?
@ShaVuklia @AkivaWeinberger is there even any non-trivial homomorphism $\Bbb C \to \Bbb Z$ @_@
@AkivaWeinberger Understood clearly. I can solve from here. Thanks. :)
$gg^{-1}\langle[a,b]\rangle g h h^{-1}\langle[c,d]\rangle h = \langle[a,b]\rangle g\langle[c,d]\rangle h \in NgNh =gNhN$
bleh...
14:42
@LeakyNun $\Bbb C^\times$, you mean?
Multiplicative group?
@AkivaWeinberger never mind
@BalarkaSen hi
watch out your lolz guys, Balarka is here:P
replace all the lols by woahs
14:45
41 mins ago, by Secret
Error: Insufficient group theoric knowledge
Returning to infinities
loooooooolll (oh shit..)
I mean
whoooooaaahhhh
@ShaVuklia lol I answered your question
I mean, woah I answered your question
HAHAHAHAHAH
omg
yea I know
that's why I'm here
confused 'n all
1 hour ago, by Secret
An interesting question to ask in general is that: For a countable, linearly ordered set $S$ and some countable subset $X \subset S$, what property $X$ must have such that $S-X < \aleph_0$
Actually...
what's confusing about my answer?
14:47
like, why didn't you answer it in the chat :P but let me check your answer
lol not confused about that
at least, not yet
maybe
cofinite and cocountable sets requires more study in topology
should the latter read "untable"?
Only if the untable is finite dimensional.
what is an untable?
It's a joke on double dual being isomorphic to itself.
"co" and all.
14:51
inb4 coconuts inb4 ffee
@Secret it's a joke interpreting "countable" as "co-untable", so the dual version should be "untable"
inb4 a
lol
Anyway, before going back to to chemistry, let me put down a conjecture:
Some conjecture:
@ShaVuklia you should just consider the other answer
lol yea I was doing that
14:55
you've accepted it now
I've just finished checking his answer
yes true
so my questions?
I'll skip for now. I have a lot of exercises left :P
sorry, time pressure
unless you have like a proof somewhere online
then I'm willing to read it
but I'm not going to do it myself now
14:56
do you want the answer?
both are false
oh right
well then I don't even have to bother looking for a proof :P
I'll just ditch that nonsense :P
just have to look for counterexamples :)
@ShaVuklia find a counterexample
14:58
I'm first going to finish my compulsory exercises :P
you're just trading one co-ex for another co-ex :)
Let $A,B$ be infinite sets with $A \subset B$. Then $|B-A|<\aleph_0$ if $A$ is cofinite, $|B-A|=\aleph_0$ if $A$ is cocountable and $|B-A|=\aleph_{\alpha}$ if error, circular definition
How to move to next line instead of sending message again and again?
Shift-enter
15:03
Ok
@AkivaWeinberger sad
everyone knows England is not a country
it's a city
ah, a demonstration of the one-color theorem
The joke is New Zealand does not exist
bleh, should've taken my allergy med an hour earlier. I always forget how miserable September is without it
September is the cruelest month / bringing allergies out of the dead land ...
3
@BalarkaSen I approve.
heh, now I've got fragments of The Waste Land bouncing around in my head
It's definitely the kind of work that sticks around with oneself
@Leaky lol a classmate of mine responded that $\mathbb T=[0,2\pi)$. I was like: no son :P your making your life too easy.
ehhh, that's not really so different
$e^{i \theta}e^{i\phi}=e^{i(\theta+\phi)}$
so multiplication of complex numbers in modulus 1 is isomorphic to addition of angles in [0,2pi)
15:19
I know, but the point is that he just said it's a group that fits him in this exercise
hm. Well, depends on what definition you have I guess
yea
ok, it hurts a bit to have a starred message with a grammar mistake that you can't fix anymore:P but whatever. I'll just dab if I have to.
lol, I read a comic, yay
:l
$y^2(1-4a)+y(2a+12)+(a^2-12) \ge 0$
Does this imply $1-4a>0$?
Why or why not?
PS: This is a part of another big question I am solving.
15:26
@ShaVuklia well they're isomorphic
is $\Bbb R^\ast$ isomorphic to $\Bbb R \times \Bbb Z_2$?
I knoooooww:P
but I had a point, ok :P
@ShaVuklia alright
you mean $\mathbb R_{>0}\times\mathbb Z_2$ ?
@ShaVuklia no.
anyhow. I just finished an exercise that took me 2 seconds, while this previous one took me 2 hours:l how am I supposed to make a schedule, when the time it takes to do an exercise can be so arbitrary.
I donno if they are
15:28
Is $\langle \Bbb R^\ast , \cdot \rangle$ isomorphic to $\langle \Bbb R , + \rangle \times \langle \Bbb Z_2, +_2 \rangle$?
that was an isomorphism that popped up in my mind just now when I was taking shower
so it IS isomorphic :P
3 mins ago, by Abcd
$y^2(1-4a)+y(2a+12)+(a^2-12) \ge 0$
Does this imply $1-4a>0$?
Why or why not?
PS: This is a part of another big question I am solving.
the real question is, show it
@ShaVuklia not going to give away :P and I am not 100% sure
Ignore the "why/why not?" in the message/
15:29
shower thoughts, you know
loooll
why don't you just give the alleged isomorphism
and then I'll try to check if it is indeed an isomorphism
@LeakyNun Are you actually asking
@ShaVuklia because I also want you to try
@AkivaWeinberger half asking half verifying
I think it's yes
Note that $\langle\Bbb R_{>0},\times\rangle$ is isomorphic to $\langle\Bbb R,+\rangle$
@AkivaWeinberger ugh, don't give it away
15:31
:D
yea I agree too then
:'(
no fun
much fun
:P
@LeakyNun I thought you just said you weren't sure if it was correct
If it was a puzzle for Sha you should've said that
no worries @Leaky, I still have 2 exercises to go for this chapter. so there's enough to do for me for today :P
@AkivaWeinberger ><
I never thought $\Bbb R^\ast$ could be isomorphic to any addition groups lol
15:33
well technically it isn't right
it's $\mathbb R_{>0}$
@ShaVuklia ??
5 mins ago, by Leaky Nun
Is $\langle \Bbb R^\ast , \cdot \rangle$ isomorphic to $\langle \Bbb R , + \rangle \times \langle \Bbb Z_2, +_2 \rangle$?
right, but that's a direct product
but alright
I get your point
what?
never mind
15:34
I'm not referring to $\Bbb R_{>0}$.
yea nm:P
This is also kinda why $\ln$ is multivalued
R^* = R-{0} ?
yep @Semi
15:35
Yeah, with multiplication as the group operation
it's a strange isomorphism
think of the lines $y=\pm1$
and then they are isomorphic to the x-axis sans origin
Does anyone happen to know of a free and good integral calculator? Need to approximate an integral out 12 places, and WA doesn't manage
Does writing "to 12 decimal places" at the end work?
gp/pari has a relatively good numerical integrator but i forget the comments
At the end of the W|A query
15:43
commands
Maybe Google "numerical integration online"
@AkivaWeinberger I don't really trust it...
I'm wary that WA might be trustable only to the places it shows, I can get it to show more places, but... :-/
@MikeMiller intnum I think
16:00
So $[\Bbb R^\ast:\Bbb R_{>0}]=2$ :O
Guys. Let $G_1,G_2$ be groups. I have to show that $(G_1\times G_2)/H_1\cong G_2$, where $H_1=G_1\times\{e_2\}$. The exercise explicitly says that I need to use the following theorem to show this: Let $G$ be group, and $N\subset G$ a normal subgroup, $H\subset G$ a subgroup. Then $H/(H\cap N)\cong HN/N$.

My attempt: Let $N=G_1\times\{e_2\}$ and $H=G_1\times G_2$. This would give us $(G_1\times G_2)/(G_1\times\{e_2\})\cong HN/G_1\times\{e_2\}$. Now it seems to me that $HN=G_1\times G_2$, so we get a trivial result. I don't see how I can use this theorem to show it (I've already shown it wit
Let $H=\{e_1\} \times G_2$ and $N=G_1 \times \{e_2\}$
ohh
say no more
ain't planning to :P
hehe
yay
16:30
@ShaVuklia have you found the counterexamples?
no I'm still doing my regular exercises
16:44
@Leaky
are you up for a final exercise? :P
Let $G$ be a group, and let $N$ be a normal subgroup of $G$. Let $G/N$ and $N$ be abelian. Show that for each subgroup $H$ of $G$, there exists a normal subgroup $N'$ of $H$ such that $H/N'$ and $N'$ are abelian. Well, obviously, $N'$ is going to be $H\cap N$. So $H\cap N$ is abelian.
All I need to show is that $H/(H\cap N)$ is abelian. One thing I thought I might use is that $G/N$ is abelian implies that $[G,G]\subset N$. So I'm guessing I need to show that $[H,H]\subset H\cap N$. Maybe I could show that $H\cap[G,G]=[H,H]$? I'm not sure.
16:56
$[H,H] \subseteq [G,G] \subseteq N$
$[H,H] \subseteq H$
@ShaVuklia
lol how does that help
that says nothing about $H\cap N$
errrrr
oh it does? :P
try again
@ShaVuklia draw a venn diagram based on the two information I gave you
16:58
ohhhh yea
hahah
oops

« first day (2588 days earlier)      last day (2731 days later) »