« first day (2841 days earlier)      last day (2475 days later) » 

00:05
Where can I find the result that states that the minimal surfaces immersed in R^3 are of revolution?

I was searching the internet but I do not think so.
Morning everyone
Hi @Faust
Has your noncommutative geometry project started?
How you been @MatheinBoulomenos
Pretty good, thanks
technically i tried to die during april so im a bit behind on things
going to start end of june
or earlier depending how bored i am. not much to do when ur sick
00:08
Do you have professional help? It's no shame to seek for help when you need it
my doctor wants me to do stuff but nothing too stressful
so im doing a small math class i already know and a language class till june 28th or something
then supposed to start on the project if im feeling well enough
yeah my doctor is also worried when he hears about my workload
when i say tried to die i meant my body tried to kill me not some suicide or something
oh
it sounded like suicide
they still dont know whats wrong with me
i realized that afterwards
they got me on some new medication and im feel a bit better for the last 10 days
but still waiting to see a specialist
00:11
glad to hear you're a bit better
Mr @Faust!! hugs
doctor doesnt want me laying around the house all summer
@TedShifrin Im still alive!
I'm glad to hear it.
lethargy isn't healthy
hi @Ted
00:12
Hi @Ted
Having things to do makes one heal better than having nothing to do.
Hi Leaky, Mathein.
i dont think i can handle just doing a langauge class and the math class is a joke im not going to it anymore
i dont even know why im allowed to take the course
What math class is a joke?
its called finite mathimatics 251 or something at my university
Oh yeah, you're too advanced for that.
00:13
its a touch of combinatorics programing finite math probality and linear algerbra
but i have done all those classes
That's like a basic discrete math class for computer science majors or something.
Yeah, you shouldn't be in that.
yeah
But your health is more important.
there not alot of options
Having something I enjoy and I'm even partially good at helped me a lot to recover when I had my share of problems
00:14
im supposed to be doing hard math as of the 7th
but doctor wont let me
Well, listen to the doctor.
yea
hoping ill be allowed to for july
If you want to do a little unofficial geometry or something and talk with me about it, we could do that gently until you're better.
i can still learn things he just doesnt want the pressure
I get it.
00:16
module theory book would be nice
Well, if we're gonna talk, I have to have whatever you're thinking about :)
lol if your like my profs you probally have alot fo books
can u reccomend one?
No, I got rid of almost everything when I retired and moved.
lol ok
you got anything on hyperbolic geometry?
Seriously, probably got rid of $10,000 worth of math books.
00:18
yeah i woulda loved to buy some off you
I gave 'em all away.
I have a section on hyperbolic geometry in my diff geo notes.
@Faust what do already know? One of the first major theorems one encounters for modules is the structure theorem for finitely generated modules over a PID (this has e.g. striking applications to linear algebra and finite abelian groups), but a specialized book on modules probably won't prove that since the most accessible proof is basically a complicated LA algorithm and it's a bit tedious to describe it and show that it terminates
at least the "module theory books" I know don't prove that
Did we convince you to get Artin's algebra book, @Faust? There's all sorts of cool stuff in there.
you could always start with the section on modules in a general algebra book you like, e.g. Artin, Aluffi, Jacobson ...
I have done 3 undergraduate classes on abstract algerbra and 2 on linear algerbra so i know a decent amount about groups and rings and linear algerbra
00:21
that will be helpful. Do you know the structure theorem I mentioned?
maybe not in those words
Artin's book might be a good thing, Faust. Have you looked at it?
we did cover some of that but we avoided using upper level abstarct algerbra in our linaear algebra classes
lemme see if my libary has it
You used to like geometry stuff, so that's why I mentioned it.
yeah its avalible i think i have seen it before at a glance
00:25
What's cool about Artin is that he shows how algebra fits in mathematics as a whole. Lots of cool stuff in there.
Plus he integrates algebra and linear algebra.
ill pick it up tomorrow
i love geometry and algerbra
can't spell worth a dam though
so im taking a japanese language class i say down and learned hiragana, katakana and 80 kanji in a couple weeks before the class started. turns out the goal of the course is to learn those 2 and 86 kanji :\
OK, and we could work on a bit of my diff geo notes if you want.
the textbook is misleading as it starts asssuming you know katakana and hiragana so i just assumed i had to know it
I like how the first example of a monoid other than $\Bbb N$ that Lang gives is the homeomorphism classes of compact connected surfaces under connected sums (he gives generators and relations). When I read that as a freshman, it was a bit over my head
turns out the book stats on page 171 not page 1
00:29
Well, if nothing else, Lang was a mathematician who prided himself on knowing everything. I like that.
@TedShifrin that would be great at my own pace relaxing summer now that i know analysis and linear algebra it should be alot easier
keep your brain challenged without too much stress, @Faust
just so you're having fun struggling with a bit of challenge without grades :P
i still gotta talk to the prof of that math class and ask if i can not show up or do something else during the class
I'm shocked they let you sign up for that, @Faust.
it doesnt technically conflit in any major way with anything i have taken
the problem is th enumber of classes i have taken
00:32
We never let people take lower level courses when they'd had more advanced. It would count as an elective, but not for the major.
Bayes’ formula, random variables
and geometric approach
to linear programming,
I mean, that's ok stuff :)
im just taking it for credits
i have all my math classes
need mroe credits to graduate
@Faust if you really want a dedicated book on module theory, note that the theory of modules depends heavily on your base ring, so most books will also treat some amount of ring theory (it really doesn't make sense otherwise). Books that do more on modules than on rings are "Rings and categories of modules" by Anderson-Fuller and "Lectures and Modules and Rings" by Lam. The second book is the sequel to "A first course in noncommutative rings", though, which focuses more on rings
well, that should count fine as an elective, Faust
Check out the answer using probability on this. I thought it was cool.
00:33
that list was all that i havent learned for the class
I also have to take a joke course which is a waste of time
@TedShifrin nice answer
Hey everyone!
Morning
Hi Demonark.
00:36
the requirements for my major are actively hindering me from taking classes relevant for the research I'll presumably do
makes sense
Mathein: A college degree is not a Ph.D.
abstract algebra and complex analysis are electives here, but probablity and numerical analysis are required. It doesn't make a whole lot of sense
half of the units i need can be taken as electives for my degree so im just taking all math classes
Well-roundedness. I approve.
00:39
How is it well-rounded if you can get a math degree without abstract algebra or complex analysis?
complex analysis is pretty heavy
I would include complex analysis (but not if it requires a year of real analysis).
i dont think everyone needs complex analysis
I wouldn't necessarily include algebra.
its a 4th year elective class for us
algebra is required though
00:41
We required algebra for our general math degree at UGA, but not for the applied.
its so useful though
AA is ftw
Useful for pure math or for cryptography, but there's a ton of applied math that has nothing whatsoever to do with algebra.
And a ton of pure math, too :P
not number theory
what pure math
i cant think of it
Analysis stuff ...
the numerical analysis has barely any proofs, at least for the students. you just learn a dozen of different algorithms for solving linear equations and evaluating polynomials in different bases
There's ton of math that has nothing to do whatsoever with numerical analysis
00:43
and i think both of those topic's are bleh :p
They should do ODE and PDE in that numerical analysis course, not just linear algebra.
PDE's scare me
Well, I'm not going to fight with you guys. Have your own opinions, even if they're wrong.
you like numerical analysis?
If I look at the courses offered, there are a ton of courses that are related to abstract algebra
and the only ones related to numerical analysis are advanced courses on numerical analysis and maybe optimization
00:46
that can vary alot by university though
Hi
whats offerd i mean
@nbro morning
I'd be happy if they would allow me to take the measure-theory based probability course instead of the elementary one, but no, I have to waste my time
I have the following problem
The definition of AIC is
thats not my department
00:48
Bear with me, it's just the mathematics where I am stuck with
I can't think now
It should be very easy
I will show you what I did
@TedShifrin homological algebra has been applied to functional analysis
I did what's in this last picture
What I am not able, right now, to do is to go from step 17 to 18
@MatheinBoulomenos Didn't I tell you that
It's not clear how that n + 4 pops up from the summation of the realizations squared all multiplied by $\frac{1}{\sigma^2}$
I am just super tired now to think
So, a little help, in this situation, would be appreciated
00:53
Im super fatigued im going have to sleep for a bit seems the first day of school took alot out of me
gnight guys
good night @Faust
@MatheinBoulomenos I would gladly use algebra if it simplified what I'm doing rn
I have no idea how this becomes n + 4
Provided my assumptions and calculates are correct
not saying everyone should do algebra all the time, but an intro course would make sense for a math major. Even if it's just a bit of language you use, knowing that bit of language and trivial results can help a lot. I remember some stuff in complex analysis and topology where people had trouble because they didn't know what a group action is
the point is I'm desperate enough to hope some crazy German guy has done all of this algebraically...not likley though
00:59
I also think everyone should learn a bit of topology because it comes up so often
Is it even possible to do interesting things without it
@nbro Well, I'm not 100% on how $x_{j}$ is calculated (I'm a novice, especially compared to most people here), but I would look for instances in which $\frac{1}{\sigma ^2}$ would end up canceling out something in $x_{j}$
depends on what you consider interesting
...topology
01:00
@Rithaniel Yeah
I think that the mean and variance, in this case, can be expressed in a certain way which may simplify things
(Again, provided my calculations so far are correct)
Yes
Exactly
The mean in this case is given by
well for algebra, you probably won't see topology in a first course and depending on waht you do, not even in a second course, but it will come up at some point when you do infinite Galois theory or algebraic geometry. You can do the topological part of algebraic geometry really early, e.g. Atiyah-Macdonald introduces that in the exercises to the first chapter
and that's just a commutative algebra book, not even algebraic geometry
But we know that the mean is 0
It seems like geometric analysis is very far removed from "pure" algebra
The most algebraic thing in my thesis is the MV sequence
Actually, this is the estimate of the mean
More precisely, the maximum likelihood estimator of the mean
I don't think I ever used something from my real analysis sequence in the advanced courses I took except maybe definitions of metric spaces, Cauchy sequences etc.
01:07
The most algebraic thing in my nonlinear hyperbolic PDE course was a matrix...
actually a bilinear form
fancy
But at least for real analysis I see the relevance for pure math and when I get deeper into e.g. the Langlands program, I will need it indirectly for harmonic analysis. But for numerical analysis I have no idea how I will need that ever, not even for the programming object I'm about to do
Ok, I think I got where the $n$ comes from
Now I just need to understand where the 4 comes from
A bit strange to me but eh, such is life I guess
How many such classes are they making you take?
@Mathein
@0celo7 if we're talking about bilinear forms you're practically an algebraist at this point. You should prob quit analysis
01:15
intro probability, intro numerical analysis and for the third I have the choice of the agony
3 semester courses? That's a bit excessive. Hopefully it's not too much of a setback
Dude I just used a freaking Kelvin transform
Hmm, what are your choices?
I'm too deep in this analysis to stop now
advanced numerical analysis, statistics, linear optimization, non-linear optimization
01:16
Ah, rip
why not something like riemannian geometry
that's applied math
:thonk:
we can choose 2 from: algebra 1, algebra 2, complex analysis 1, complex analysis 2, differential geometry 1, differential geometry 2, algebraic topology 1, algebraic topology 2
Lol but yeah I dunno, maybe it's just me but somehow it feels like some of the stuff is just a bit less interesting unless you're already interested in the contexts where they're applied. I feel like applications of math to computer science, stuff like cryptography, would be my speed
Also I'd probably prefer probability to statistics I feel, but maybe that's not true. :shrug:
I'm fine with CS courses
the ones I took were more interesting than the applied math courses I took
01:24
Hmm, hopefully there aren't too many classes that you intended to take but can't because of this?
well, I could finish my bachelor degree one semester earlier if I didn't have to take those classes and could get credit for more pure classes that I took anyway
I see
so far I just delayed those classes as far as I could, but I have to take them eventually
Is there a process for being exempt?
but I won't let that prevent me from taking classes that are essential for what I want to do, e.g. algebraic geometry
no
01:28
Rip. I've heard that at some places there are ways to petition waivers, in fact the way I executed my biology requirement required getting special permission
German bureaucrats are legendary
I really wonder what is more important for a math major to know, group actions or Clenshaw–Curtis quadrature
For a second I merged your two messages in my mind
are u a psychic '
And I was really confused by the phrase: "I really wonder what's more important for a German bureaucrat to know..."
01:37
The intelligent bureaucrat uses group actions to more effectively annoy people
@MatheinBoulomenos is there a categorical maximum principle
like for harmonic functions?
Not that I know of
there's no way of defining a PDE on the class of categories or something
@0celo7 I think even algebra can't help you here, only your tears can save you with whatever madness it seems like you're dealing with :P
I have been working on this proof for hours
I am so ready to be done
Help me
but maybe there's something that starts with "let $C$ be a smooth topos"
01:40
bah, smooth
I'm working in the weighted Sobolev category
I don't even know what that is
if you want to force decay at infinity you put a weight in your norm
it lets you solve equations but you have to sacrifice a kidney to Fredholm
that actually sounds interesting. Decay at infinity is important for modular forms. How does this modification of the norm look like?
I'll email you the current draft
do I have yours
I don't think so
01:47
ok
did you choose the most german email ever for a reason
that's just my name
oh wow
so Mathein is a pseudonym
til
well, I don't use my middle names much
Mathein is Ancient Greek and means "to learn". It's the word that "mathematics" derives from which is something like "the art of learning"
I see
finishing this sentence then emailing
check section 6.2
@0celo7 interesting, but I don't really understand how the weight forces decay at infinity
02:02
@MatheinBoulomenos the weight grows for negative $\delta$ and large $p$
so for the integral to be finite, the function has to decay to counteract this
okay make sense
there's quantitative decay in the same spirit as Morrey's inequality, which you might know
I don't know that inequality
@MatheinBoulomenos It says that if $f\in W^{1,p}(\Bbb R^n)$ for $p>n$, then $f$ is bounded
(also that $f$ is Holder continuous)
but here you get that the function is $o(r^{-\epsilon})$ instead of being just bounded
you can get decay from the usual Sobolev spaces as well but you need like $n/2+1$ derivatives
and I don't think it's quantitative...
@Mathein hmm, so there was one point in time where I Googled the definition of a modular form. A weight k modular form being one where $f(\frac{az+b}{cz+d}) = (cz+d)^k f(z)$ for $\begin{pmatrix}a&b\\c&d\end{pmatrix} \in SL(2,\mathbb{Z})$ if I remember right. What does that have to do with number theory?
(Also, is there a "meaning" to such a function, if that makes sense?)
02:09
it's kind of unexpected how this is related to number theory
not sure about "meaning", modular forms feel like magic
coefficients of fourier expansion of modular form <--> L-functions and identities for arithmetic functions
That transformation rule is the most important thing, you also need holomorphic and holomorphic at infinity.
"There are five elementary arithmetical operations: addition, subtraction, multiplication, division, and… modular forms." - Eichler
Wow that seems cool. (Also the more time goes on, the more it seems like not knowing Fourier is prob gonna bite me in the ass)
@Daminark Reed and Simon Vol 2 or Hormander Vol 1
02:13
Fourier stuff is much easier when you assume that everything is holomorphic. Fourier series are just a change of variables of a Laurent series
Well, so I've been told that depending on who you are, stuff like proving L^p convergence of Fourier series and all that can be either really interesting or really boring. Given that i might be interested in algebra more, I've been told to check out Folland's "Abstract Harmonic Analysis" book
We did all the Fourier stuff we needed for basic modular forms in 2 lectures in our second complex analysis course (and the lecturer is kinda slow)
One of the first results you do for modular forms is you compute the dimensions of the spaces of modular forms of weight $k$. They are all finite-dimensional and you can also give generators. Actually, the modular forms form a graded algebra over $\Bbb C$, graded by weight (i.e. the weight of the product is the sum of the weights) and you can give generators for that algebra. The algebra of all modular forms is isomorphic to the polynomial ring over $\Bbb C$ in two variables
did you type all that just now
or was it prepared
was typing that, but I replied to the question first
the two generators corresponding to the variables are Eisenstein series $E_4$ and $E_6$. The Fourier coefficients of those are number-theoretic functions, multiples of $\sigma_3$ and $\sigma_5$, where $\sigma_k(n) = \sum_{d \mid n} d^k$
you can prove some pretty nontrivial identities of number-theoretic functions by proving equalities of modular forms
Interesting
02:24
I guess one answer to "meaning" is that, for example, weight 0 modular forms of level 1 (i.e. the definition that you have there, with k=0) correspond to functions when you mod out the upper half plane by the action of $SL_2(\mathbb{Z})$ (the resulting space, if you compactify it by including cusps, is the modular curve $X_0(1)$ - which is isomorphic to $\mathbb{P}^1$ - so it's no surprise that functions are just constant (i.e. weight 0 modular forms are just constants).

Modular forms of weight 2k corresopnd to k-fold holomorphic forms on $X_0(1)$ (so for example there are no weight 2 modu
For example, $E_4^2$ and $E_8$ are both modular forms of weigth $8$ and the space of modular forms of weight $8$ is one-dimensional, so they are multiples of each other. Since they have the same first Fourier coefficient, they are actually equal.
Now $E_4(z)=1+240\sum_{n \geq 1} \sigma_3(n)q^n$ and $E_8(z)=1+480\sum_{n \geq 1} \sigma_5(n)q^n$, where $q=e^{2\pi i z}$, so comparing coefficients in the equality $E_4^2=E_8$ gives $\sigma_7(n)=\sigma_3(n)+120 \sum_{m=1}^{n-1}\sigma_3(n-m)\sigma_3(m)$
(you can multiply these Fourier series like power series)
no idea how to prove that without modular forms, seems really difficult
but it just "falls out" of the theory of modular forms
I see, that sounds insane
Oh also I just realized that these guys are only holomorphic on the upper half plane, okay now I'm happier with your first thing @loch
In general, because the space of modular forms for a given weight is finite-dimensional and we know the dimension exactly, to prove that two modular forms are equal we only need to compare finitely many Fourier coefficients (in the previous case, just one, which was basically $1$ by definition since these Eisenstein series are normalized that way), but if we prove that two modular forms are equal, we get equality for all Fourier coefficients which are often interesting for number theory
another example is Jacobi's four square theorem (en.wikipedia.org/wiki/Jacobi%27s_four-square_theorem) which gives a formula for the number of ways that each natural number can be written as a number of four squares. It's not even obvious that this number is always greater than $0$ (that would be Lagrange's four square theorem, which just states existence)
@Daminark It's possible that I may have made some mistake in what I said though (unfortunately I don't know this off the top of my head all that well..). But you can look at section 2 in https://www.math.wisc.edu/~hast/Notes/Math847-modular-forms-notes.pdf , or more generally "modular forms" by Miyake ( in particular chapter 2). The modular interpretation can be found in Diamond and Shurman's book.

But this story doesn't quite explain (at least I don't see it) all the other miraculous stuff that modular forms somehow prove, e.g. ones that @MatheinBoulomenos is talking about.
the proof is similar, you show that two modular forms are equal and compare Fourier coefficients
there are some other connections. You have some special operators that act on those spaces of modular forms of a given weight, called Hecke operators. It turns out that the eigenvalues of those operators are actually algebraic integers. And if you have an eigenform for a Hecke operator that satisfies some other conditions (it vanishes at infinity, which is equivalent to having zero $0$th Fourier coefficient), then this can be used to define a Galois representation
the last fact is central for Wiles proof of FLT
02:46
Well that was some witchcraft
since Galois representations can also be related to elliptic curves. Wiles proof doesn't go from elliptic curves to modular forms directly, but has Galois representations as an intermediate step
here I am, just a humble farmer tending to my harmonic functions
meanwhile Mathein does FLT in his sleep
Hecke also found some connections between modular forms and Dirichlet series. If $f$ is a modular form with Fourier series $f(z)=a_0+\sum_{n \geq 1} a_n e^{2\pi i n z}$, then we can look at the Dirichlet series $L(s)=\sum_{n \geq 1} \frac{a_n}{n^s}$. Hecke proves some facts about this Dirichlet series just using the fact that the coefficients $a_n$ come from the Fourier series of a modular forms. Various properties of the modular form (e.g. weight, $a_0$) also appear in the properties
Such Dirichlet series are generalizations of the Riemann $\zeta$ function and Hecke also proves some similar properties to the Riemann $\zeta$ function for those where the coefficients come from a modular form, e.g. analytic continuation, formula for the only possible residue, functional equations.
03:02
Huh, in complex analysis recently we've been talking a bit about Dirichlet series :P
there can be a lot of information in a single residue of a Dirichlet series
that's some crazy magic
basically you can associate something like the Riemann $\zeta$ function to each number field and this has analytic continuation with a simple pole at $1$ and there's a formula for the residue at that pole that involves basically any invariant we can associate to a number field, even some which are really hard to compute
That's insane
So basically, modulo a lot of algebraic and analytic theory, the fact that $\Bbb Z$ is a PID is a consequence of the fact that $\lim_{s \to 1} (s-1)\zeta(s) =1$, where $\zeta$ is the Riemann zeta function
See I now want to see a timeline where someone never learns that $\mathbb{Z}$ is a PID except as a consequence of this business
03:19
@0celo7 u humble?
in The h Bar, Jul 20 '17 at 5:16, by The Raiders of Las Vegas
behumbole
@Mathein oh so, I talked to someone in the other Galois class, the one where they had the homework I sent you about finding a primitive element and its minimal polynomial for every intermediate extension where the Galois group was $S_4$
He sent out that solution but where the Galois group was $D_4$ instead of $S_4$
having a degree $8$ polynomial sounds a lot more reasonable than a degree $24$ polynomial
Good grief ... Mathein needs sleep.
also you have 10 subgroups instead of 30
sleep is overrated
Yeah for sure, it's just like, 6 pages only for that, I wonder if someone actually did the whole problem
03:34
6 pages and they skipped the computation for the largest degree minimal polynomial
Our pset was luckily much more pleasant
Wait wow
I don't think anyone computed a polynomial like $x^{24} - 90x^{21} - 70x^{20} + 5695x^{18} + 18690x^{17} + 34895x^{16} - 225900x^{15} -
1544060x^{14} - 3867780x^{13} + 18840027x^{12} + 62876100x^{11} + 228621050x^{10}
+ 222888810x^9 + 999415025x^8 - 9907474500x^7 - 24575577355x^6 -
34467394920x^5 + 232838692457x^4 + 705674357100x^3 + 2030693398335x^2 +
2155371295770x + 1779496656001$ by hand
chokes
Yeah I highly doubt that
The guy's a mad lad
you either spend probably over 2 days doing some uninspiring calculations or you write 6 lines of magma code in under 1 minute
03:41
Hi yall
._.
Hi @KasmirKhaan
hides from Kasmir
Mathein :D
Ted :D
I am not gonna ask Q's! just hanging around for a min ._.
Kasmir gotta go see doctor
Not feeling so hot
Oh oh ... what have you done?
Bad caugh for a week now
03:43
Oh oh ... could be pneumonia
I wake up at night and cant breath properly
@Daminark in case you have do some ridiculous stuff like that, enter the following code on http://magma.maths.usyd.edu.au/calc/

P<x>:=PolynomialAlgebra(Rationals());
f:=x^4-x-1;
K:=SplittingField(f);
a:=PrimitiveElement(K);
g:=MinimalPolynomial(a);
print g;

Replace x^4-x-1 with any polynomial you want
:(((((((
yeah, get thee to a doctor
Will do !
03:44
No sickly people!
if you want two lines, you can also use:
P<x>:=PolynomialAlgebra(Rationals());
print MinimalPolynomial(PrimitiveElement(SplittingField(x^4-x-1)));
I will keep that in mind
This proof involves the numbers 3.4 and 3.6
strange
But yeah I think most people I talked to just didn't do the really bad parts of that problem
(To be fair, basically none of us know about magma)
But yeah thanks!
it's useful if you want to e.g. check calculations
the syntax is pretty easy and there are built-in functions for a lot of things you encounter in algebra
03:50
That'll be real convenient
04:49
@GFauxPas Nice! but I was thinking of some analytic procedure or through some method of contradiction by which I can prove that $f$ is unbounded
 
2 hours later…
06:25
@0celo7 "note that thanks to lemma 3.4 we can apply theorem 3.6 and conclude"
06:36
sin⁡〖(A+B).sin⁡〖(A-B)= 〖sin 〗^2 〗 〗 A-〖sin〗^2 B How to prove this?
07:13
Hello!!

A DIN A4 sheet is divided into thirds. A rectangle is the root of the tree, the other two rectangles are each divided into thirds again. Two rectangles form the branches - one to the left, one to the right - the others are again divided into thirds and so on.

How can we calculate the area of that tree?

Do we maybe do the following?
Let x be the area of the paper, then we divide this into 3 parts, so into three parts ith area x/3.
Then each of the two parts is again divided into three parts, i.e. now we have x/3 + 2*(x/3)/3.
anyone here rn
 
2 hours later…
09:05
hello
@Kawaiiii hello
09:17
@Abhinav helo
09:55
@AlessandroCodenotti hello

« first day (2841 days earlier)      last day (2475 days later) »