« first day (3250 days earlier)      last day (1761 days later) » 
00:00 - 15:0015:00 - 23:00

12:48 AM
76284: One more of these posts and you are out. I do not interact with machines
 
To whom is that addressed?
 
His chat log indicate he has a long history of asking trivia type questions that are googlable
with the exception of game theoretic questions
 
I'm not sure that makes him/it a machine.
Think of all the prealgebra, algebra, calculus, and algebra problems that people post all the time ... all of which have answers everywhere.
 
I... don't know, but the way he asks questions seemed to just want an answer and no thing else, which is quite different from other users who does similar things...
 
It's hard to know without interacting IRT.
 
12:53 AM
I guess you are right. Well if the chat is fine with that, then it should be fine
 
The trouble with these "user ...." names is that it's hard to recognize if one has encountered them before ...
I only remember a few.
 
yeah, one need to go quite a deep search to id them via their chat patterns. Not many are as easily identified as 21820 (the logician)
 
I guess I'm just not interested in working that hard.
 
That's pretty normal, its much better to spend time on maths than trying to piece together a history of interaction for er... curation purposes...
 
1:13 AM
@Secret What’s your problem?
How is that a “trivia type question”?
@Secret What gave you that impression?
 
 
1 hour later…
2:42 AM
The list of questions looks so random and pairwise unrelated to each other. Even JEE and Ultradark's questions follows a certain theme. I think my reaction is definitely too overeactive in response to that weird feeling
@user76284 I guess the only fair thing I can say is I never seen such a question pattern where each question does not seemed to be remotely related to the previous one, and possibly that weird feeling lead me to misinterpret as repetative and trivia like (regardless of the truth that none of the questions are actually trivial)
 
3:04 AM
If you're looking for themes, the most recent ones are related to regularization of divergent series. The others are related to representations of ordinals and generalizing their arithmetic to subtraction and division (Grothendieck group of a commutative monoid and field of fractions, respectively).
The game theory stuff is a third theme, though somewhat related to the ordinals (through nimber arithmetic).
They finally added a built-in ComplexPlot function to Mathematica. Nice.
 
3:20 AM
I see
 
4:08 AM
https://math.stackexchange.com/a/84646/76284

I wonder if there's a way to "break through" this formidable barrier of poles.
 
4:22 AM
Had no background on that level of complex analysis, but if those nontrivial poles are only dense, then in theory certain transcendentals may be able to allow some contours to slip past
still, that they have something to do with the nontrivial zeros seemed to suggests it is going to be riemannian hypothesis difficulty in getting past them
 
5:08 AM
I don't know what to do even after reading the hint. Problem source: Visual complex analysis.
 
well, it can be anywhere depending on whether the analytic function reverses orientation
 
What is point of asking the question, then?
 
Hold on
 
Reverse orientation implies reflection, I guess. Then , no.
 
I think you'd have to use the complex conjugate to reverse orientation.
So just think about what happens with $z \mapsto 1/z$.
Around the unit circle.
 
5:15 AM
$ \bar z$ is not analytic
 
Yep.
 
@user76284 what do mean what happens with?
 
If $z$ moves counterclockwise around the unit circle ($|z| = 1$) starting at $z=1$, where does $1/z$ move?
 
$|z|$ is preserved. What happens to $\arg z$?
$1/z = 1/(|z|\exp(i \arg z)) = 1/|z| \exp(-i \arg z)$
 
5:18 AM
@user76284 reflection like conjugates
 
?
It moves in the opposite direction if that's what you mean.
 
My statement was argument of $1/z$ is $ - \theta$ is $\theta$ is argument of $z$
 
Another way to look at it is to think about the Riemann sphere if you're familiar with that. $z$ always rotates counterclockwise with respect to the center of the shaded region.
Yeah, that's right.
 
In the book, reimann sphere hasn't been introduced yet.
By the way, why you are talking about $1/z$ ?
 
Yes, $1/z$ is an example of what that map could be (since it maps the inside of the unit circle to the outside and viceversa).
Since analytic functions preserve orientation, we know there's a single right answer, so you just need to think of an easy example.
 
5:22 AM
is $1/z$ analytic?
 
I'll try to draw a picture, one sec.
Well, it's meromorphic.
Analytic everywhere except at the pole ($z=0$).
Wait, is the function analytic everywhere?
 
5:37 AM
My short reasoning is this: On the Riemann sphere, $z$ is rotating counterclockwise with respect to the shaded region, and clockwise with respect to the unshaded region. Therefore, its image $w$ will also be rotating clockwise with respect to the unshaded region.
Another way to think about it is to draw an arrow perpendicular to the direction of motion, say to the right, which for $z$ would point into the unshaded region. Such an arrow will continue to point into the unshaded region for $w$.
 
$1/z$ is analytic as a self-map of the Riemann sphere. You'd call it meromorphic as a complex function on $\Bbb C$.
 
Apologies for the terrible drawing.
Here's a better one.
You can see that moving counterclockwise (say, around the red equator) with respect to 0 (i.e. the interior region) is the same as moving clockwise with respect to infinity (i.e. the exterior region).
 
Your reasoning makes sense.
 
6:24 AM
According to jstor.org/stable/44237555 "Borel has shown that there exists a kind of analytic continuation which differs from the usual kind (the theory of the usual kind is due to Weierstrass), and Borel made use of examples such as the ones studied here in showing that in some cases it is possible to continue beyond what Weierstrass called natural boundaries".
Where can I read more about this type of continuation? They cite Borel's 1917 Leçons sur les fonctions monogènes d'une variable complexe. Is there an English translation somewhere?
 
@cdt lol
 
6:38 AM
@BalarkaSen i'm proving that the fundamental groupoid exists
imagine doing all those homotopies by hand
 
7:08 AM
Did you decide whether that theorem is true? @Leaky
 
@AlessandroCodenotti no
 
I think it's false and only holds in the $\sigma$-finite case
I'm sure it holds in the latter case
 
@LeakyNun ? which homotopies
oh like identity/associativity and shit
 
yeah
 
u can draw a picture for those proofs
its really about concordance of configuration of points on $I$, if you want
 
7:19 AM
@BalarkaSen i'm formally verifying it
 
like in lean lol
 
Is @BalarkaSen drinking lean
 
@BalarkaSen exactly
 
yo @Slereah
 
yo
 
7:21 AM
@BalarkaSen and I came across this amazing thoerem
continuous_if :
  ∀ {α : Type u_1} {β : Type u_2} [_inst_1 : topological_space α] [_inst_2 : topological_space β]
  {p : α → Prop} {f g : α → β} {h : Π (a : α), decidable (p a)},
    (∀ (a : α), a ∈ frontier {a : α | p a} → f a = g a) →
    continuous f → continuous g → continuous (λ (a : α), ite (p a) (f a) (g a))
 
if you say so my dude
 
I think the last two lines are readable
the last term says that the function if p(a) then f(a) else g(a) is continuous
 
ok gluing lemma
I think thinking of the crux of $\Pi_{\leq 1} X$ being a category as a fact about concordance of configurations of points on $[0, 1]$ has some merit to it - for example $A_\infty$-spaces are something you get by forgetting about homotopy-coherence
Aha, it's called the "little intervals" operad
 
Morning boiz
 
7:38 AM
hey... morning fella
 
Hey @Adam
 
Remember, an important part mental health is therapeutic behavioural exercises, for example offline trolling of humours institutions like scientology, there is one in almost every city, and they are there to make life a pleasure to live, remember it's ok to hurt their feelings and play tricks on them, they are differently abled, and much like a goldfish resumes orbit seconds after being frightened by something, so will they too meg, so will they too.
humorous*
 
Morning @ÍgjøgnumMeg
 
It's actually 4 pm but I did wake up just now so relative morning I say to all
 
7:54 AM
Hi @Mathein
@Adam I dunno if you're being ironic or not
lol
 
Good morning.
 
Morning @Rithaniel
 
8:09 AM
@Mathein what do you recommend taking as an applied module? lol
 
So, sanity check: In any integral domain, if $ab=bc=-1$, then $a=c$, right?
 
(to avoid as much applied mathematics as possible)
@Rithaniel integral domains always have the cancellation property
 
@ÍgjøgnumMeg probability theory 2 counts as applied
 
@Mathein and it's just measure theory or what?
 
8:11 AM
Fair
 
@ÍgjøgnumMeg Asking Mathei, who took all the master courses instead of finishing his bachelor because the applied courses are ugly :P
 
Yeah, but are they always commutative?
 
@Rithaniel yes
an integral domain is commutative
by definition
 
@Alessandro that's true
I still need to take some applied undergrad courses, haha
 
8:12 AM
yeah
 
most of my undergrad was applied
I had maybe 4 pure courses over 3 years
:(
 
but I can take as much master courses as I want now and it all counts towards the master
 
By definition? I'm only familiar with the definition that says a ring is an integral domain if $ab=0\implies a=0$ or $b=0$.
 
@Rithaniel if $ab=ac$, then $a(b-c)=0$
so if $a \neq 0$, then you have $b=c$
integral domain includes commutativity, yeah
 
"domain" usually has commutativity built in
(think of them as subrings of their fields of fractions, for instance)
 
8:17 AM
Well, if an integral domain is commutative then the original thing I posted is trivial. I was getting hung up on if the elements didn't commute.
 
Right
I guess "non-commutative integral domains" would be the unital subrings of division rings?
brb, doing my job :(
 
Now I feel like I need to read through all the definitions again.
 
@ÍgjøgnumMeg nope, that's actually wrong
 
Also, hopeful work is enjoyable, today, Meg.
 
localizations in the non-commutative setting are hard
and don't always exist
it is kinda surprising that there exists a non-commutative domain $A$ such that $A$ cannot be embedded into any division ring $D$
 
8:29 AM
Hmmm, really? Do we know what $A$ looks like?
 
there's a construction in Lam's "Lectures on Modules and Rings", starting on page 291, but it's a bit involved
 
I need to get caught up on reading, honestly. Summer rolled around and I started to slack off too much.
 
the idea is that there exists a monoid $H$ such that the monoid algebra $k[H]$ is a domain, but $H$ has some strange properties such that $H$ can't be embedded into a group
now if we had an embedding $k[H] \hookrightarrow D$ into a division ring, then that would induce an embedding $H \hookrightarrow D^\times$ into a group
but I don't want to spell out the details of the contruction of $H$ and of the proof that $k[H]$ is a domain here
 
This is a book, right? Lam's "Lectures on Modules and Rings"
 
8:37 AM
That will be added to the list, then.
 
it is recommended that you read Lam's "A First Course in Noncommutative Rings" first, as the other book is a follow-up to that book
 
There's another book I need to read, too. I forget the author's name, but it's titled "Commutative Semigroup Rings."
 
@Mathein fair!
I don't know anything about non-commutative algebra
lol
 
it's a fun subject imo :)
 
as soon as you start appending left and right to various subobjects I start crying
 
8:42 AM
lol
 
@Mathein I suppose it probably is, I know Venjakob does some kind of non-commutative iwasawa theory
 
it's also useful if you want to do rep theory as the group algebra $k[G]$ is non-commutative if $G$ is
 
Nice :)
 
also in group cohomology, you work with $\Bbb Z[G]$-modules
 
@Leaky I just asked it on main since I couldn't find a counterexample myself math.stackexchange.com/questions/3276858/…
 
8:43 AM
Yeah, and if $G$ is non-abelian then that guy is non-commutative?
 
@AlessandroCodenotti nice
 
the elements of $\Bbb Z[G]$ are formal sums of elements of $G$ with coefficients in $\Bbb Z$ right?
 
Also I'm quite sure Cohn's proof goes through with s-finite spaces rather than $\sigma$-finite ones
 
right, and you just extend multiplication $G \times G \to G$, bilinearly
 
8:46 AM
and addition is obv
Cool :P
 
some "magic" theorems in rep theory fall out of some basic non-commutative algebra results, namely Artin-Wedderburn
 
Do you know if there's a rep theory course running this semester? :P
or is that undergrad in Heidelberg?
 
hmm
not sure
there are some undergrad seminars offered irregularly
and sometimes rep theory with a focus on Lie groups from a mathematical physicist
 
Fair, I'm so unsure of what I do and don't know that I don't know what to take hahaha
 
Though if you want to take an undergrad seminar as a masters student, you can ask for a more difficult talk and they'll count it as a master seminar
I've seen that happen before
 
8:53 AM
Well I have a feeling I'm gonna need to take one or two undergrad courses anyway, even if I don't take the exams
just for the background :P
 
I think you get credit for one undergrad course
I really recommend "Algebra 2" next summer
 
Fair, but only certain courses
 
gives a lot of background
 
Cool, I planned on taking that
I know Lie Algebras is credited
 
actually Lie Algebras is offered next term
 
8:56 AM
I'd like to take that :P
 
I might take that as well
 
I just have huge gaping holes in analysis and topology hahaha
 
@ÍgjøgnumMeg algebraic topology is really fun
and it starts from 0, you don't even need to know point-set beforehand
 
Yeah, I've tried to learn point-set topology but it's so painfully dry
Ah nice :)
 
I don't think that's offered soon, though
the topology prof usually alternates between one year alg top and one year diff top
so the next two terms are likely differential topology
you can still take alg top in your third semester, though
Nevermind, algebraic topology I is offered next term!
 
8:59 AM
That's fine, as long as I'm not severely disadvantaged
Oh nice!!!
 
hahaha
I'd like to see the vorlesungsverzeichnis
How do you know it's offered?
 
it's not officially published
there's kind of a hack ...
 
you go to "Suche nach Veranstaltungen", then you search for something
then you delete "veranstaltung.semester=20191&" in the url on the results page
that gives you the results for all semesters
 
9:01 AM
Ha nice
 
including the next one
No idea who discovered that or why they haven't fixed it, yet, that worked since a long time now
 
and where does it say which semester they are offered in ? lol
 
nowhere
you have to click the links
it's a bit painful
 
Right, but the one you linked says WS 2018/19
not 2019/20
 
oops
nevermind
then alg top isn't offered after all
 
9:03 AM
life ruined
 
hahaha dw
 
I thought 2018/19 was the next semester lol
numbers are hard
 
yeah
agreed
 
I guess Lie algebras isn't offered either, then
 
9:05 AM
Ah I see, just searched for ANT I with the same method
that's cool :) thanks for the tip
Affine algebraic groups
looks like Liealgebren is not offered either
 
@ÍgjøgnumMeg there's a lecture on modular forms
 
I took that, it was quite cool
 
What kind of background does one require? I've had a course on complex analysis but it was pretty much junk
 
an intro course in complex analysis should be sufficient
if you know e.g. the residue theorem
and what holomorphic means
 
9:12 AM
Holomorphic is just ... complex differentiable right?
 
it wasn't very rigorous as a course
 
you just need a few key results
 
unfortunately the cohorts that come here aren't interested in proofs so the lecturers quite often just glide over proofs and go straight to applications and "exam questions"
i did also plan on taking the undergrad funktionentheorie course
 
@ÍgjøgnumMeg Kasten, the lecturer of the modular forms course has a good funktionentheorie script on his website
 
9:14 AM
I have ~2 months out of work before I start the master, maybe I'll brush up on some stuff
 
I'd recommend taking the modular forms course now, as that's offered irregularly
2 months is plenty for reading up on complex analysis
 
Okay cool :)
Thanks for the info, I gotta go and pretend I car eabout my job again
 
9:36 AM
I have a function p in $L^{\infty}(0,+\infty)$ that is measurable and bounded all most everywhere, I want to obtain that there exist a,b >0 such that $a\leq p(x)\leq b$ what can be a and b?
 
essential sup and essential inf of $p$?
The inequality $a\leq p(x)\leq b$ can hold only almost everywhere of course
 
but essential inf can be negative
@AlessandroCodenotti
 
Then it can't be done, just look at the function which is $-1$ on $(0,10)$ and $1$ otherwise, which is in $L^\infty(0,\infty)$
 
even if $p$ is positive, the essential inf might be $0$. Consider the function which is $1$ on $(0,1)$ and $1/x$ on $[1,\infty)$
 
Hi everyone
 
9:50 AM
Hi @Balarka
 
Hi @Balarka
 
what van be a good condition for p?
 
@ÍgjøgnumMeg still working on what ironic means but do recommend trolling the church of scientology in a literal sense it's one of my favourite drinking games
 
10:18 AM
Live and let live I guess
 
10:30 AM
Rant: Say I have a convex polyhedron, and I barycentrically triangulate it. Let $\Delta$ be the "fundamental triangle", reflections along whose edges generate the full thing. The three vertices each will have different "link structures", aka one there will have $p$ triangles meeting at, another there will have $q$ triangles meeting at and the remaining one will have $r$ triangles meeting at. If you "bloat this up" to a geodesic map on $S^2$, you'll have a tessellation with $(p, q, r)$ triangles.
 
it's finally over @BalarkaSen
359 lines
 
The symmetry group of my guy will be the triangle group $\langle a, b, c | a^2, b^2, c^2, (ab)^p, (bc)^q, (ac)^r\rangle$, right?
Of course, there is a severe restriction on $p, q, r$ coming from Gauss-Bonnet. Namely, $1/p + 1/q + 1/r > 1$.
I think $[p, q, r]$ is what they call a Schlafli symbol
A cube has Schlafli symbol $[2, 3, 4]$, so the full octahedral symmetry group is $\langle a, b, c | a^2, b^2, c^2, (ab)^2, (bc)^3, (ac)^4 \rangle$
This is isomorphic to $S_4 \times \Bbb Z_2$ (because the rotation group is $S_4$ and you're taking the preimage under the double cover $O(3) \to SO(3)$ which is isomorphic to the trivial cover $SO(3) \times \Bbb Z_2$ group theoretically, as principal $\Bbb Z_2$-bundles)
Er, I meant $2p, 2q, 2r$ is the link structures, the number of triangles meeting at the various vertices. The spherical angles at those vertices is $2\pi/2p, 2\pi/2q, 2\pi/2r$ i.e., $\pi/p, \pi/q, \pi/r$ thereof
It seems by convention you remove the $\pi/2$ angles when considering the Schlafli symbol and write it the order of barycenter of 2-simplex, then barycenter of 1-symplex, then barycenter of 0-simplex. So the correct notation for Schlafli symbol of the cube would be $[4, 3]$
Fun
 
11:11 AM
Ah, here's another way to think about this. The reflections can be realized in $\Bbb R^3$ by a collection of three 2-planes which pass through those edges, instead of passing to the spherical map. The angles between the (unit vector orthogonal to the) planes are $\pi - \pi/p, \pi - \pi/q, \pi - \pi/r$. Assign three vertices corresponding to these three planes, and join them by a labelled edge with label $m$ if they make an angle of $\pi - \pi/m$ with $m > 2$.
Usually you don't label the edge if it corresponds to $m = 3$ because it's quite common. This is the Coxeter-Dynkin diagram of the polyhedron
This generalizes to regular polytopes, since their symmetry groups are generated by reflections as well (this is a theorem of Coxeter, but it's not hard to believe)
 
11:25 AM
So if we have a regular $n$-tope, it's fundamental chamber is an $n$-simplex and we have a corresponding hyperplane arrangement of $n$ planes in $\Bbb R^n$. If $v_1, \cdots, v_n$ are the unit normal vectors, then the matrix $M = ((v_i \cdot v_j))_{1 \leq i, j \leq n}$ encoding the angles between these hyperplanes is a positive definite symmetric matrix. This is the necessary and sufficient condition for the Dynkin diagram to be realizable by a polytope
 
11:45 AM
Hello gents, I have tried to ask a question twice which wasn't clear enough, so I thought I could pop it in here and ask for a formula ?
I am a programmer (an average one at that) and not a mathematician so may be that's why I am facing a challenge in explaining my problem
but I will give it a go here
Say, I have a range of natural numbers from 1 to 5. If a user enters say the number 2 then I need to print "0 + 1 + 2 + 6 + 12 = 21"
the way this is calculated is, by multiplying the 2 numbers before the number I am processing (which will be 1 followed by 2, followed by 3 until 5)
so keep the processing bit aside.
is there a formula which I can use to calculate the sum of the product of a group of n numbers ?
so if the range of numbers is 1 to 10 and the user inputs number 3, then I need to be able to find the sum of the product of a group of 3 numbers
ex : (9∗8∗7)+(8∗7∗6)+(7∗6∗5)+......=Z
 
12:20 PM
I get how they can bust people for insider trading but how to they prevent people from insider pre-trading
?
for example project Jedi, all of the quality flock control programs that could be developed from the research of mk ultra, seems like an unfair advantage
sorry wrong window
 
 
1 hour later…
1:29 PM
My favorite new fact for tonight is: The Jewish Kabbalah acknowledges the existence of infinity as being the only entity to have truly existed for eternity, which sets it apart from all other monotheistic belief systems in that if no entity existed when God was created, it stands to reason that we should be in an infinite regress of God production, thereby nullifying God as the supreme entity, but in the Kabbalah, they acknowledge the existence of "Ein Sof" which is unusual
because the Old Testament from the Christian Bible was from Jewish scripture, and it clearly states at the beginning of Genesis that there was nothing prior to the point of God's creation
 
"fact"
 
 
1 hour later…
2:37 PM
@skillpatrol - of course. If I say, in multiple places, "I believe X", you can then say "It's a fact that Joe believes X." The underlying belief doesn't need to be proven true or false.
 
@Adam Nothing is bigger than infinity:
-nothing...-infinity...zero...infinity...nothing
1,2,3...infinity...nothing
 
@Balarka are you here?
 
@MatsGranvik - reminds me of - Pretzels are better than nothing. Ice Cream is better than Pretzels. Nothing is better than Ice Cream.
 
@AlessandroCodenotti what do you need
 
Apologies to the lactose intolerant.
 
2:49 PM
Someone who understands connections on vector bundles better than me
 
Let $K$ be some compact Hausdorff space. Recall that if $C(K)$ is finite dimensional, then $K$ is finite. The standard way of proving this is by contrapositive and an application of Urysohn's lemma. I was wondering, is there a way of proving this based on the idea of showing that $K$ is a discrete? I.e., first show that $C(K)$ being finite dimensional implies $K$ is a discrete space?
 
@AlessandroCodenotti I probably count
@JoeTaxpayer I had some amazing ice cream yesterday
 
@RyanUnger - proves my point...
 
Definitely since I'm trying to make sense of the definition! So I have a smooth bundle $E\to M$ and a connection on it is a linear map $\nabla\colon\Gamma^\infty(E)\to\Gamma(E\otimes T^\ast M)$ satisfying $\nabla(\psi f)=(\nabla\psi)f+\psi\otimes\mathrm{d}f$. I don't have an intuitive idea of $E\otimes T^\ast M$, but this is fine
 
if you want to think like an analyst, think of vector bundles in terms of their sections
 
2:53 PM
So what's a section of $E\otimes T^\ast M$?
 
the sections of $E\otimes T^*M$ are just tensor products of sections of $E$ and sections of $T^*M$
 
Ok but...
 
(well, linear combinations of those)
 
So they're like pairs of a section of $E$ (what that is intuitively depends on $E$ of course) and a $1$-form basically
 
yeah...you can think about this in coordinates too
 
2:56 PM
Now suppose $\mathfrak X\in\Gamma(TM)$ is a vector field, I want to define this covariant derivative $\nabla_{\mathfrak X}\colon\Gamma(E)\to\Gamma(E)$
 
suppose we write sections of $E$ as $s=s_\mu e^\mu$, where $e^\mu$ is an orthonormal frame
 
@AlessandroCodenotti I don't know if this is helpful or not but while studying somethings about connections of vector bundles, I thought of $E$ as $TM$ and then like tried to understand it using this basic example.
 
I don't know what a frame is...
 
then sections of $E\otimes T^*M$ are $f_{\mu,i}e^\mu\otimes dx^i$
@AlessandroCodenotti oh no
Ted is about to kill you
 
2:57 PM
@AlessandroCodenotti ok...
well you need to know what an orthonormal frame is
 
Apparently for $\sigma\in\Gamma(E)$ we want to define $\nabla_{\mathfrak X}\sigma=(\nabla\sigma)\mathfrak X$
 
it's just a set of sections that at each point gives an orthonormal basis of the fiber
 
@RyanUnger Oh ok, I didn't know this is called a frame but I've met those before
 
00:00 - 15:0015:00 - 23:00

« first day (3250 days earlier)      last day (1761 days later) »