« first day (1596 days earlier)      last day (3418 days later) » 

11:04 AM
Not upvote, but FAVORITE :D
 
r9m
@Venus That's already my favorite ! (really nice question !!!)
@Chris'ssis nope :( cluelessatm ..
 
@r9m Come from a favorite guy too :P
 
r9m
@Venus ya ! I consider sos440 a great mentor ! (I get inspired by his awesome answers !) ... unfortunately he doesn't talk to me anymore =P
 
@r9m do you recall we once had an argument about division
 
@r9m I consider Ramanujan a great mentor (the world is like that: Ramanujan and the rest) :-)
 
11:10 AM
@r9m I talked with him yesterday when he answered 2 of my questions. Not much.
@Venus, thank you! I tried many ways until the hat fits her perfectly. Your compliment rewards my effort :) — sos440 yesterday
 
r9m
@skullpatrol :o division ?! I can't remember ! :o
 
@r9m ok, nvm it wasn't important :-)
It was about the difference between $\div$ and /
 
@r9m Seriously speaking, don't you see the solution wpap?
 
r9m
@Chris'ssis nope :(
@skullpatrol I am scratching my head atm .. I totally forgot what was it about ! :O
 
@r9m Split the integral, then let $a x\mapsto x$ in the first integral, $b x\mapsto x$ in the second one, then bring the integrals together again. That integral you get has a nice primitive easily got by the integration by parts. Done.
I plan to propose it for a math contest.
 
r9m
11:18 AM
@Chris'ssis 'kay thanks for the hint ! ^^ (honestly I haven't tried it yet ... I try later plz ... when I'm done with the problems at hand that are beating my brain into jelly)
@Chris'ssis you shouldn't reveal these in any forum chat like this if you are proposing it for a contest ! :O
 
@r9m Didn't you try it?? Well, I thought you tried that ...
@r9m Well, the major part of the questions I see in the local contests are known to me ... (it's just a local contest).
 
r9m
@Chris'ssis I promise I'll forget about the hint when I try it again ^_^ (I can do that .. ;) ..)
 
@r9m Well, you might be surprised to find out that NO ONE found an elementary solution so far, and I gave it to many people.
 
r9m
@Chris'ssis I see ... how do you know I don't have a friend in your locality that I share problems with ? ;)
 
@r9m Yeah, I see your point, but that's less probable. :-)
 
r9m
11:23 AM
@Chris'ssis you don't know that for sure ! :P like you don't know for sure if I actually know your real name ! ;)
 
Gotta go
 
r9m
I suck at being creepy !
 
@Venus Ah, yes, I have known about that.
 
@r9m :-)
 
Is this chat all about math?
 
11:29 AM
I hope so.. I have a big math problem..
 
Me2
 
r9m
@Chris'ssis but in this case its a negative ! I haven't exchanged problems with any Romanian friend other than you :-)
 
@Stefan is that a calculation?
 
Yes M*e+2=1208
 
@Stefan ohhhh.. Nice calculation bro
@Stefan I love it
 
11:32 AM
@MZeinstra I know! I bet no one knows how I just did that calculation
 
@Stefan I do..
 
But now my huge math problem
I can't figure out what 1+1 is :( Can someone help me with this huge problem
 
@r9m :D
 
@Stefan 1+1 = 11 .. You dumass
 
r9m
@Stefan 10 ?
@Chris'ssis you didn't collect any new hats ? :)
 
11:39 AM
@r9m No time for hats, I'm working on a new article I wanna publish. :-)
 
r9m
@Chris'ssis Cool ! :D
 
10?
ok thanks
so 2+2 = 900?
 
r9m
@Stefan I added them in base 2 !
 
What do you mean?
 
r9m
11:48 AM
@Stefan $1_2 + 1_2 = 10_2$
 
idk i think $1_2+ 1_2 = 1_1241241$
but not sure
 
r9m
hmm .. I don't know what + is anymore ! is it the cross we bear across ?!
 
idk
im getting confused by math
 
12:03 PM
Congratz
 
Sheesh, I have wrote up a lot of stuff on this answer.
 
You are a lot of stuff
 
12:27 PM
So, I'm trying to click these start mathjax link in chat, and I can't seem to figure out how to get the stuff to rendor here. I'm running chrome
 
12:39 PM
@BalarkaSen Written, not wrote.
@Alan Render, not rendor.
 
I reserve the right to make silly typos when sleep depped and theoretically supposed to be studying for quals :).
 
Maybe this community should pay me for proofreading their theses and papers.
 
lol
@JasperLoy proofreading their these s and papers.
 
1:25 PM
@Chris'ssis That's lots of removed messages :O
 
@Hippalectryon lol, I posted a solution to a simple integral, but an exceptionally interesting, very interesting from some points of view. I'm amazed that no one of the people that tried that managed to finish it elementarily.
 
taken thanks
 
@Chris'ssis Why can you not use frullani here?
 
@N3buchadnezzar You can, but I wanted to finish all in a very simple way.
 
Huy
I have 10 operators, let's call them p, p^t, k1^t, k2^t, k3, k4, l1^t, l2^t, l3, l4. I want pair operators without ^t with operators with ^t such that 5 pairs result. There are many ways to do so, however I have certain rules I want them to follow:

I want to pair p with k1^t or k2^t and p^t with l3 or l4 or the other way around.
The remaining of k1^t or k2^t can only be paired with the remaining l3 or l4.
The remaining ks have to be paired with ls, e.g. k3 with l1^t and k4 with l2^t.

If I'm not mistaken, there are a total of 16 ways to pair them. I am too lazy to write down all possible p
 
1:28 PM
@Chris'ssis Well frullani is a very simple method :p
 
I cant figure out how to calculae 1/8 of 37
 
@N3buchadnezzar I didn't learn Frullani integrals in high school (hmmm, or I did learn? If that happened I'm sure the rest of my colleagues had no idea about it since we have nothing like that in ourtextbooks)
 
Swapetiy swopety i am coming for that $\frac{f(ax) - f(bx)}{x}$ integral
@Chris'ssis You should probably ask them to demolish the school then. It is not worthy of exsistence.
 
@Chris'ssis How do you justify the last step ?
 
@Hippalectryon What last step?
 
1:31 PM
@Chris'ssis Well you find the result for $a,b\rightarrow1$ then you expand it to all $a,b$
 
I know what 1/10th is but how come 1/8th is higher than 1/10th of 37
 
@Hippalectryon I don't know what you mean. I think everything is very clear there.
 
@Chris'ssis You first compute the integral for $a,b\rightarrow1$ right ?
 
@Chris'ssis the presentation was very clear =) Imho. You skipped out proving the convergence, but that iseasy.
 
@Hippalectryon Note that I let the variable changes $a x \mapsto x$ and then $b x \mapsto x$
 
1:33 PM
oh I get it im going the wrong way
 
@Chris'ssis O_o I thought that by $ax\rightarrow x$ you meant "ax tends to x" lol
 
@Hippalectryon no. That is my notation. :-)
 
Ok then :)
 
@Chris'ssis I like that notation. I use it myself
 
@N3buchadnezzar OK :-)
 
@Hippalectryon Now I finish it using one of the Ramanujan's integrals ... (writing up the proof)
 
@BalarkaSen WHy don't you upvote
@BalarkaSen did you use data.SE to find it ?
 
:P
nah @Hippa was just going through the alg topo posts
 
@Chris'ssis You've seen the ramunjan pdf's ? They is nice yes
 
@N3buchadnezzar I saw some, and, yes, they are very nice. Nothing can be compared to Ramanujan's work in the area of integrals, series and limits. He's the god of them.
 
1:38 PM
@Chris'ssis Well ryzik and Gradzyk has some nice ones.
Yeah I spell that different everytiem
 
@N3buchadnezzar Yeap, indeed. I think they also have some by Ramanujan.
 
@Chris'ssis Loads. I like how they added a couple of thousand integrals without proofs. then everyone just accepted it...
 
If somebody is free then please look at this stupid question, I talked about this to @BalarkaSen yesterday.
 
After many years people started asking questions, and they were like. I dunno where I found this. Dont remember :p
 
@N3buchadnezzar This year, in October, it appeared the new table of integrals ... I don't have it yet ... :-(
(the 8th edition)
 
1:41 PM
@Chris'ssis Luckilly people like Victor Moll has started publishing proofs for the integrals. He recently published a book filled with proofs.
 
@N3buchadnezzar Which book? I think these guys should publish lots of books, they have the potential of doing it ...
 
@N3buchadnezzar Wow, awesome!!! Do you have it??
 
@Chris'ssis Nah. But I have read the papers. They are accesible online. Thinking about buying it though
But I am more looking forward to volume 2 ;)
 
@Huy Not directly. And by "efficient", do you mean that the programme should do few calculations, or that writing the code should be fast? The answers would be different, and if you have small numbers of things, the efficiency of the algorithm doesn't matter much (unless you're doing it really often).
 
1:45 PM
@N3buchadnezzar I wanna have it ... :-(
 
@Chris'ssis Wait a sec
 
@Chris'ssis Then buy it :p
 
@N3buchadnezzar Yeah ..........
 
@Chris'ssis Here a present
 
@Chris'ssis It's soon Chris'sistmas
 
1:47 PM
@Hippalectryon Awesome! Thank you! :-)
 
bbl
 
2:10 PM
how many different arrays of length n are there with x 1, y -1s and n -(x+y) 0s?
 
@Lembik What do you think the answer is ?
 
n!/(x! y! (n-x-y)!)
 
Why ?
 
that's the problem :)
it's really a guess
 
-___-
 
2:13 PM
Wow @DanielFischer 750+ votes :D
 
I'm sure @Sawarnik will help you :D
 
:O :O
@Hippalectryon I have to go .. just dropped in to check .. but bye.
 
Hello!!
Show that the knapsack problem (Given a sequence of integers $S=i_1, i_2, \dots , i_n$ and an integer $k$, is there a subsequence of $S$ that sums to exactly $k$?) is NP-complete.

Hint:Use the exact cover problem.



The exact cover problem is the follwing:
Given a family of sets $S_1, S_2, \dots , S_n$ does there exist a set cover consisting of a subfamily of pairwise disjoint sets?
First of all, to show that this problem is in $\mathcal{NP}$ do we have to do the following??


A nondeterministic Turing machine can first guess which the subsequence of that we are looking for is and
 
@Lembik I don't wanna give a wrong answer, I'm not sure :/ ask on main
 
I think I got it in the end
by googling more :)
but thanks
 
2:18 PM
@Lembik Ok :)
 
Huy
@DanielFischer: Writing the code should be fast and simple.
@DanielFischer: Eventually, I might need it for more operators, but not e.g. 1000, the order will stay rather low (10 right now). But I just thought since I will be doing these calculations many times, I might as well write a program.
 
Off topic, @Huy, you look younger than 22 on that photo.
 
Nah, he looks at least 73
 
@DanielFischer Heya. If $f$ is periodic ($f(x) = f(x+T)$. How do I prove that $\int_0^\infty f(x) \mathrm{d}x$ diverges?
 
@N3buchadnezzar You start by requiring that $f$ is regular enough, and $f\not\equiv 0$. Then look at $g(x) = \int_0^x f(u)\,du$. Look at $g\lvert_{[0,T]}$.
 
2:28 PM
So $g(\infty) = \lim_{n \to \infty} n \int_0^T f(u) \,\mathrm{d}u$ which diverges
 
@Huy The most difficult part will be handling constraints of the form "if x is paired with y then ...". I have no immediate idea how to do that. And choosing a good representation for the constraints. What programming language, by the way?
 
@DanielFischer I just think that argumentation looks funky. It proves that the integral goes to infinity, but clearly it should not since f is periodic. where is my logical flaw?
 
@N3buchadnezzar Not necessarily so easy. Consider e.g. $\sin$.
 
@DanielFischer Right. So then we have $\infty \cdot 0$.
 
@N3buchadnezzar Well, kind of sort of. To get a rigorous proof, distinguish between $g(T) = 0$ and $g(T) \neq 0$. In the latter case, which argument immediately shows the divergence of the integral?
 
2:38 PM
What is 1+1?
 
@Stefan e__e
 
@Stefan That depends. If it is a trick question, the expected answer might be 11.
 
@DanielFischer The latter, since then the integral blows up?
@DanielFischer In which base? :p
 
@N3buchadnezzar Formulate it precisely?
@N3buchadnezzar String
 
@DanielFischer Its 2
 
2:41 PM
@Stefan So apparently not a trick question.
 
It was a huge trick question.
 
If one says 11, it'll be 2. If one says 2, it'll be 11. Trickery indeed.
 
Ok a harder one now what is 1208124+159127521*1241221412421/12512521
 
$\approx 15785188411459.94951337144609$
 
Thanks!
i can just throw away my calculator and ask it here now :-)
 
2:45 PM
I used a calculator.
 
I know
 
what is the meaning of pie?
 
@N3buchadnezzar Principle of inclusion and exclusion? French for magpie?
 
The definition of a plane in $\mathbb{R}^3$?
 
@N3buchadnezzar ?
 
2:50 PM
@Hippalectryon I also have a present for you ...
 
@Chris'ssis :D
 
One often define planes in R3 by for example $\pi:=2x + 3y + z -5$
 
@N3b yeah I have seen that a lot in multivar calculus and highschool
 
My carefully written loooong answer got a single upvote.
Meh.
 
@BalarkaSen One upvote to rule them all
 
2:54 PM
LEL
Gross.
 
@Chris'ssis Thanks :D
 
@Hippalectryon The result is awesome. :-)
@Hippalectryon Did you take it?
 
@Chris'ssis Yep :D
 
@Hippalectryon OK
@Hippalectryon $2/\log(2)$ :-)
 
O_ooo
Mindblown
 
3:00 PM
@Hippalectryon :D
 
Huy
@DanielFischer: I'm flexible regarding programming language. I usually just use Python or even PHP for a simple task like this, depending on which PC I am using.
@DanielFischer: Off-Topic - Asians always look younger than they are.
 
@Huy Except when they reach 40. then they look 120 for the rest of their lives.
 
That^ is only for the women ;D
Hey @Hippalectryon our avatars look as if we are brothers :D
 
@skullpatrol In thailand that image represents both.
 
Sanic doesn't have any siblings
 
3:15 PM
Yesssssssssss, Baby Rudin came in the mail!!!!!!!!!
 
Nice! You gotta love that new mail feeling, @teadawg1337!
 
@teadawg1337 Sending babies in the mail is unadviced.
5
Even though baby rudin is gooood.
 
Could someone explain to me the definition of a vertex cover??

A vertex cover of G=(V,E), an undirected graph, is a subset $S\subseteq V$ such that each edge of G is incident upon some vertex in S.
 
@Huy I guess Python would fit the task better. Your constraints are always of the form "x can only be paired with a,b,c,..." and "if x is paired with b, then y must be paired with e,f,g (and z must be paired with h,i)"?
 
3:31 PM
Bah I messed up some open cover argument and now I am too lazy to patch it up.
Maybe I should delete it.
 
3:45 PM
You shouldn't leave up incorrect answers, no.
But if yours is the only answer you should fix it.
 
Fixed, I think.
This works, @Mike?
 
Huy
@DanielFischer: Yes, they will always be of that form.
 
Hello! Can anyone suggest me which of these is a good book for math olympiad prep.?
a)Winning Solutions
 
Really off topic, but is that L-Drago in your avatar @MathGod?
 
Oh boy.
 
3:58 PM
b)Mathematical Olympiad Treasures by Titu Andreescu
 
It'll take me a while to look, but I'll do so in maybe half an hour? @BalarkaSen
 
It's been years since I watched that anime @Khallil.
@MikeMiller OK.
 
@KhallilBenyattou Yes.
 
It was so great. Did you watch it subbed or dubbed, @Balarka?
 
dubbed.
It was great.
Metal riders was also good.
I have forgotten it @Khallil. It was so cool. I want to watch it again.
 
4:04 PM
Metal riders?
I never saw that one. I saw Metal Fight Beyblade, MFBE and 4D all subbed. It was so awesome!
 
@KhallilBenyattou Did you see the old beyblade series which had dragoon, dranzer etc.
 
Yes! That was my gateway drug to the Metal Fight/Fusion series.
The original, V-Force and Revolution were so good, @MathGod!
 
Dragoon, yeah
The names all make sense
@KhallilBenyattou Metal fighters yeah LEL
 
It was all about the Tyson vs Kai face-offs!
 
@BalarkaSen @Huy Could you help me to FAVORITE this problem of mine? I need 2 more fav mark. Hehe
 
4:10 PM
However none of those animes beat Inazuma Eleven.
 
@Huy Okay. I'd say you make two types of constraints, one the simply (x,[a,b,c,d]) type saying x can only be paired with one of a,b,c,d, and the if-then type, ((x,a), list of constraints). Keep a dictionary of these - note: for the recursion, the easiest way to "undo" modifications to the dictionary when you go up the recursion tree again would be to copy the dictionary and pass the copy to the recursive call.
Then, for each x, look up which of a,b etc it could be paired with, for each possibility, add the pairing to a list, look up the resulting constraints and add them to the dictionary [the copy you're going to pass], and recur. When all things have been paired, add the list to the big list of all possible pairings. If no partner is available, bubble back up.
 
Huy
@Venus: +1.
 
Hi all
 
Hi Daniel
 
Hello @DanielRust
I have set a puny bounty on the inverse limit of graph question of yours.
Looked interesting.
 
4:19 PM
@Huy @BalarkaSen Thank you guys. I love you so much. Haha
 
Yes I noticed. You should have mentioned it first because I actually solved the problem a while ago, and I was planning on self-answering :P
But I'd feel bad now taking the bounty from someone else.
 
Ah @DanielRust.
@DanielRust That's nonsense. Answer it. I'd love to see a solution.
 
It's essentially just an application of the (known) analogue in the category of finite sets, and the fact that a simplicial map is determined by its action on vertices.
I used a version of the statement in my paper (although didn't provide a proof).
 
Inazuma Eleven < Death Note, Fullmetal Alchemist (Original or Brotherhood), Code Geass, ZOIDS and Naruto to name a few.
@BalarkaSen
 
I am actually working on a version where $G_n$ are cyclic graphs @DanielRust.
Of course, in that case, $G$ is no longer finite.
Psst : I am thinking of a way of defining a notion of \delta-hyperbolicity in profinite group ;)
@Khallil DBZ > all of them
Ba Dum Tss
 
4:24 PM
You mean something like $G_n$ is a cyclic graph with $p^n$ vertices?
 
mmhmm
 
Yeah, the statement wouldn't apply in that situation because they don't all belong to a universal supergraph
 
yup
 
How can DBZ be better than all of them, @Balarka?
 
i am actually thinking of making Z_p hyperbolic in a nontrivial sense somehow. maybe i should think of infinite union instead of inverse limit dunno
 
4:26 PM
It's so overrated. I mean, I enjoyed it so much, but it's not as good as Death Note or FMA Brotherhood.
 
@DanielRust in that sense the cayley graph of Z_p would be the boundary of a standard p-adic tree
which makes sense, as the gromov boundary converges to the cantor set
my ultimate goal is to define hyperbolicity in Gal(\bar Q/Q) and thus take over the world :P
 
The reason I needed it was because I had cellular maps between finite graphs in an inverse limit, but then each approximant had a distinguished subgraph $S_n$ with the property that $f_n(S_{n+1})\subset S_n$ and such that $f_n|_{S_{n+1}}$ is a simplicial map. So I have this simplicial complex appearing as a subspace of my (generally non-CW) inverse limit.
 
@Hippalectryon
 
@DanielRust actually i think this guy $\varprojlim K_n$ is the abstract simplicial compelx assosiated to some open cover of the solenoid
 
@Chris'ssis
 
4:30 PM
i.e., the nerve
 
taken
@Chris'ssis It isn't even that much shorter :D
 
@BalarkaSen yep
 
@Hippalectryon :D
 
@Chris'ssis High School maths can be powerful at times
 
@Hippalectryon Yeah. That integral emphasizes the big difference between ways when using a brilliant idea. The idea is to try to be brilliant in everything you do and never be content with solving a problem only.
 
4:32 PM
@BalarkaSen wow you just taught me \varpojlim...... I've always used \displaystyle{\lim_{\longrightarrow}}
 
heh
 
@Chris'ssis :)
 
@MikeMiller O_o
 
risky click
 
4:34 PM
@Mike What the.........
 
ok i gotta go. meanwhile, i've made a puny progress in my understanding of grothendieck-teichmuller theory. i've added it here.
 
Apparently I'm a little bit ready for Christmas. I need to prepare more before the night comes.
 
@MikeMiller you hat.
:P
 
I love how many bounties there currently are... so many people earning hats
 
Turn down for hat
 
4:37 PM
@DanielRust In principle it's good; but it means fewer things will get attention - there's too many!
 
@MikeMiller agreed
 
For my Christmas bounty I'm just going to award an answer I like +100 rather than seek new answers.
 
Don't you need to wait a day before awarding a bounty?
 
@MikeMiller Answer my hard unanswered questions :D
 
I actually earned a bounty that way once for an answer, it's always nice to see someone willing to give you rep for a good answer
 
4:39 PM
Yes, @DanielRust, but I think the hat is for placing a bounty.
Congratulations on your Naruto, @Behaviour.
 
So what hats are you trying for at the moment @Mike?
 
I'm looking for a -3 question to turn good and give a good answer to.
(I also need to get 30 Minutes or Less - but figuring out what to ask for that is a hard problem.)
 
what's that one?
 
Ask a question that gets 5 answers within 30 minutes.
 
o.o
The only time I see something like that is a PSQ
 
4:43 PM
I figure the right question is a good reference request, but I need to think of a reference request that a) I actually want to know b) is accessible
Right, I'm not going to stoop to asking PSQs.
 
I suppose a biglist question might work too
 
Right
But at the same time, I'd like to make a good question
 
Yeah
 
So I don't just want to go the easy route like this guy did.
 
haha wow
 
4:49 PM
Lol
 
They are really crazy in the ELU room :D
 
That's Pizza Hat
 
@MikeMiller Thanks. Got to 21 with Aztec. Will try some bountied question next... but need more bounties, why don't more people get Gelt?
 
@Venus That's a very beautiful lady.
 
4:55 PM
@JasperLoy I wanna use my own picture but I don't have pizza hat
 
@Venus You don't need a stupid hat.
 
@MikeMiller The one for starting a bounty between Dec 16 and 24. Easy hat for the low price of 50. Everybody go start bounties!
 
There are lots of bounties out there right now.
 
@JasperLoy But I like hat.
 
I gave a stupid, awful answer to a Linear Algebra reference request. I feel dirty.
Luckily it also got two good answers that will actually get the bounty
 
4:57 PM
@MikeMiller Where is it?
 
If you want to see it, you can find it.
 
3
A: Text suggestion for linear algebra and geometry

Mike MillerAs darij grinberg comments above, there's Linear Algebra and Geometry by Suetin, Kostrikin, and Manin; it's fairly difficult, but it should be accessible, given the time between now and when you originally asked this question. I didn't read all of it, but quite liked what I did. It has plenty of ...

 
Is she really beautiful, @Jasper?
 
@KhallilBenyattou Yes. I wish I can find a girlfriend like that.
 
The bounty I offered I actually would really like to see an answer to
It's a question I wondered about before
 

« first day (1596 days earlier)      last day (3418 days later) »