« first day (3066 days earlier)      last day (1969 days later) » 

1:26 AM
Why must a function $f:(0,1) \to [0,1]$ such that $f$ is continuous and surjective exist?
 
@user330477 I'm not sure if I could come up with a non-constructive way to see it, but there's an elementary example you can construct.
 
@Fargle Is the example you have in mind something simple like involving polynomials, exponentials, etc?
 
Yeah
 
Piecewise linear
 
@Fargle I think whatever you have in mind is probably a piecewise function each of whose parts are piecewise linear?
 
1:33 AM
Cut the domain at 1/4, 1/2, 3/4
 
Actually, no, but piecewise linear totally does work.
The one I have in mind is in fact differentiable on its domain.
 
Ah I see
 
@Fargle @MikeMiller Actually the piecewise function I happen to construct is bijective (more than surjective) but not continuous.
This is the function I have in mind $f(x)=\frac{1}{n+1}$ if $x=\frac{1}{n}$ and $f(x)=x$ otherwise.
 
I disagree that this is surjective---what maps to 0? to 1?
 
@Fargle This is the function I have in mind $f(\frac{1}{2})=0$, $f(x)=\frac{1}{n-2}$ if $x=\frac{1}{n}$ for $n>2$ and $f(x)=x$ otherwise.
 
1:45 AM
This just moves the problem though---now, what maps to 1/2?
 
I don't get your point anyway. Why are we talking about discontinuous functions?
 
Yeah, I'm not sure how this gets at your original question
 
@Fargle @MikeMiller This is my problem. I am a having a hard time finding a function which is both continuous and bijective.
 
The problem says surjective, not bijective.
 
So as not to send you searching endlessly, there is no continuous bijection $(0,1) \to [0,1]$.
 
1:50 AM
@Fargle But why?
 
I don't want to give it away, but I'll give two hints:
1. By invariance of domain, a continuous injection from an open subset of R into R has to be a homeomorphism onto its image.
2. Try to use the sequence formulation of continuity to reach a contradiction.
 
That wasn't your original question. Anyway, a continuous injection defined on some interval is increasing. (Prove this yourself.) Show that f(x) = 1 is therefore impossible.
@Fargle InvDomain way too big here
 
I like nuking flies. :(
I like your way better though.
 
What I said constitutes a proof of invariance of domain in 1D
 
Yeah.
 
1:55 AM
Sorry, it should say "a continuous injection is either increasing or decreasing".
 
Bob
2:08 AM
could somebody who is good with ODEs look at my post: math.stackexchange.com/questions/3051716/…
is there a bijection (0,1)→[0,1]
it seems to me that has to because $|(0,1)|$ = $|[0,1]$
Do I have that right?
 
There is indeed a bijection. In fact, that's why we know that those two sets have the same cardinality
Because that's just the definition of cardinality
 
Bob
do you know of one?
 
It's something close to what the other guy was getting at
Identity on most points, shift 1/n's by 2, and let 1/2 map to 0, 1/3 to 1
Seems that's what their example became, actually
 
Bob
@Fargle that makes sense, thanks
How are you with ODEs?
 
Not good at all, sorry.
 
Bob
2:23 AM
that is okay
I hope you had a Merry Christmas
thanks your help
good night
 
Merry Christmas. :)
 
 
2 hours later…
4:49 AM
hi anyone on here
 
Wish everyone a very merry Christmas! :)
 
merry Christmas
 
:)
A cayley graph with respect to a generator set S has a degree of $2\vert S \vert$, only if S has involution- free elements, right?
 
question about set theory: if A is an element of B and B is not included in C, then A is not an element of C is not necessarily true, correct?
because: B = {A, 1} and C = {A,2}, so B is not included in C and A is an element of C.
 
 
7 hours later…
12:18 PM
@MacroGuy Do you think there's a problem with this reasoning?
 
1:13 PM
@MacroGuy That looks right.
 
1:53 PM
Inclusion is not a transitive relation @MacroGuy. What you've said proves this.
 
uhh.. yes it is
membership is not a transitive relation
 
2:10 PM
@MacroGuy If B is not a subset of C, then there exists an element of B which is not an element of C.
 
$B\subseteq C$ $\Leftrightarrow$ $(\forall x) (x\in B\Rightarrow x\in C)$
 
Also hey Martin
I saw a link in the MO chat to something here
 
And you can get by negation:
 
and then I saw this chat was kind of active
What was that graph in the MO chat, I didn't understand the conversation
 
2:12 PM
$B\nsubseteq C$ $\Leftrightarrow$ $(\exists x) (x\in B\land x\notin C)$
 
Is it about how all the chatrooms died because it's christmas?
 
Wow, did I just get told to read up =(
 
@HarryGindi I've posted this for others, who might not be familiar with using MathJax in chat. Surely, for a chat-veteran like you, this is a familiar thing.
 
2:29 PM
Random meme:
Yo mama so impossibly fat that she is a proper class
 
Is the most 'abstract' definition of a vector space $V$ over a field $F$, simply that $V$ is an $F$-module?
 
That's just another way of saying the same thing
 
@Secret what is the difference between meme and a joke?
 
Jokes aren't funny.
I read it in an article once
 
And memes are?
Since when?
 
2:37 PM
@HarryGindi Yes I erred.
 
Someone maybe knows if there an easy proof that amorphous set is Dedekind finite?
 
@Holo D-infinite is equivalent to having a countable subset
@Perturbative A vector space over F is a unital ring homomorphism from F to the endomorphism ring of an abelian group V
 
2:52 PM
Too easy... I completely forgot I can use... Well anything
 
3:52 PM
hi @BalarkaSen
 
did you see my question?
 
yes
 
and?
 
i dont have useful comments
 
3:52 PM
ok thanks
 
4:17 PM
let $R$ be a non-commutative ring with unity, and let $r \in R$, and let $(r)$ be the left ideal $\{xr \mid x \in R\}$
for a left $R$-module $M$, what is $\operatorname{Hom}_R(R/(r),M)$?
 
4:51 PM
Let $X=Cay(G,S)$, be a cayley graph of a group G, with respect to a generator set S. Then, $degree(X)= 2 * \vert S \vert $, right?
If S does not have any elements $s \in S$, such that $s^2=1$.
 
What's the degree of a graph?
 
Then can we say $degree(X)= 2 * \vert S \vert $, if S has involution-free elements?
The degree is same as valency of a graph
The maximum number of edges incident on a vertex
I just wanted to know whether using the term "involution-free" is correct?
 
Problem: If $J$ is uncountable, then $\Bbb{R}^J$ is not metrizable. Attempt: If $\Bbb{R}^J$ were metrizable, then the subspace $X := [0,1]^J$ would also be metrizable. In fact, it would be a compact metric space by Tychonoff's theorem. If so, then it is limit point compact and the set $A := \{e_i \mid i \in J\}$ has a limit point in $X$, call it $(x_i)$....
Here's where things breakdown. My strategy was to show that $A$ cannot in fact have a limit point. I was able to show that $(x_i)$ cannot be in $A$, but unfortunately it seems that $0$ is a limit point of $A$. Is there any way of salvaging the proof?
 
 
1 hour later…
6:16 PM
@LeakyNun omg that big hat I adore it!!!
 
lol
omg i'm an idiot, $\operatorname{Hom}_R(R/(r),M)$ is still $M[r] := \{x \in M \mid r \cdot x = 0\}$
 
Yeah.. I noticed that too...
 
hmm that can't be true
let $x \in M$ such that $r \cdot x = 0$. then for any $cr \in (r)$, $cr \cdot x = c \cdot (r \cdot x) = c \cdot 0 = 0$
maybe $M[r]$ isn't a submodule of $M$ then
 
 
1 hour later…
7:27 PM
In 3D, four line segments in a loop do not necessarily determine a quadrilateral (e.g. if they're not coplanar).
As such, the rule that the angles sum to 360 no longer works.
Is there a way to fix this, by adding a term that measures the twisting at each edge?
Hm. I believe you can…
Thinking in 3D is hard…
…I am suddenly less sure
 
7:49 PM
Hey guys!
Does anyone know where I can find examples to build a regular expression of a deterministic finite automaton that has at least two final states?
Because I am struggling with the following question:
2
Q: Given a transition table do digraph, determine if it is DFA or NFA and build grammar

manoooohFor the next transition table: $$\begin{array}{|c|c|c|c|}\hline&0&1&2\\\hline a&a&b&d\\\hline b&a&b&c\\\hline c&c&d&a\\\hline d&c&c&a\\\hline\end{array}\\a\text{ is initial state}\\\{a,d\}\text{ are final states}$$ Make the digraph of the finite automaton and indicate if it is DFA (determinist...

 
 
2 hours later…
Eli
9:27 PM
hey peeps, is it possible to solve an equation, to find x when there's rounding aspects to in the equation?
 
2
Q: A closed ball in $l^{\infty}$ is not compact

Bear and bunnyDefinition of compact in Real Analysis, Carothers, 1ed said that: In Example 8.1 (c), he claimed that closed ball $\{x: \|x\| \leq 1\}$ in $l^{\infty}$ is not compact. Why?

See GA316's answer. In it, he makes the following claim: "Actually you can prove stronger result that, they are discrete. But the only discrete subsets of a compact sets are finite." Isn't this false? The set $\{\frac{1}{n} \mid n \in \Bbb{N}\}$ is an infinite discrete subspace of the compact space $[0,1]$, right?
If he/she is saying what I think he/she is, I believe it is false...right?
 
9:45 PM
@LeakyNun first of all, the notation $(r)$ doesn't make sense in the noncommutative setting, so let's use $R/Rr$ instead. You're right that $M[r]=\{x \in M \mid r \cdot x =0\}$ isn't necessarily a submodule. But it is still true that $\mathrm{Hom}_R(R/Rr,M)=M[r]$. In general there is no reason to expect that $\mathrm{Hom}_R(M,N)$ is a $R$-module for two (say left) $R$-modules $M,N$
 
@MatheinBoulomenos ah you're here!
 
did I send you my question?
 
1
Q: $\operatorname{Ext}^\bullet_R(R/rR,M)$ and $0 \to A[r] \to B[r] \to C[r] \to A/rA \to B/rB \to C/rC \to 0$

Kenny Lau$\newcommand{Ext}{\operatorname{Ext}}\newcommand{Hom}{\operatorname{Hom}_R}$Let $R$ be a commutative ring with unity and $r \in R$. Applying snake lemma to the following diagram: $$\begin{array}{c} 0 & \longrightarrow & A~~ & \longrightarrow & B~~ & \longrightarrow & C~~ & \longrightarrow & 0 \\ &

 
9:47 PM
Hey @Mathein @Leaky :)
 
Hey @ÍgjøgnumMeg
 
hmm if $\operatorname{Hom}_R(M,N)$ isn't an $R$-module then what is Ext doing?
 
it's still an abelian group
 
but what exactly is Ext a cohomological extension of?
 
Ext is still a derived functor of the Hom functor
it's just that the Hom functor has values in abelian groups
 
9:50 PM
uw0tm8
my life is a lie
 
@Mathein how was your Christmas? :P
 
very nice, thanks!
and youself?
 
Yeah it was cool, didn't really do much but drink and eat.. of course
did some cycling etc.
with all this free time lol
 
@MatheinBoulomenos froeliche weihnachten?
 
danke!
man würde eher "frohe Weinachten" sagen, aber frag nicht, warum :P
 
9:54 PM
wollen Sie meine Frage sehen
denk, dass nur Sie konne' tue' es
@ÍgjøgnumMeg willst du meine Frage sehen?
gott kvold?
 
Auf jeden Fall, heißt aber nicht dass ich eine Ahnung hab
ja gott kvøld !
Also ich schreibe gerade Stipendienbewerbungen
so langweilig :(
 
German is the most beautiful language on earth.
 
"is" might need some qualification there
 
10:19 PM
@LeakyNun about your question: if $r$ is not a zero divisor, then your six-term exact sequence is just the exact sequence $0 \to \mathrm{Tor}_R^1(R/rR,A) \to \mathrm{Tor}_R^1(R/rR,B) \to \mathrm{Tor}_R^1(R/rR,C) \to R/rR \otimes_R A \to R/rR \otimes_R B \to R/rR \otimes_R C \to 0$
 
I've also noticed this pecularity
the tor sequence and the ext sequence are the same for R/rR, r not zero-divisor
 
What's the easiest way to show that $x \mapsto (\cos (2 \pi x), \sin (2 \pi x))$ is a quotient map from $\Bbb{R}$ to $S^1$? Do it directly or show that it is a covering map, and then show every covering map is open?
 
what's a quotient map?
 
Eli
0
Q: How to: Step by step equation solving

EliI have a few equations that needs solving, but I'm scratching my head because it involves rounding ups and rounding downs. Plus, it's been like a decade ago since I've done this kind of thing. I have realized that there ain't good libraries online to solve any equation given and seeing that I'm ...

 
@LeakyNun $p : X \to Y$ is a quotient map if $p$ is a continuous surjection such that if $p^{-1}(B)$ is open, then $B \subseteq Y$ must be open.
 
10:25 PM
hmm
I mean, it can get quite deep depending on the things you're allowed to assume
or maybe we can use some compact->hausdorff trickery
 
@user193319 overkill approach: the map is a surjective continuous group homomorphism from a locally compact $\sigma$-compact Hausdorff group to a locally compact Hausdorff group, hence it is open
 
see, I'm right on some terms :P
make your second "locally" disappear
 
why?
I was quoting the general theorem
 
I'm just jealous :P
anyway what do you think of my question?
 
it's interesting but I have no idea
 
10:29 PM
also doesn't sigma-compact imply locally compact?
@MatheinBoulomenos great... who else can I ask
@MatheinBoulomenos maybe I should add the tor-ext sameness to the list of questions that nobody will answer
 
maybe try MO?
 
how do I do that? do I flag for migration?
 
@LeakyNun $\Bbb Q$ is sigma-compact, but not locally compact
 
I don't know if I should make my first MO question an open-ended one...
@MatheinBoulomenos what?
aha
very sneaky
 
Eli
Can somebody pls tell me why this question has been downvoted? Am I missing something, or do I need to elaborate?
https://math.stackexchange.com/q/3053372/333955
 
10:44 PM
@MatheinBoulomenos so $rA$ isn't a submodule of $A$?
 
@Eli you could solve them without the rounded numbers then try the integers next to the solution
 
$rA$ is a right submodule of $A$
 
@MatheinBoulomenos so I can't form left submodules?
my life is sad
 
well, $Ar$ is a left submodule
$ArA$ is both a left and a right submodule
 
oh and what's the map $H^2(G(K/F), K^\times) \to H^2(G(L/F), L^\times)$ for $L/K/F$?
@MatheinBoulomenos oh $A$ is a left module
 
10:55 PM
ah, then $rA$ is just a subgroup of $(A,+)$
 
my life is hell
 
if $R$ is commutative though or more generally $r$ is central, then $rA$ is a submodule
@LeakyNun inflation
 
why does CF say restriction
 
cassels-frohlich
 
10:58 PM
restriction is a map $H^2(G(L/F),L^\times) \to H^2(G(L/K),L^\times)$
 
and inflation?
 
$H^2(G(K/F), K^\times) \to H^2(G(L/F), L^\times)$
more generally if $H$ is a (closed, for continuous cohomology) normal subgroup in $G$ and $A$ is a $G$-module, then $A^H$ is a $G/H$-module and inflation is a map $H^\bullet(G/H,A^H) \to H^\bullet(G,A)$
 
what next, they're adjoints?
 
no, but there's the inflation-restriction exact sequence
where do you think that CF is using restriction where it doesn't make sense?
 
P.131 (P.74 on pdf)
 
11:09 PM
but the situation with $\mathrm{Br}(K)$ and $\mathrm{Br}(L)$ is different. we have $L^{sep}=K^{sep}/L/K$ and the restriction map $H^2(G(K^{sep}/K),K^{sep})=\mathrm{Br}(K) \to H^2(G(L^{sep}/L),L^{sep})=\mathrm{Br}(L)$
 
sneaky
 
note that the proof of Cor. 2 is actually the inflation-restriction sequence + Hilbert 90
see prop 1.8.11 on p.25 here: math.ucla.edu/~sharifi/groupcoh.pdf
 
@MatheinBoulomenos vielen Dank
 
All around me I see nerds
 
then you have perfect eyesight!
 
11:35 PM
@Daminark All around me are familiar faces
 
There is nobody around me.
 

« first day (3066 days earlier)      last day (1969 days later) »