« first day (1476 days earlier)      last day (3535 days later) » 

1:00 PM
@anon I think it is time for you to retire from the moderator post.
 
I am not a moderator. You mean chat co-owner with robjohn?
 
It's too much for your old bones.
 
@DanielFischer: Oh, about the partial fractions thing. I think I fixed my answer. Is it okay now?
 
@anon Yeah, that.
 
I'm just being dramatic.
My bones are still young.
 
1:01 PM
@Nick well, that depends on your interval.
 
LOL, I knew you'd take the offence, @anon
 
I sincerely believe that @robjohn, @Chris'ssis and @anon are the few golden members that make M.SE the best site in the stackexchange network
 
daww
 
@Nick lol, I'm not active on main, but just a bit ... once in a while :-)))))) @robjohn is like that, not me! :-)
 
@Nick I like @blue better than @anon
 
1:05 PM
@Chris'ssis: I can't $\int$ in an interval where there's a jump. So, I have to choose an interval where the function is continuous?
 
@Nick Well, how would you integrate $\int_0^3 \lfloor x\rfloor dx$ ?
 
@Nick Can't be right. You cannot have $A = 0$, since that would give you a $Cx^2$ in the numerator. $1 = (Ax+B)(x-3) + C(x^2+1) = Ax^2 + (B-3A)x - 3B * Cx^2 + C = (A+C)x^2 + (B-3A)x + (C-3B)$ gives you $A+C = 0$, $B-3A = 0$ and $C-3B = 1$.
 
@Chris'ssis: If I were seriously asked that question, I'd ping you and have you answer it for me. But I think you can't because the function in the interval is discontinuous.
 
So $C - 9A = 1$, using $A = -C$, we get $10 C = 1$, so $A = -\frac{1}{10},\, B = -\frac{3}{10},\, C = \frac{1}{10}$.
 
@Chris'ssis Chat is a part of the site, and you have been a lot more active here than I have recently.
 
1:09 PM
@robjohn Yeah I noticed that. Did something happen (btw)?
 
@Chris'ssis 3
 
@BalarkaSen: I like @anon better than @blue because I know @anon better.
 
@Chris'ssis Life happens.
 
@robjohn Exactly.
 
$0+1+2$
 
1:10 PM
@robjohn I hope everything is fine now.
 
@Nick He's getting quite a few useless pings like that, you know.
=P
 
@Chris'ssis Life goes on, but I hope to be here more than in the past week. I am on vacation next week, so I should be here a good bit
 
@BalarkaSen did you ever meet benjalim? that guy'd paint the room orange, it was amusing.
 
@blue who?
 
I'll take that as a no. last I recall he was using the alias grothendieck, or maybe he reverted to user####. if I still used skype I could prolly talk to him again.
 
1:14 PM
@blue Why do you wanna talk to him? Why not me?
 
If I did want to talk to him, it'd be about math (which I rarely talk about with you, or the many others that have asked why I don't talk to them in chat over the years now...), but I never said I wanted to talk to him again. I suppose it'd be nice to at some point, just don't have any plans to.
 
who was that guy? an ex-co-owner?
 
Nope, just another guy.
 
nah, just another guy who'd come to chat
 
I actually have some problems with him.
 
1:16 PM
Me too
 
@robjohn @r9m is also missing. I didn't see him in the last period of time.
 
I can even list them as problem 1,2,3,...
 
@JasperLoy: I can list them (i) , (ii), (iii),..
 
@Chris'ssis I saw him over the weekend.... but I wasn't here much yesterday to miss him ;-)
 
1:18 PM
@Nick However, my greatest enemy on Math SE is someone with over 150k, lol.
 
@blue have you seen my fresh attempt at group theoretic galois theory?
 
@JasperLoy lol, I liked that :-))))))))))
 
@Chris'ssis What, you mean you know who it is?
 
@BalarkaSen nope.
 
@JasperLoy: ... that means he's my enemy too. (Get what I'm doin here?)
 
1:19 PM
@JasperLoy It shouldn't be hard to guess ... Did or amWhy? :-)
 
@Chris'ssis The latter, lol.
 
:))))))
 
@Chris'ssis: So, about integrating $\int^0_{-1} \text{floor}(x)$ .. how do I go about it. Or just link me to something if it's too long and boring to explain.
 
@blue the galois groups are mimicked by Aut_H(G) for some group G and subgroup H. the galois extensions are mimicked by characteristic subgroups. Claim : if N <= H <= G is a chain of characteristic subgroups, then there is a left-exact sequence 1 --> Aut_H(G) --> Aut_N(G) --> Aut_N(H).
 
@Nick look at the graph of the function ...
 
1:23 PM
@Chris'ssis: I just plotted it on some graph paper.
 
hey,
how to unblock?
 
@blue I won't post a proof if you're convinced.
 
i cant ask to SO
 
@user44517 What do you mean?
 
stackoverflow wont allow me to post question
 
1:24 PM
@user44517 Oh, then we can't answer you either.
 
@BalarkaSen try generalizing to more general categories, for fun. (e.g. if the category isn't concrete, how do we define the relative automorphism groups arrow-theoretically? enjoy.)
 
Sorry, we are no longer accepting questions from this account. See the Help Center to learn more
 
@Nick you need to understand what happens there - mathworld.wolfram.com/FloorFunction.html
 
damn, why u sucks?
 
@blue hmm hmm categorical automorphism group hmm
 
1:25 PM
@Chris'ssis: Its just $y + 1= 0$ constricted to $x\in [-1,0)$ . How do I integrate that?
 
@user44517 What did you do?
 
idk
 
@user44517 Nobody can help you. You may send an email to SE staff. Bye.
 
actually someone can
consider yourself banned
 
@Nick I recommend you to learn some more and then answer these questions alone.
 
1:27 PM
@user44517: Trolollololololo
 
uh, sucks you know
 
relax
kid
 
@Nick Let us ignore the troll. Don't feed him.
 
u shouldnt if you cant
 
@user44517 you gotta problem
 
1:29 PM
with what?
 
with me
 
-_-
 
there is no greater troll than skullpa troll
 
>8(
 
@BalarkaSen You mean Scallop Troll.
 
1:30 PM
god i am serious
 
@Chris'ssis: I deeply apologize for any stupidity I may have said. And after a bit of thinking $\int^{0}_{-1} \text{floor} (x)= -1$ .. Is that right?
 
Why do we have $$\lim_{n\to\infty} \frac{1}{n} \sum_{k=1}^{n} f\big( \{\frac{n}{k}\}\big) = \int_{0}^{1} f\big( \{\frac{n}{k}\}\big) dx$$ ?
 
Riemann sums.
 
Is that a Riemann sum?
 
Well, @skullpatrol was always very nice with me.
 
1:31 PM
Yes, skullpatrol is a nice boy.
 
@blue doing that needs a lot of category theory, I presume?
 
looc si lortaplluks
 
@BalarkaSen "a lot" is relative; I doubt an actual category theorist would say so. think of extensions L/K as arrows K->L, and relative automorphisms as maps f:L->L that fix the map K->L after postcomposition. then one can try the same exact sequence trick you brought up.
 
Oops can't edit above msg anymore
 
@rehband: Copy the latex, edit and repost
 
1:34 PM
@blue then what's the problem with defining a totally category-theoretic galois theory?
 
@skullpatrol: Did you get my last message?
3 mins ago, by Nick
looc si lortaplluks
 
@blue also, an interesting question is when that exact sequence is short exact.
 
@BalarkaSen It is short exact when it is short and exact, lol.
 
everyone is trolling. i think i'm going to sign outta here.
 
@BalarkaSen I am not trolling, I am just joking.
 
1:37 PM
@BalarkaSen from what I understand, nothing: while we often want more from a galois theory than just that small exact sequence, we can generalize its structure to more general categories. search for "galois categories"
 
@BalarkaSen Yeah, I also feel that. :-)
 
@BalarkaSen: Hey, My message to skull was different than what @user44517 is pooing out.
 
@Nick yes, what does it mean?
 
@skullpatrol: Read it backwards, silly.
 
1:39 PM
@skullpatrol: and I mean it. :D
 
@user44517 Do you want to be suspended from chat? If not, keep quiet.
 
lol i am joking
is that funny?
 
so, it is
never open that link
 
1:41 PM
cool pic
that seems funny
 
Click on it
 
too tired
 
sleep
 
Me too
 
good night then
 
1:42 PM
later
 
good night :D
 
good
 
@BalarkaSen: The video, I linked it for a reason, $$\int_0 ^{\pi / 6} \sec y \cdot dy = \ln(\sqrt3)(\text{What})^{64} + C $$ Quickly, find $\text{What}$
 
Why does the following equality hold: $$\int_1^\infty\frac{f(\{x\})}{x^2}dx=\int_0^1f(x)\Big(\sum_{n=1}^{\infty}\frac{‌​1}{(n+x)^2}\Big)dx$$ ?
 
split up the domain of integration into intervals [n,n+1) for naturals n
 
1:48 PM
okay, thanks
:)
 
@Nick I guessed well you're trolling there ... :-)
 
@Chris'ssis: Wait, now you're the guy those messages get sent to?
 
did something get flagged? we seem to have attracted an alien...
 
@anon: It was the first time I attempted to flag.
 
@Nick what did you flag?
 
1:52 PM
Something which was neither English nor MathJax...
 
Something. I was testing out what all the buttons do.
Wow, did I attract all the mods!
 
well, for better or worse, the troll got trolled.
 
I apologize
 
@Nick when you flag something in chat, it gets echoed across the entire SE network and interrupts a lot of people's business (this includes all high-rep users, not just mods). this math room in particular is known to be the worst offender with frivolous flags.
 
@anon: OMG O_O So sorry. so so so sorry
 
1:53 PM
@anon Really? I thought it was The Bridge...
 
@robjohn that result I got is pretty impressive, and it wasn't the first time I tried to compute that. It requires some work though.
 
@nick - no worries, disruptive behavior is actually a valid use of a chat flag though
 
@MichaelHampton heh, that's what I thought too, but another mod told me it was this room. I guess we're in competition. or maybe I misremembered that chat.
 
@blue O_O that's Grothendiek's Galois theory!
 
@Nick Yeah, no worries, the flag was fine
 
1:54 PM
@AJHenderson: I don't think we've even met. But sorry
@Michael: What are you guys supposed to do when someone flags something?
 
@Nick Figure out what's going on and then crack skulls.
 
skull?
2
 
oh boy
 
@Michael: No, you wouldn't O_O
 
you shouldn't have used that word.
 
1:56 PM
@robjohn I've been working (pretty hard) on some modified versions. That evaluates to $\zeta(2)\zeta(3)-\zeta(5)$.
 
Oh no, did I open the wrong can of worms?
 
@Nick no, I'm saying the flag was fine, yes we sometimes get superfulous flags, but that one was valid. Mass giberish posts is a problem worth flagging
though a better description would be helpful :)
@BalarkaSen you keep using that word, I don't think it means what you think it means
 
@AJHenderson In my defense, I did not expect this big of a wave to be made. I was just looking at all the fun features.
 
Dee Dee will no longer press the Red Button in this Lab.
 
1:58 PM
@AJHenderson I never said "skull".
 
yeah, I was more adding to your statement
but I wanted to make sure you saw the joke :)
 
ah.
 
joke accepted :-)
 
Joke missed, so joke unaccepted.
 
@blue or @anon whoever is there : Is there any real use of galois categories other than geometric galois theory?
 
2:00 PM
@BalarkaSen dunno
 
i.e., why should i think galois categories are of interest?
 
why should anyone think galois theory for groups are of interest?
 
@RickDecker: If you're here about the flag, I'm sorry.
 
@Nick yeah, that's why the comment about better explanation. Flagging is the correct thing to do if there is a disruptive behavior that needs attention (either for clean up or for remediation) but if the flag description isn't clear, a whole lot of blue is going to show up to figure out what the heck it actually meant
 
Sorry, I have to ask again: Why is it only $f(x)$ and not $f(\{x\})$ on the right hand side? $$\int_1^\infty\frac{f(\{x\})}{x^2}dx=\int_0^1f(x)\Big(\sum_{n=1}^{\infty}\frac{‌​1}{(n+x)^2}\Big)dx$$
 
2:02 PM
i dunno @blue. personally i think it's a piece of trash.
 
> a whole lot of blue is going to show up
aww yiss
 
lol
 
@blue: ... Is blue a bad word I don't know?
 
@rehband this is elementary, split the interval and use the variable change ...
 
@Nick not unless you're talking about waffles, no
 
2:04 PM
@AJHenderson: So, you get to delete unnecessary things from the chatlogs?
 
@Chris'ssis I did that, but shouldn't we have $$\int_0^1f(\{x\})\Big(\sum_{n=1}^{\infty}\frac{1}{(n+x)^2}\Big)dx$$ on the right hand side?
 
@blue: Is waffles a bad word I don't know?
 
@rehband {x}=x for x in the range [0,1)
 
@rehband Well, just focus on this one $$\int_n^{n+1}\frac{f(\{x\})}{x^2}dx$$
 
@blue Of course, thank you, and thank you Chris
@Chris'ssis Got it :)
 
2:08 PM
@rehband did you? How can you write $\{x\}$? So, you have $\{x\}=x-\lfloor x \rfloor$
 
@blue a positive point about galois theory for groups is that the Fix operator transforms any short exact sequence 1 --> A --> B --> C --> 1 of groups into 1 --> Fix(A) --> Fix(B) --> Fix(C), which apparently is the key object of cohomology theories.
also, the left-exact seq 1 --> Aut_H(G) --> Aut_N(G) --> Aut_N(H) might or might not be related to the theory of derived functors, in the process of extending it to the right.
that's all the interesting stuffs i could think about.
 
@rehband but on that integral $\lfloor x \rfloor=n$. The rest is a piece of cake (well, all is a piece of cake).
 
@Nick among other things
exciting isn't it ;)
 
@AJ: Can you delete this line <---
 
@Nick yes
 
2:12 PM
@rehband and that inner series Id' write it in terms of polygamma function.
brb
 
@AJ: ... hmm, care to demonstrate?
 
@Nick nah ;)
 
:D
@AJ: Then atleast tell me this:
28 mins ago, by Nick
@BalarkaSen: The video, I linked it for a reason, $$\int_0 ^{\pi / 6} \sec y \cdot dy = \ln(\sqrt3)(\text{What})^{64} + C $$ Quickly, find $\text{What}$
 
@Nick I am not in the least interested in that.
 
@BalarkaSen: I know
 
2:15 PM
Already told you that I have a strong dislike for integrals.
 
@Nick btw, @ messages only work with atleast 3 characters
so I don't actually get notified by @AJ
 
@Nick I am going to help @blue farming an army of 10k+ers to co-rule over the world and hang the analysts all over the world.
 
@BalarkaSen: I was shooing away #user44517 earlier with something silly that looked complicated
 
mad scientist laughter
 
mwhaha
 
2:17 PM
@robjohn I also proved that $$\sum_{n=1}^{\infty} (\psi^{(1)}(n))^3=9\zeta(2)\zeta(3)-\frac{25}{2}\zeta(5)$$
 
@Nick huh?
 
@Chris'ssis: Congratulations
 
@robjohn an easier version you find here math.stackexchange.com/questions/882621/…, but the crazy thing is that I did this version without pen and paper :-))))
I know, it's hard to believe, but it's true! (no need for integrals at all) :D
This question makes me thing of a very intriguing question though ...
Find the closed form of $$\sum_{n=1}^{\infty} \frac{\psi^{(-1)}(n)}{n^3}$$ where $\psi^{(-1)}$ is the negapolygamma of order $-1$.
 
2:38 PM
Hey, everyone!
 
@Chris'ssis Sorry, logged off right after I understood. Thank you!
 
@rehband Welcome. :-)
@robjohn In general, these series are hard to be evaluated by integral means, I've convinced myself about that.
brb
 
@Nick I don't know where you got the constant of integration from, but the 'what' is simply equal to $1$ or $i$. $$ \displaystyle \begin{aligned} \int_{0}^{\frac{\pi}{6}} \sec y \text{ d}y \ \ \overset{u = \sec y + \tan y}= \int_{1}^{\sqrt{3}} \dfrac{1}{u} \text{ d}u = \log ( \sqrt{3} ) \end{aligned} $$
@BalarkaSen Why do you dislike integrals?
 
2:55 PM
$$\int \sec(x) \ dx =\int \frac{\cos(x)}{\cos(x)^2} \ dx=\int \frac{(\sin(x))'}{1-\sin(x)^2} \ dx$$
 
3:13 PM
Nice! That's a pretty neat method. It reminds me of the integral of the tangent function. $$ \begin{aligned} \int \tan x \text{ d}x = \int \dfrac{\sin x}{\cos x} \text{ d}x = - \int \dfrac{-\sin x}{\cos x} \text{ d}x = - \int \dfrac{(\cos x)'}{\cos x} \text{ d}x \end{aligned} $$
Cotangent too, in fact.
 
@BalarkaSen The four-term sequence you wrote down is exact. But I see no reason for the last map, in general, to be surjective
 
4:12 PM
@Chris'ssis You said you were going to show me a proof yesterday, but I don't think I saw one.
 
Breaking news: tag was removed.
 
@robjohn Here
 
@MartinSleziak Congrats!
 
@JasperLoy Why? It wasn't my effort to remove the tag.
Although I am glad that we don't have to remove it manually.
 
@MartinSleziak Because you seemed very concerned about it!
 
4:15 PM
Maybe it was because I wanted to keep it. When there were no changes for a long time, I still had hope.
 
@Chris'ssis got it...
 
@robjohn I hope you like it. :-)
 
@Chris'ssis were all three of those the same? one was so small, I couldn't read it
 
@robjohn Yeap. The key is to cleverly split the initial sum and get some smaller series easy to manipulate.
 
@Chris'ssis What tex editor do you use?
 
4:21 PM
@JasperLoy TeXworks
 
@Chris'ssis Yay! Same here.
 
Does anyone know of an algebraic topology text that treats infinite-dimensional manifolds?
 
@Chris'ssis Is that not just zeta derivative?
the negapoly part
 
Is it true that if $A = \{ 1, 2 \}$ and $B = \{ 2, 1 \}$, then $A = B$?
 
4:34 PM
@Khallil Yes.
 
@Khallil Sets are equivalent up to order and multiplicity of the elts, so yes.
 
@BalarkaSen What do you mean by the multiplicity of the elements? Oh wait, I think I know where you're going with this.
 
$\{1, 2, 2\} = \{1, 2\}$
 
@BalarkaSen Wait, how is that true?
 
@Alizter $$\log\left(\Gamma(x)\right)$$
 
4:36 PM
@Khallil That is the definition.
 
@Chris'ssis I realised dw.
@Chris'ssis I was confused because polygamma(z, q) and i mistook zeta(z, q) as q being the expoenent...
 
@BalarkaSen I've never seen that definition before. :O
 
@Alizter btw, have you ever seen the way the negapolygamma of order $-2$ looks like?
 
@Khallil It is a standard definition. Otherwise, you'd get a multiset (google it)
 
Yes I literally just plotted it :)\
 
4:38 PM
@Alizter look here
 
What is g?
 
Barnes G-function.
Ah, @Chris'ssis beat me to it.
 
@BalarkaSen It's easier to me to post links. :-)
 
I keep marvelling at the fact that Chris Sis is a GENIUS.
I cannot understand how she comes up with her formulas.
 
4:41 PM
@Khallil I don't know why I dislike integrals. Why do you like mathematics?
 
@robjohn as a matter of fact, because I was talking about Barnes G-function, the series $X$ in that proof can also be computed using this function (Barnes G-function).
 
@BalarkaSen I honestly have no idea. We might as well be going in around in circles, questioning each other as to why we like/dislike things that we just like/dislike!
 
@Chris'ssis I am still working on how you got
$$
\sum_{n=1}^\infty\left(\psi(n)+\frac12\psi'(n)-\log(n)\right)\\
=\lim_{n\to\infty}\sum_{n=1}^N\left((n+1)\log(n+1)-n\log(n)-\log(n+1)-1+\frac1{2n}\right)
$$
 
@JasperLoy Are you kidding, right? :-)
 
@Khallil It's similar here. I have no idea why I dislike integrals.
 
4:43 PM
@Chris'ssis No. I also suspect that you are Cleo, lol.
 
@JasperLoy lollllllll :-)
 
@BalarkaSen I did some reading on multi sets (the wiki entry at least), and I can't find why $\{ 1, 2, 2 \} = \{ 1, 2 \}$.
 
@robjohn Well, consider the Abel's summation, and the rest flows naturally. See there I put $a_n=1$, $b_n$= the remaining part
 
@Khallil You are asking, like, 'why does 1 + 1 exist?'. It's the definition of a set.
 
Balarka has confused Khallil.
 
4:45 PM
@BalarkaSen I found something on the axiom of extensionality. I'll read that. ^_^
 
@Khallil A set is defined to have distinct elements
 
"In mathematics, the notion of multiset (or bag) is a generalization of the notion of a set in which members are allowed to appear more than once."
^ that is the relevant quote from wikipedia which should stand as a plausible reference if you don't want to believe my word for it.
 
@Khallil As sets in higher math, they are equal. As sets in elementary math, they may be considered distinct as the two copies of two are taken to be distinct.
 
@BalarkaSen Mhm, I read that. I'm still not satisfied.
@JasperLoy What kinds of distinctions are there between elementary and higher math?
 
I don't get what's to be dissatisfied about.
 
4:47 PM
@Chris'ssis Abel summation, like integration by part, plays the difference of one sequence off the sum of another. I don't see how this follows from Abel summation, or perhaps I don't see the two functions you are using.
 
@Khallil So maybe many elementary math texts should be rewritten to avoid this confusion later on.
 
@JasperLoy Then it's not {1, 2, 2}. It's {1, 2, 3}
 
@BalarkaSen I am sure you know what I mean.
 
No, I don't know what you mean.
 
@Chris'ssis give me a second
 
4:48 PM
How could two copies of 2 be distinct? 2 = 2, for heaven's sake!
 
@robjohn $a_n=1$ and $b_n=\psi(n)+\frac12\psi'(n)-\log(n)$
 
@BalarkaSen Cardinality is one of my issues. If $\{ 1, 2, 2 \} = \{ 1, 2 \}$, then $|\{ 1, 2, 2 \}| = |\{ 1, 2 \}|$ so we might as well be saying that $3 = 2$.
 
@Khallil That is not the cardinality. You are overlapping two elements.
 
@BalarkaSen Have you seen the way some elementary texts are written?
 
@Khallil The real cardinality of P1, 2, 2} is 3 for the count - 1 for the overlap = 2. Free. Your. Mind.
@JasperLoy No, I haven't.
 
4:51 PM
@BalarkaSen Like I said, in some contexts, they may take the two copies of two to be distinct elements, so let's relabel them as a and b instead of 2.
 
So really, that notation as 1,2,2 is BAD notation.
 
@JasperLoy Ah, that's precisely the point of confusion made then.
 
@robjohn see above
 
It should be 1,a,b where a refers to the first 2 and b refers to the second 2.
 
4:52 PM
Right.
 
What's the difference?
You've just called the same thing by two different names.
 
@Khallil You are thinking of the Venn Diagram here. Don't take the 2s as two different points. They are the same point.
 
However, in higher math, we define A=B to mean (x in A) iff (x in B)
Using this formal definition of equality of sets, 1,2,2 = 1,2
 
Hmm.
I think I read the wiki passage incorrectly.
 
Don't even read about multisets now.
Just read the transcript, what I said.
 
4:54 PM
@robjohn the limit in the right side goes to $0$ and we remain with that sum only.
 
I somewhat agree with @Jasper
 
Of my limited knowledge of sets, $\{ 1,2,2 \} = \{ 1, 2 \}$, whereas for multi sets, those two are distinct.
 
@BalarkaSen Of course. =)
 
@Khallil Yes.
Sets and multisets are two different beasts.
 
@JasperLoy That makes sense, according to this axiom which I half read. (en.wikipedia.org/wiki/Axiom_of_extensionality)
 
4:56 PM
@Khallil Go through the standard axiom : en.wikipedia.org/wiki/Axiom_of_extensionality
Whoa, JINX!
 
How I hate pokemons.
 

« first day (1476 days earlier)      last day (3535 days later) »