« first day (1181 days earlier)      last day (4135 days later) » 

05:00
@user60887 Well, start by using the definition.
well there is a theorem that says floor(x + y) <= floor(x) + floor(y)
yeah like that. How would I prove floor(x + y) <= floor(x) + floor(y)?
and since you know n is a natural number then the floor of it is just the number itself so floor(x + n) = floor(x) + n
@anon Good bye, fellow stranger.
good ol' django
05:06
sorry it should be the case that floor(x) + floor(y) <= floor(x + y)... i thinl
think*
yeah that way. the other way is wrong
@user60887 Use dem definitions. $\lfloor x\rfloor=n$ is the unique integer such that $n\leqslant x<n+1$.
Note that if $x'=x+m$ with $m$ an integer, then $n+m$ has this propertyw.r.t. $x'$, simply by summing in the above. Thus $\lfloor x+m\rfloor =\lfloor x\rfloor +m$
so do i let x=floor(x)+phi and y=floor(y)+alpha where 0<phi,alpha<1 to prove that t floor(x) + floor(y) <= floor(x + y)? and break it up into cases
@FernandoMartin Yiss. Nos vemos mañana?
No sé, yo voy a las 5
05:10
Greetings the great ones!
@FernandoMartin Ah, yo voy de 11 a 4 casi 5.
De ultima podemos tomar un cafe entre las 4 y las 5.
Que tenes, Topo o NT?
i dont feel so great yet
@PedroTamaroff do you like my problem you may see in the right side of your screen?
05:11
suelo llegar 20 min antes igual, así que dale
@FernandoMartin Coool.
floor(x) + floor(y) = m + n <=> x + y >= m + n <=> x + y >= floor(x + y) <=> floor(x) + floor(y) >= floor(x+y)
@Chris'ssis I don't know, haven't looked at it. To avoid tiring my eyes, I zoom in the page a lot, so the starts are invisible.
i think I am making a logical leap ...
Now I'm off.
05:15
yeah and it should be floor(x) + floor(y) <= floor(x+y) at the end
uhh sorry I am a little tired ... that was a bad mistake
yeah its ok. i got an idea. ill try it tomorrow
thanks for helping though
no problem dont feel discouraged, proof writing takes time to learn
its like any other skill, requires lots of practise
practice* ... like spelling :P
yeah i know. haha yeah. i just want to become good with floor funcitions because they seem pretty cool when you apply it to number theory.
lots of cool stuff in number theory :)
05:19
well im going now. peace
have a good night!
$\floor(x)$
$\ve$
i guess latex doesnt work here
$\lfloor x\rfloor$
what is \ve supposed to be?
isnt that what latex calls the OR symbol
(you are using the bookmark linked on the starboard right?)
you mean $\vee$
bookmark link?
yeah $\vee$
05:21
see "LaTeX in chat" linked on the starboard
ohhh
hehe got it!
$\vee$
so much better!
thank you anon!
mmhmm
so... how are you?
... do I know you?
i dont think so :P ... I have come here once before
05:41
Hi
I am a newbie here, is there a list of codes so i can write math expressions like they would normal like 2^2/2*3
would look normal
Wow thats Anon (bookmarked the page) :)
thanks
@gbox. Don't use this lightly, but MathJax tutorial should also prove useful for you. Everyone appreciates well written posts.
thanks @J.W.Perry
 
6 hours later…
12:10
Hi! I have a little question on the $\TeX$ notation on the main site. It isn't possible to mark more than one line by "\tag{n}", what should I do to enumerate more than one line in a formula? What I'm doing now is adding "\quad (n)" after each line to be enumerated
12:52
Greetings great people
@robjohn have you seen my last problem posted? It's on the right side of the screen. I had a lot of fun with it. Maybe you like it.
What does $$\lim_{x\to\infty}\frac1{\arctan x}$$ converge to?
@Chris'ssis I will look. I was gone last night and then came home to sleep.
@robjohn Some party? :-)
I also need to attend some parties ... for a long time I haven't attended one
$F(x)=\mathrm{W}(e^x)$
@robjohn It was given on a high school contest. I think we might avoid the use of the Lambert W function.
13:08
@Alizter $\frac2\pi$
@robjohn Thanks it struck me shortly after writing it
Is this correct then? $$f(x)=\frac{F(x)}{1+F(x)}$$
@Alizter It is.
@Chris'ssis Im a terrible person though :(
I used Lambert D':
How is a high school student suppose to know this?
@Chris'ssis I get the limit of $0$, but that seems uninteresting.
13:17
@robjohn you mean you find the problem as being uninteresting?
$x=F(x)+\log(F(x))$
so $F(x)\sim x$ and $f(x)\sim1$
@Chris'ssis I thought it would give a non-zero limit.
@robjohn this is the way some mathematicians created the problem. I also initially thought I'm wrong, but this is the limit, that is $0$.
@Chris'ssis Okay. I thought I was missing something. :-)
@skullpatrol it's meant to avoid the use of Lambert function.
After some work I got 0 too but I cannot see how this is possible without $W$
13:21
@Nimza The align environment allows you to \tag more than one line.
@Alizter Take logs of both sides and then differentiate
@robjohn exactly.
When $x$ is large, $e^x=F(x)e^{F(x)}$ forces $F(x)$ to be large, that eventually means that when $\lim_{x\to\infty} F(x)=\infty$
indeed
@Alizter After doing what robjohn said, you get $$f(x)=\frac{F(x)}{1+F(x)}$$
@Alizter taking the limit to $\infty$ we get that $\lim_{x\to\infty} f(x)=1$. Hence, our limit is $0$.
@Chris'ssis I got it now thanks :)
Did any of the students get it right?
13:27
@skullpatrol No one of those I personally know.
How about partial marks?
@skullpatrol It's not hard if you pay attention at details.
@skullpatrol partial marks? What do you mean?
Do you think this will ever be answered? math.stackexchange.com/questions/508829/…
@Chris'ssis If the question was worth 10 marks what would it take to get 5/10?
Partial marks means part of the marks
@skullpatrol As regards this question all I know is that some of the students that received this question made no progress. They didn't even know how to tackle it. It's sad but it's true.
13:34
@skullpatrol hi skull
@Charlie hi Chuck
@skullpatrol how are you?
@robjohn how would you rate it as difficulty level?
@skullpatrol how would you rate it as difficulty level?
@Charlie Fine thanks, how are you?
@skullpatrol I'm fine :)
13:38
@Chris'ssis Out of 10, I would say it is an 8.5
Definitely an A- student question.
@skullpatrol what question?
@Charlie The one with 4 stars on the right panel ------>
13:52
@robjohn thank you!
@Charlie hehe, hi Marilinha!
@Nimza Hi leshikinho!
@Charlie :)))
@Nimza how are things today?
13:55
@Charlie I'm at work now, and what about you?
@Nimza work too
@Charlie I also predict that cocoa and milk make you smarter. In my case such a drink works great! :D
@Chris'ssis the only natural thing that make ou smarter is to study
@Charlie TRUE
hmmm, I think I prepare right now some cocoa & milk.
brb
@Chris'ssis I prefer milk derivatives
14:02
@Charlie Can milk derivatives be integrated back to milk?
@skullpatrol I hardly think so XD
@Charlie :D
@skullpatrol :D
@Chris'ssis I believe that being breastfed is also a very important step to grow up as a smart child
Anyone spot anything to make this easier to compute?
$$\int_0^e x^{1/x}dx=2e\sum^\infty_{n=0}\sum^n_{k=0}\frac{(-1)^{n-k}}{e^nk!}+C$$
14:21
@Alizter Your double sum is: $$\sum_{k=0}^\infty \tfrac{(-1)^k}{k!}\sum_{n\geq k} \frac{(-1)^n}{e^n}$$ use the geometric sum there, to get $$\sum_{n\geq k} \frac{(-1)^n}{e^n} = (-1/e)^k / (1+1/e)$$ and then sum the exponential
@PaulRS Are you sure about the variables on the first double sum?
@Alizter You were summing first over $n$, and then over the $k$ satisfying $k\leq n$, right? .. if you do it the opposite way, sum over $k$ and then for the $n$ greater than that $k$.
So you'd get $$\sum_{n=0}^\infty \sum_{k=0}^n \frac{(-1)^{n-k}}{e^n k!} = \frac{e^{1/e}}{1+e^{-1}} $$
@Alizter What is your constant $C$ ?
@PaulRS It snuck in there when I was doing the definite integration sorry its not supposed to be there
@Charlie I know, it is important.
@PaulRS I am not too familiar with geometric series so can you go through the steps a bit more? On your second step can you explain how that becomes that?
@PaulRS If it is easier math.stackexchange.com/questions/508829/… is the question which you can answer
14:36
@Alizter Okay, I am going to explain it there then.
@Alizter You need to check this because the left side is approximately $2.6618257053804178285$ and the right side is approximaely $5.7417551233587916696$
@Chris'ssis :)
@robjohn You are right!, @Alizter It'd be nice to see how you arrived at that expression then.
@skullpatrol I also proposed for a high school contest the following question. Compute $$\sum_{n=1}^{\infty}\arctan\left(\frac{1}{(2+\sqrt{3})n^2+(2+\sqrt{3})n+2-\sqrt{3}}\right)$$ and express the answer in terms of $\pi$
@skullpatrol from the feedback I received so far I can say that most probably it will be rejected.
@skullpatrol the teacher I talked to failed to compute it and then I sent him my solution. (or maybe he was too lazy to compute it)
It's said the mathematicians are lazy persons, you know. :-)
14:56
but to make it, most people are
15:11
fale @marcelo
@skull skulls don't cry
Can anybody offer any input on my question here? math.stackexchange.com/questions/542921/…
15:41
leo
leo
@agent154 the multiplication table is full of mistakes
@leo It wasn't intended to be a multiplication table; It's supposed to be a power table... but I think I realize now that the top row should have been $1\cdots 6$ instead of the elelements in $\mathbb{Z}_9^*$
leo
leo
@Charlie como você está?
@leo Is a "Primitive Root" synonymous with generator?
leo
leo
@agent154 what is a primitive root?
15:47
If $a^n\equiv 1\mod{m}$ and $(n,m)=1$, then $a$ is a primitive root $\mod{m}$....
@robjohn I do not see any mistakes but here is the derivation math.stackexchange.com/questions/459147/…
@leo bem e você?
It would appear that when I draw a table, all the generators of $\mathbb{Z}_9^*$ appear to have order $\varphi(m)$
leo
leo
@Charlie bem :-)
Let me correct my definition above: A primitive root is one where $a^n\equiv 1\mod{m}$ and $n=\varphi(m)$
leo
leo
15:49
@agent154 also, what do you mean by $\Bbb Z_9^\ast$?
It's the set of all integers $\mod{9}$ that are invertible; so all the integers relatively prime to $9$
$\mathbb{Z}_9^*=\{\overline{1},\overline{2},\overline{4},\overline{5},\overline{‌​7},\overline{8}\}$
leo
leo
the set of primitive roots?
@agent154 and you want to find a generator for it?
@agent154 I've posted an answer. Tell me if you understand it.
leo
leo
@agent154 If you want to find a generator for a finite cyclic group, you can just try with all its elements, and which one works
I'm aware of that, but I was wondering if there's a direct way to find one instead of guessing and testing each element in the group...
15:58
@robjohn
@Alizter but is $\log( x) / x$ integrable in $[0,e]$, it has a nasty singularity at $x=0$?
@PaulRS I'm aware how to test if a number is a generator... just raise each element in the set $\mathbb{Z}_m^*$ to every power $n$ such that $1\leq n\leq \varphi(m)$ and see if you get every element in the set back... but I was wondering if there's a direct way to find a generator without having to test every element in the group.
Ah! an algorithm?
@PaulRS You're right but then what am I producing?
Was looking for more like a formula, because my aformentioned method works as an algorithm already. Not very fast though.
16:01
@agent154 There is a randomized algorithm I know... which works well if you can factorize $\phi(n)$, it should be pretty fast, but... that is good for a computer, not so much for a human I think
@PaulRS Is my calculation right up to here? $$\sum_{n=0}^\infty n!^{-1}\int x^{-n}\log^nx\space dx$$
I'm only worried about being able to solve questions on a test quickly... but I guess the questions won't be hard enough such that I need to test that many possiblities
I have a test in 30 minutes and I'm trying to practice how to find generators of cyclic groups
@Alizter I am going to check your solution in detail
@PaulRS Inform me when you have something :)
@agent154 A detail, note that, since $4 = 2^2$ and $8=2^3$ by checking $2$ you've also checked the other 2.
@agent154 If you can factorize $\phi(n)$, then there is a nice way to check whether a number if a generator, it is here math.stackexchange.com/questions/542921/…
16:06
@PaulRS So for $\mathbb{Z}_9^*$, then I only need to check $2$, $5$, $7$, since $1$ is never a generator, and $4$ and $8$ are powers of $2$? Could it ever be a case where $4$ is a generator but not $2$?
@agent154 No, since 2 generates 4 trivially (independent of your modulo)
So if $g$ doesn't generate a group, then any power of $g$ will not do so either;
So I only need to check the primes
@agent154 exactly
Can someone help me with this
0
Q: Point Group of a pattern

user43418I need to determine the point group of the following patterns: I think the one on the left is $C_1$ and the one on the right is $D_2$

@agent154 no,
16:09
So in that vein, can generators only ever be prime numbers?
@agent154 you have to check products of distinct primes, because, say
Oh, ok.. so $2\cdot 3$ could be a generator
@agent154 exactly, it may happent hat 3 is not a power of 2, or 2 not a power of 3 (in your module)
But I don't need to check $2^2\cdot 3$
@agent154 in this case, you don't ahve to consider $3$ since is not coprime to 9 :)
16:11
OK... thanks a lot for the help. I'm gonna run off to my test now.
@agent154 what the trick means is, if you have an element $b$ that satisfies $b = a^k$ in your module, if $a$ is not a generator, neither is $b$.
@agent154 indeed, if $\{ b^j : j\in {\mathbb Z}\}$ was the whole thing, then also $$\{ b^j : j\in {\mathbb Z}\} = \{ a^{k\cdot j} : j\in {\mathbb Z}\} \subseteq \{ a^j : j\in {\mathbb Z}\}$$
and so, if the first set were the whole group, so would $\{ a^j : j\in {\mathbb Z}\}$
leo
leo
@agent154 good luck
You guys. November 08 is now officially going to be the biggest day of my life.
why?
@leo what have you been studying, leo?
leo
leo
@Charlie Galois theory
@DonLarynx good for you :-)
16:25
@leo nice :)
@agent154 good luck!
@DonLarynx $\huge \text {WHY?}$
leo
leo
@Charlie nice indeed. I've no arived to the main results yet, but I'm trying to understand
@Charlie :D
@leo there are subjects that fascinates me, even though I understand almost nothing
16:28
@PaulRS hi
@Charlie Why so serious? (the cat)
Haha, that cat is so badass
I like her
we have a lot in common
@Alizter Try integrating in $[\epsilon,e]$ and then letting $\epsilon\to 0$. I am worried about the case $x=0$...
16:37
@charlie my first advanced calculus (real analysis) test is in less than two weeks! Nov 08is when I get to show whether I'm the real deal or I am only a squeal
@DonLarynx hmm...
@Alizter I think the sum $\sum^\infty_{n=0}\sum^n_{k=0}\frac{(-1)^{n-k}\log^kx}{x^nk!}$ converges only for $x>1$.
What I am trying to get at is if my method was right until the integration of $\log^n x$
@charlie hmm?
leo
leo
ha ha haç
grumpy cat
16:43
:D
@DonLarynx nothing
I want to no of solutions of the following equations hold .

x+y+z+t+w+m+n = 8
hmm?
@osimerpothe can you explpain it a bit better?
I am doing a programming assignment:
Write a Python function, evalQuadratic(a, b, c, x), that returns the value of the quadratic aâ‹…x2+bâ‹…x+c.

This function takes in four numbers and returns a single number.
---I am not sure what it is asking, I knwo how to find the root, but there are two of those, so that can't be what they are asking for
I have an equation: x+y+z+t+w+m+n=8. I want to know how many solutions does this equation have? x,y,z, and t, w,m,n are positive integers.
@osimerpothe it can have many solutions
@osimerpothe you have one equation for 7 variables
16:53
I want to the no of solutions
yes .
Count them.
leo
leo
no of solutions?
no = number
@osimerpothe: what have you tried?
I have come to know that if an equation has n variables whose sum is k then the no of solutions is $ \binom{k-1}{n-1}$
4
Q: How many solutions of $ x+y+z+t = 7 $

Way to infinityI have an equation: $ x+y+z+t = 7$. I want to know how many solutions does this equation have? $x,y,z$, and $t$ are positive integers. I have no idea how to solve this. Can you please help me to solve this question?

16:55
Well, then it's solved
But according to this formula ,$ x+y+z+t+w+m+n=8 $ eqn has only $ \binom{7}{6}$ or 7 solutions .
@Alizter You do realize that that expression blows up at $x=0$?
It has 7 solutions up to rearrangement
@robjohn Yes people have been telling me that :)
@Alizter Sorry, I've been working and just peeked back to see your comment. Didn't mean to repeat that at you :-)
16:57
@robjohn It's fine! At least you are now aware
But I think that the eqn has more than 7 solutions .
That is my confusion .
It has exactly 7 solutions
well . let me explain .
They are of the form 2,1,1,1,1,1,1
1,2,1,1,1,1,1,1
and so on
You have exactly 7 places to put your 2 on
if all of them are zero except one .
16:59
You said your solutions are positive integers
Then they can't be zero
oh Sorry . I have not noticed that .
Thanks
I want to know the proof of this formula .
Yes . I have that this problem is known as classic stars and bars combinatorics question.
There's a link to a wiki article on the first answer
which has a proof
So I have found the proof .
 
1 hour later…
18:31
If you guys saw the crystals on these nugs you'd go insane. They're sugar dipped
@EnjoysMath hmm?
I have home grown weed
you can come over and smoke some if you want
ah, fine :)
@EnjoysMath where you live?
Sedona, AZ
18:40
medical mj is legal here
@EnjoysMath ah, so you are very ill
lucky rascal.
it's "legal" everywhere...socially
Suppose $Y \subset X$. A subset $E$ of $Y$ is open relative to $Y$ iff $E = Y \cap G$ for some open subset $G$ of $X$. Does this definition imply $E = Y$?
hello
@what'sup hi :)
18:48
how are you ?
@what'sup fin, fine, and you?
fine too thanks
great
19:00
Don Larynx, no it doesn't since that's the standard definition of subspace topology
@Don
Well actually, does $E = G$?
No
Think of a blob in 3D
intersect it with a bunch of open balls
so some will intersect on the boundary and bey "half-open" and "half-closed" if our blob were closed
so those intersections are no longer necess open in $X$, but by definition they're open in $Y$
if $G \subset Y$ then it's an open set in both $Y$ and $X$
but $G \subset X$
I recommend you draw a blob on your paper and intersect it with open balls from its mother space
I did......
so $E \subset Y$ and all points in $Y$ are also in $E$
19:06
What's your question again?
X = reals. Y = [0,1]. E= [0,1/2). G=( -1,1/2).
Is that statement correct?
If that's the case then $E = Y$, yes but that's not the case in the setting you presented
What's the difference?
19:09
Just look at the example I posted.
Draw a picture of $Y\subset X$ should be two circular objects one enclosing the other
draw a typical open set in $X$ that happens to touch $Y$
That intersection area forms $E$
why would $E$ then necess be $Y$?
makes no sense pictorially
$$\int\left(\frac{\log x}{x}\right)^ndx$$ Anyway to integrate this?
@EnjoysMath show him using paint
heh was going to
@Don, still need help on that?
19:31
I get it now
I was getting confused
I thought the entire radius had to cover $Y$. It only has to cover $E$
hi @anon
hello
how are you doing?
meh
why? what happened?
19:37
nothing has happened
meh =/= bad
we never know, maybe something caused your mehness
So, what's an interesting topological ring on the integers?
@anon
@Alizter Integrate by parts $n$ times?
@EnjoysMath 15:31
I get it now
I was getting confused
I thought the entire radius had to cover Y. It only has to cover E
True?
@Bitrex If only it was that simple
I am trying to find a general solution
19:45
Oh
@EnjoysMath well, there are the topologies on Z that come from lifting the discrete topology on quotients Z/n
not only does it have to cover $E$ it has to be $E$
$E = Y \cap G$
you mean the smallest topology such that $Z \to Z/n$ is continuous?
yes
basically where the cosets of nZ are the open sets
19:47
what topology on $Z/n$ then to start with?
or do you mean the other way
@EnjoysMath discrete
set of all subsets?
google tells me that Weber classified all "locally bounded" topologies on number rings
@EnjoysMath yes
why?! isn't there a more interesting one?
sure
one can pick any family of subsets and then speak of the topology they generate
in general they will not be so simple as "discrete modulo n"
19:50
@Alitzer the coefficients of $x^{n-1}$ in the denominatior of the indefinite integral seem to be related to this sequence: oeis.org/…
@Alizter but the coefficients in the numerator are weird and OEIS has got nothin'
Taylor series of $$\left(\frac{\log x}{x}\right)^n$$
I can integrate if its a taylor series

« first day (1181 days earlier)      last day (4135 days later) »