« first day (2219 days earlier)      last day (2809 days later) » 
00:00 - 20:0020:00 - 00:00

12:06 AM
@Semiclassical evening
 
@0celo7 No, but it seems believable.
 
@BalarkaSen Unsurprisingly, the proof is via partition of unity.
 
For compact manifolds it is of course obvious so who cares :P
 
Because it's crucial to the Nomizu-Ozeki theorem!
(which is nontrivial only for noncompact manifolds)
 
You and your noncompact manifolds :) But yes, I can see a partition of unit proof. Good point.
 
12:11 AM
Why is it obvious for compact ones?
 
Image of $g$ is compact. Since it's positive, there is an $\epsilon$ such that $g$ is always $\epsilon$-away from $0$.
Take $f = g - \epsilon/2$.
 
I went to the bathroom and it hit me when I sat down.
Partition of unity is a cheat card
It's too powerful
 
The whole point is that if $M$ is noncompact, $g$ might get closer and closer to $0$ so that trick no longer works. That's why you do partition of unity to do the same trick but varying $\epsilon$ (by doing it on compact subsets of $M$ inside charts, say) and gluing those $\epsilon$'s togather.
This is the idea for all partition of unity proofs: do on small pieces, then glue.
 
@BalarkaSen Actually, one can prove the more powerful: Let $F:M\to \Bbb R$ be cont. and $\delta:M\to\Bbb R$ be positive and cont. Then there is a smooth function $G:M\to\Bbb R$ such that $|G-F|<\delta$.
 
@0celo7 Agreed. That's why complex manifolds are harder, I have been told.
No partition of unity there.
 
12:19 AM
Are they analytic?
Dumb question.
 
Holomorphic functions are analytic, yes.
 
THAT'S WHY I SAID DUMB
 
So, given that theorem, pick $F=\delta=\frac{1}{2}g$
Then you have a smooth $F$ such that $|F-\frac{1}{2}g|<\frac{1}{2}g$
To my surprise, this doesn't seem to be in Hirsch
 
Probably because all of that is not really hard.
 
12:22 AM
But the result was known in '61
He should at least state it
Maybe it's an exercise
I'm analyzing the Nomizu-Ozeki paper for my advisor and he doesn't like when I reference Lee
If it's in Hirsch, I'm good...
@BalarkaSen long shot, but do you know what the symbol $d(x,X)$ on the bottom of page 55 of Hirsch means?
 
@0celo7 I don't have Hirsch with me but it usually means the infimum of $d(x, y)$, $y$ varying over $X$.
 
@BalarkaSen Agreed but I was asking what $d(-,-)$ means.
 
Eh. The metric?
 
What metric?
 
Dunno. I'd have to look at Hirsch.
 
12:37 AM
it's a manifold, not a Riemannian manifold
and he never talks about Riemannian distances
it's not in the list of symbols
oh, he seems to use it in some proofs
@Huy Jesus christ go to bed
 
I should sleep too
 
I should finish this proof
I should buy my prof Lee for his birthday
So he can get some good math
 
12:54 AM
poopoo
 
eat da poopoo
 
dont do that
i think
 
@ForeverMozart Nah. Builds the immune system.
 
i am quasi-famous now
but there are still many problems in my area that I cannot solve
 
you're famous?
 
1:05 AM
not quite
its hard to get people interested in my research area
but I think it is fascinating
 
what is it
 
weird topological spaces
but they can be metric
or even in the plane
 
@ForeverMozart did you publish your paper
@Soham Hey
 
I sent it off
no word yet
 
aha
 
1:13 AM
i have no idea when I will hear back
 
@ForeverMozart lol
my topology prof said there are people like you
your dream is to make textbook exercises
 
@ForeverMozart you should do shape theory
 
no i dont care about books
 
the goal is the same: to study weird topological spaces. but a lot more tools - useful if you want to get more people interested
also good research possibilities
 
> Shape theory is a branch of topology, which provides a more global view of the topological spaces than homotopy theory.
...
that's pretty global
 
1:18 AM
what is an example
of a shape
 
It tries to generalize algebraic topology to weird topological spaces
 
this is a circle
 
@ForeverMozart I don't know much about it. You should look at Mardesic's works.
The basic example I think is the closed topologist's sine curve.
All of it's standard homotopy groups are 0 - yet it's not homotopy equivalent to the circle
 
can you apply it to non-compact spaces?
 
I think one proves that by showing the first shape group (analog of fundamental group) is nonzero or something
@ForeverMozart I'd bet.
 
1:22 AM
the hell is a shape group
 
you should be able to do this for very disconnected spaces too - that's Cech theory
homotopy theory for non-path connected spaces. good stuff.
 
where do I learn that?
 
dunno; google is your friend.
 
1:42 AM
Um.
You'd think people would lower their tone down a few notches, given who's answering the question?
 
lol that is overkill and circular
 
2:22 AM
> Your password may not contain more than 3 consecutive characters from a singles class (Upper, Lower, Number, Special). Please enter a different password.
smashes keyboard
 
that sounds absolutely awful. It seems to make the password easier for a brute force attack and harder to remember.
 
why easier for brute force?
I'm not remembering this
It's scribbled on this paper with some geometry
 
You're more likely to try to use a shorter password, also there are fewer combinations of length $n$ than just random. for example, qqqqqqqqqq is a 10 character string not included in the set of passwords required.
 
min of 14 characters
 
allowed, not required
 
2:31 AM
also they make you use upper, lower, numbers, and special
How does brute force work on websites anyway
they lock you out after 3 tries
 
I'm just saying that putting these requirements makes the list of possible passwords much smaller.
 
yeah I guess
 
It's probably a good idea to use upper, lower, numbers, and specials anyway
 
my go-to password is geometryBest
 
Well I hope you don't use it on anything important, because these chat logs are public, I'm pretty sure
 
2:34 AM
that's not actually my password
 
Hello
 
user61230
3:29 AM
DID YOU KNOW?
 
user61230
If you take any two coprime integers $a$ and $b$, their product is an integer that has a square.
 
do you mean a square root
 
 
1 hour later…
4:44 AM
@Emrakul $gcd(a,b)=\frac{a\cdot b}{lcm(a,b)}$, coprime means $1=\frac{a\cdot b}{lcm(a,b)}$ but there's nothing in terms of $c^2$ or $a^2$ or $b^2$ here?
 
@Secret every integer has a square (the square of x is x^2, which exists)
#humor
 
O i see, I mistaken as the integer obtained above is a perfect square of another integer
 
 
5 hours later…
9:21 AM
good @daniel you saved the situation
 
 
1 hour later…
10:49 AM
hello, i want to study the continuity of \begin{align*}
\varphi: & E \longrightarrow \mathbb{R}\\
& f\mapsto \varphi(f)=f(0)
\end{align*}
in $(E,||.||_{\infty})$
 
What is $E$?
 
a normed space
(E,||.||_{\infty})
 
then $f(0)$ does not make sense
 
sorry
$E=C([0,1],\R)$
Can i say that
 
@TobiasKildetoft hmm identicon falsification ?
 
10:51 AM
@Agawa001 ???
 
just see how do u look from the icon-lineup
 
@Agawa001 there is apparently some problem with gravatar (I have different icons in the lineup and next to where I write this)
 
can i say $|f(0)|\leq \sup_{[0,1]} |f(x)|=||f||_{\infty}$ then there exists $c=1$ such that $|\varphi(f)|\leq c||f||_{\infty}$ ?
@TobiasKildetoft ?
 
yes appearently , but members as balarka arent affected
(nepotism)
or he does pay a regular fee maybe
 
can someone help me ?
 
11:00 AM
@Agawa001 lol
@Tobias I fixed it by getting a snapshot of the "right" avatar and saving it as gravatar instead of the randomly generated identicon.
Maybe you'd want to try that out. Mike did the same.
 
This was put on hold :/ . Its a good question ( 2 upvotes , yet on hold ?! )
2
Q: Find $ ? = \sqrt[3] {1 + \sqrt[3] {1 + 2 \sqrt[3] {1 + 3 \sqrt[3] \cdots}}} $

mickI wonder about a closed form for $ ? = \sqrt[3] {1 + \sqrt[3] {1 + 2 \sqrt[3] {1 + 3 \sqrt[3] \cdots}}} $ I tried solving the related equation $ f(x) ^3 = 1 + (x+y) f(x+1) $ for various fixed values integer $y$ , but I failed. It appears $$ ? = \sqrt[3] {1 + \sqrt[3] 5} $$ But I am not able...

:/ meh
 
11:19 AM
@BalarkaSen Nah, either it corrects itself or I end up with a new icon. Either way is fine by me.
 
11:39 AM
@TobiasKildetoft Fair enough.
What're you upto?
 
@BalarkaSen Preparing a grant application as well as a job application
 
I see.
 
What about you?
(ohh, and my coauthor just dropped by with the print copies of our Advances paper)
 
Reading a lemma I need to prove the Poincare-Hopf theorem. Almost done.
 
11:55 AM
@BalarkaSen Suppose $f:E\to F$ (Banach spaces) is continuous along every continuous curve through $x_0\in E$. Is $f$ continuous at $x_0$?
 
Yes: if it was discontinuous at $x_0$, there is a limit of points approaching $x_0$, say $\{a_n\}$ such that $f(a_n)$ does not approach $f(x_0)$. Look at the piecewise path joining $a_n$.
 
That's the idea.
But making "piecewise path" precise takes a bit of work
 
Nah.
Your spaces are all vector spaces.
 
So?
 
So you think.
I am not going to work it out for you.
It makes sense to join $a_k$ and $a_{k+1}$ by a line is the point.
 
12:01 PM
@JasperLoy Indeed, I was pretty busy. Then I spent a part of the night on very important research.
 
Huy
@BalarkaSen: there's an even more elementary solution to the geometric problem. :P
I can't believe none of us saw it
 
hm?
 
@TobiasKildetoft related to a previous message of yours, imagine this situation: you have at a large table the most important mathematical figures (from the mathematics you're interested too) from all history, all alive, and waiting for you to join them. Would you dare to bother them with what you produced so far in mathematics? Would they need to see your work?
No need to answer me.
 
@user1618033 For most of them, no, they would not care or even have any idea why the sort of questions I look at could ever be called interesting.
 
@TobiasKildetoft OK
 
12:06 PM
On the other hand, Lusztig would probably belong on that list (even though he is still alive) and he has cited some of my work and will certainly care about several of the things I have done
 
Oh, Lusztig has cited you? Then you're semi-famous, @Tobias.
 
@BalarkaSen Heh, yeah. Now I am just waiting for the money to come rolling in (not sure what is keeping it).
 
@BalarkaSen I don't need you to work it out for me, I solved it.
 
@0celo7 OK, good to hear.
 
But proving that thing is continuous without the gluing lemma is not fun
 
12:09 PM
Why not use the gluing lemma?
 
And proving it is continuous at $x_0$ requires some convexity magic
@BalarkaSen because we haven't covered it in class?
 
Oh well.
Yes, maybe you'd take off a point because my curve doesn't pass through $x_0$. But one can fix that.
 
@0celo7 Then prove it by writing first a proof of the gluing lemma, then follow with a proof using the gluing lemma
 
@TobiasKildetoft Nice.
Money as in, you're getting some prize or a grant?
 
@BalarkaSen that was meant as a joke (being famous should equate to getting money, right?)
 
12:11 PM
Ah.
Yes, indeed.
 
Thinking seriously to get out of any kind of mathematical community, I've seen enough, and continue mathematics absolutely alone (away from any nonproductive opinion). It would be enough to publish once in a while some papers, books (if the case).
 
@TobiasKildetoft No, I just said "due to the gluing lemma"
I doubt the prof will object to it
he might say he doesn't know what it is though...
I don't think it's a standard name
 
I think it doesn't have a standard name
Or rather, it's so standard it doesn't have a name
 
Gluing lemma is what Lee calls it
Munkres calls it the pasting lemma?
 
I thought Munkres also calls it the gluing lemma. Maybe I misremembered, you're probably right.
 
12:16 PM
It's my topology prof who calls it the pasting lemma
 
Yes, I have seen both being used.
 
gluing lemma is not in the index of Munkres
yup, pasting lemma
Someone sent me an email with the wrong name attached, but it's definitely directed at me
hmm
 
12:35 PM
@user1618033 have you ever planned to wake up early in daily basis to practise some yoga in middle of the forest, recall some juvenile moments, process some mathematical calcs with perception alone, try to forget that you are in forest or even in romania or earth itself, just focusing on the abstract, then see what wonderful results would come into you
 
@Agawa001 never tried practising yoga. Do you practise yoga?
 
@user1618033 i learn with internet lol
bad student i am
 
@Agawa001 hehe, well, it's a beginning.
 
@user1618033 it's wonderfuuuul, once you get to that stage of 'forgetting the surrounding' you can consider that you made halfway of the session
 
(It might help, as some claim)
 
12:39 PM
it is like a "just-in-mind picnic"
 
@Agawa001 But this takes time, and I need time to invest in math, especially these days when I have a lot of stuff to do.
 
@user1618033 well i do this because sometimes my mind refuses to turn on
 
@Agawa001 And you feel it works?
 
@user1618033 at some point yes
 
I see. Interesting.
Despite my belief, I actually believe in almost nothing that cannot be proved in a rational manner. I need to convince myself that Yoga works, and then I don't believe in those energy acumulation points called in Yoga.
 
12:42 PM
@user1618033 this is better than self-shocking one's brain
 
@Agawa001 :-)))
 
@user1618033 its good for constant thinking, and uninterrupted concentration
 
@Agawa001 I reached this stage by self-education, then being able of constant thinking and uninterputed concetration for many hours in a row
But if there is room for more, it's great.
 
also for that phenomenon when ur thinking switch to something else as another topic somehow connected to the main one unwillingly
@user1618033 gooood, that is an achievement
 
It seems I have a problem with concentration when typing in English. :-)
 
12:49 PM
@user1618033 well it is different for me cuz of the irresistible noise i have here
 
@Agawa001 you live in the city?
 
@user1618033 no! but there is unwanted noise
 
@Agawa001 I see.
 
hallo
Why does the following equality hold where $W_k=\text{diag}(w^{(k)}_i=\lvert x^{(k)}_i+\sigma\rvert^{p-2})$?
$$\sum_iw^{(k)}_ix^2_i=x^TW_kx$$
 
1:45 PM
I'm honest, I want to revolutionize my mathematical area, to make the life of people studing this area much much better than ever before. I have results, of course, and there is still an incredible amount of work to do.
 
135 messages moved to Trash
Time to cut it out, guys.
This discussion was completely unconstructive, and it's bad for the general atmosphere of the chat room.
 
i ll get lost to pepare some coffee @chriss : my beforelast note , if you want to take depart, do it with ur own dignity, by
 
8 messages moved to Trash
 
@Agawa001 OK
@Agawa001 I'm at risk of becoming pretty unproductive today. Back to my work to finish some proofs.
BBL
 
2:02 PM
Is there any one interested in Fractional calculus here?
 
I've always thought that was a party trick
 
I do not know what that is, Pentapolis. My knowledge goes as far as Integral calculus' basics.
 
At the beginning of this week I sent by email to some students I train the following integral
$$\int_0^1 \left(\frac{\log(1+x)}{1+x}-\frac{\log(1-x)}{1+x}\right) \arctan(x) \textrm{d}x$$
They have a week to come up with a very simple solution to it that use no complex numbers.
So far I received mail that it seems impossible, of course it is not, but it might be just a bIt more difficult.
Please let me know, if the case, how you teach in a class your student to very easily calculate such an integral.
Thanks.
I'm around working on my research stuff.
BBL
i change my mind.
Please calculate it without pen and paper. How?
BBL
 
2:26 PM
Another drama?
 
@BalarkaSen sorry if my integral is a drama to you. Just skip it as usual.
 
They aren't.
But apparently I missed a big fight.
 
@BalarkaSen This is only a place of peace and love. Don't think negatively anymore. ;)
BBL
 
::steps over corpse::
whoa...
 
I think it would be positively nice to also give you the closed-form
$$\int_0^1 \left(\frac{\log(1+x)}{1+x}-\frac{\log(1-x)}{1+x}\right) \arctan(x) \textrm{d}x=\frac{\pi^3}{192}+\frac{\log(2)}{2}G$$
So lovely, isn't it!
@Agawa001 give it a try when you're back :-)
OK, back to my research, I feel I spend too much time on this chat. BBL
 
2:39 PM
I forget what $G$ is
ah, catalan's constant
 
never heard of it
 
I've heard of it, but only that. reference: en.wikipedia.org/wiki/Catalan%27s_constant
 
looks to be both not general enough and circular at the same time
i.e. it applies only to numbers which are products of prime powers
@user1618033 i think a good hint for that would be which integral representation of Catalan's constant they should be looking for
 
tries to solve with u sustitution, then by parts. Gets stuck. Finds out this involves infinite series. Flips table.

Please restrict your problem professor. A warning of sorts @user1618033
 
2:47 PM
e.g. $G=\int_0^1 \arctan{t}\,\frac{dt}{t}$
 
3:07 PM
@user1618033 Playing around with Mathematica, I find that the substitution $x=\frac{e^u-1}{e^u+1}$ gives the equivalent form $$\frac{1}{2} \int_0^\infty \frac{u\,\text{gd }u}{e^u+1}\, du$$ where $\text{gd }u$ is the so-called Gudermannian function.
which is cute if likely useless
 
3:54 PM
@user1618033 i didnt recognise u fom ur new look, ur identicon is misbehaving buy a new one
 
there's been a bunch of that going around
 
yes appearently , it is better to adopt a fixed picture
ok what we 'v got here
i could solve this without that annoying arctan
log(1+x)/(x+1) , variable substitution e=log(x+1)
for log(1-x)/(1+x) , same
 
@MithleshUpadhyay Thanks for providing the link. Yes, that accepted answer with currently score at -4 should be deleted. I've put my vote (to delete) in the mix. But, as long as it remains the accepted answer, we'll need 27 more votes to delete
 
4:13 PM
d = x_B y_C - x_C y_B

Guys, please, can i get an explanation about the scalar d in the second answer of this? It is not explained in the answer. (Maybe someone can add a edit).
4
A: Determining if an arbitrary point lies inside a triangle defined by three points?

ja72Method To test any ppoint $P=(x,y)$, first move the origin at one vertex, like $A$ such that $$ B \rightarrow B - A $$ $$ C \rightarrow C - A $$ $$ P \rightarrow P - A $$ Then I calculate the scalar $ d = x_B y_C - x_C y_B $ and the three Barycentric weights $$ \begin{matrix} w_A = \frac{ ...

 
4:26 PM
Back.
@Semiclassical OK
@Agawa001 To me it seems all is unchanged.
 
4:39 PM
I wonder if robjohn ever cared to calculate this integral $$\int_0^1 \left(\frac{\log(1+x)}{1+x}-\frac{\log(1-x)}{1+x}\right) \arctan(x) \textrm{d}x=\frac{\pi^3}{192}+\frac{\log(2)}{2}G$$
 
4:59 PM
Maybe I should say once in a while what I do in mathematics, that is making artwork, turning my mathematics into artwork (to avoid possible misunderstandings).
The integral above shoud be included in some textbooks at least (together with a shining proof).
 
user227867
5:20 PM
@user1618033 Today I went out for a haircut and walk, lol.
 
@JasperLoy You made yourself beautiful? :D
@JasperLoy Walking is GOOD!
 
user227867
@user1618033 Yes, even though I am already very beautiful, lol.
 
@JasperLoy hehe, sure. But to add some more I meant.
 
user227867
@user1618033 Was it a Monica who came yesterday to your house?
 
@JasperLoy lol, not really ::D
@JasperLoy How is it going? Rest of the things. Some date then? (after haircut and walk)
 
user227867
5:26 PM
@user1618033 So so. I hope to sort out more thoughts in September. No dates. =) There is no Laura here.
 
@JasperLoy Too bad then. :-)
 
user227867
@user1618033 When the time is right, Laura will appear, and so will Monica, lol.
 
@JasperLoy I'm not concerned about such things these days (anyway I find it very easy to solve :-)), but about finishing more important objectives from my research.
 
user227867
@user1618033 I am trying to ignore that user but her stupid comments keep appearing everywhere on the site, hehe.
 
@JasperLoy I also received a lot of strange comments today, and I think it comes from an old users that uses a different account. But I won't talk more about it. I have some ideas about that who.
 
user227867
5:32 PM
@MithleshUpadhyay It seems that the answerer cannot delete the answer if it is accepted.
 
user227867
@user1618033 You mean ideas about your strange comments or my user?
 
@JasperLoy hehe, my comments might be sometimes strange, it depends on what one understands, no matter what I say. ;)
 
user227867
@user1618033 Sorry, what I mean is you think you have ideas about who I am referring to?
 
@JasperLoy No. I was only referring to comments that were addressed to me.
 
user227867
@user1618033 I see. Anyway, no more 'negative energy'! I am flagging that user's stupid comments. Hope she gets suspended from the site forever...
 
user227867
5:38 PM
Everywhere I go, I see her attacking other users using sarcasm.
 
@JasperLoy Forever? It means a lot. :-)
 
user227867
@user1618033 Well, this user does not contribute to the site at all, even though she has a huge rep. The points mean nothing.
 
@JasperLoy I don't contribute to the site either these days. :-) I come in chat and fight with some users. :D
 
user227867
@user1618033 I don't want to mention this user, but she keeps abusing me and other people through comments and emails.
 
@JasperLoy There is so much negativity in this chat sometimes, it almost seems unreal.
 
user227867
5:42 PM
This nonsense has not stopped for years and years and years, despite people being nice to her again and again and again.
 
@JasperLoy Well, don't pay much attention to it. Focuse yourself on positive stuff. ;)
 
user227867
@user1618033 Yes, let's talk about other things then. =)
 
@JasperLoy OK. I'm leaving now for a jogging session, but we can talk later.
@JasperLoy Very long time hard work said a word to me lastly. I need to make many changes, and follow rules like in a army.
@JasperLoy Dressed and ready to go.
Talk later.
BBL
 
user61230
6:17 PM
DID YOU KNOW?
 
user61230
For all numbers $a$ and $b$ coprime with the exception of a shared factor of 7, the Collatz sequence contains only integers.
 
what
 
dunno what collatz sequence of a pair of integers mean.
 
google+ has changed its style
 
user61230
6:21 PM
I have started to confuse the math chat. Soon, they shall be under my control.
 
user61230
Uh. Wrong room.
 
low-quality troll tbh
 
user61230
Not really trying to troll, honestly. If it's seen that way, then I'll stop.
 
Nah was joking.
I think the crackpot mathematics mathgen comes up with is better.
 
user61230
I'm glad my math is better than an algorithm designed to generate crap ;)
 
6:29 PM
Hah.
 
Folks, I am having a crisis. Is there a way to deduce from $s^2=e$, $r^n=e$, $srsr=e$, that $sr^j\neq e$?? I am running in circles here on my paper
For instance, I come up with $sr^j=e \Rightarrow r^j=s$, which I hope isn't true...but it doesn't seem to be leading to any contradictions.
I can also determine that $r^{2j}=e$, which forces $n|2j$, but I still see no problem.
 
6:44 PM
WAIT
I think I got it
go on with your lives per usual
 
7:05 PM
Let me change my username
@Agawa001 I changed my username
Let's come back to the integral above
$$\int_0^1 \left(\frac{\log(1+x)}{1+x}-\frac{\log(1-x)}{1+x}\right) \arctan(x) \textrm{d}x=\frac{\pi^3}{192}+\frac{\log(2)}{2}G$$
How I do math art? Well, it's easy and complicated at the same time. The easy part is to explain the steps, that is working extremely hard for years (mainly), and the hard art, of course, is to put that into practice. What's next? After that hard work I meantioned we do the math art. How? We look at this integral, we notice something cool that comes from those hellish research days and we are done.
 
@Idomathart nice, a beginning of new cycle ?
 
@Agawa001 After the discussion with Juan I considered that it better to simplify my life in this chat, and to better explain what I do, that is mathematical art (all this by using a proper username).
A case: the students come to you and beg you, implore you to teach them how to calculate the integral above in a very nice way. What do you tell them? Telling it's not an important question?
 
@Idomathart i doubt you are that vulnerable to online trolling
 
@Agawa001 There is a different thing here. Having a misconception on my mathematics is a really bad thing. One miss so much beauty that cannot be compared to anything (I admit I might have exaggerated a bit).
 
@Idomathart i dont think he was here for seeing you with a different suit
but i like this new name keep it
 
7:19 PM
@Agawa001 Yeah, it represents me fully.
Trying now to finish some proof.
BBL
 
00:00 - 20:0020:00 - 00:00

« first day (2219 days earlier)      last day (2809 days later) »