« first day (1888 days earlier)      last day (3126 days later) » 

12:34 AM
why is it true that $gPg^{-1} \cap gNg^{-1} = g(P \cap N )g^{-1}$
?
 
1:00 AM
$f(A)\cap f(B)=f(A\cap B)$ for any bijection $f$
 
 
2 hours later…
3:32 AM
yeah I see
@anon is it true that every p subgroup of G/N, where N is normal is of the following form PN/N ?
I need this for a proof in algebra
 
hello
 
@KarimMansour my guess would be no
hello
 
how is everyone?
 
if you assume p doesn't divide |N| it's true
fine
 
yes P is a sylow P subgroup
because I am trying to prove the following
$n_p(G/N) \leq n_p(G)$
This can be done by $\phi : Syl_p(G) \rightarrow Syl_p(G/N)$ and proving that this map is surjective
hm
 
3:41 AM
do you have the sylow theorems at your disposal?
 
yeah
I can use it
 
G acts on both by conjugation. phi is a G-equivariant map between two orbits, so it must be surjective.
 
to put this abstractly: suppose G acts transitively on two sets X and Y, and f:X->Y is such that f(gx)=gf(x) for all g in G and x in X. then f must be onto. see if you can prove it.
 
oh
actually I proved something like that in set theory before
 
3:45 AM
you had group actions in (elementary?) set theory?
 
yeah towards end of it
By the way @anon do you like category theory ?
 
it's a nice condiment.
 
we had this question in topology last week
it was very nice question
I didn't know you can define product in abstract terms not just for sets
 
lots of stuff is actually categorical
 
Do you agree with this definition of product?
 
3:51 AM
yeah @KarlKronenfeld
I asked this question to my prof
in topology but he told me its hard to find examples of such things
I asked him the following
Is there category of objects that isn't just sets with extra structure on it ?
because it seems that category is just sets with extra structure
I mean is their things that are more general than sets?
 
consider posets and groupoids
if you want something nonconcretizable, the homotopy category
 
I see
 
4:07 AM
Hi, I was wondering if someone could help me understand an answer for a particular question
0
Q: language of bitstrings with no more than 3 consecutive zeros generating function

cp101020304I am trying to find the generating function of a sequence in the language of bitstrings, $X$, where each bitstring contains no more than 3 consecutive zeros. I have come up with the recurrence relation for a sequence $h_n$ of length $n$ as follows: (I am not 100% sure that it is correct, either...

How is the second term $h_1xh_2x^2$
I would assume it to be a1*x
h1*x sorry
 
4:55 AM
I am gonna buy artin group theory book
DF sometimes sucks
 
Huy
5:39 AM
sell me your DF
 
nah I want it aswell
 
Huy
but sometimes DF sucks
 
sometimes not all the times
haha
 
Huy
:(
 
5:55 AM
heyo
 
 
1 hour later…
7:09 AM
Glad I could share it with you. He's a guy with 3 MIT degrees who left his to make this site, eventually funded by google and Bill Gates foundation.

His math practice system is all AI based ;)
 
 
2 hours later…
9:27 AM
Does anyone here like to discuss algebra sometimes?
I just want to discuss quotient rings
 
what's the question, @Understand?
 
Well I just wanted to discuss adjoining roots
So taking $\Bbb R[x] /\langle x^2 + 1\rangle$
Is that the same thing as taking $(\Bbb R[x])(i)$?
 
it's isomorphic to $\Bbb R(i)$
 
Oh sorry, yes
So if we quotient a field by an irreducible polynomial, it means the quotient ring is a field
 
that's not true. $\Bbb Z[x]/(x^2+1)$ is isom to $\Bbb Z[i]$ - not a field.
 
9:32 AM
$\Bbb R[x] /\langle x^2 + 1\rangle \cong \Bbb C$ and then $\Bbb C$ is an extension of $\Bbb R$, so then $\Bbb C$ is a real vector space of dimension two(which ignores multiplication)
 
what you mean is that if $k$ is a field, then $k[x]/(p(x))$, where $p(x)$ is an irred, is a field. in fact every field extension of $k$ arise in this way.
 
Sorry yes, I am not use to instant messaging
 
well, *finite algebraic field extension
grump, finite algebra galois field extension.
 
and if we have an degree $n$ irreducible polynomial, and we quotient a field by it as a generator, i.e. $K[x]/\langle f(x)\rangle$ we get an $n$-dimensional $K$-vectorspace as a result.

But I think I can't get that until I understand what happens with the multiplication on $\Bbb C$ with our 'two dimensional real vector space'
 
@robjohn Have you seen this one? My Mathematica is not able to calculate it. $$\int_0^1 \frac{\operatorname{arcsin}(x)}{x \sqrt{x+1}} \, dx$$
@robjohn It has a very cute closed form.
 
9:38 AM
does it converge chris's?
 
@Understand multiplication on $\Bbb R[x]/(x^2+1)$ is the same as multiplication on $\Bbb C$, so I am not sure what the question is. do you know what the isomorphism between the two is, explicitly?
 
@Understand Yeah, note that $\displaystyle \lim_{x\to 0} \frac{\arcsin(x)}{x}=1$. Near $0$ the integrand behaves like $\displaystyle \frac{1}{\sqrt{x+1}}$
 
@BalarkaSen I know of a construction that proves it must be the case, but I haven't got an explicit map
 
the map takes $p(x) \pmod{x^2+1}$ to $p(i)$.
 
I like it
Thank you Balarka
 
9:44 AM
no problem
 
@Chris'ssistheartist Mathematica 8 gives an answer with lots of Hypergeometric functions.
I will have to go to my other computer to check Mathematica 10
 
@robjohn Well, yeah, my Mathematica too, but I didn't want to call that one an answer. :-)
 
@Chris'ssistheartist I don't like people using Hypergeometrics as closed forms. Most things that can be written with a decent series can be written as Hypergeometrics
 
@robjohn Agree.
 
10:05 AM
@Chris'ssistheartist I did a bit of manipulation, but Mma 10 gets $4G-\pi \operatorname{coth}^{-1}(\sqrt2)$
 
@robjohn Yeah :D
 
I should go back to 8 and see if that manipulation helps there
@Chris'ssistheartist since it involved $\frac{\arcsin(x)}{x}$, I am not surprised with the appearance of $G$
 
@robjohn I came across this one while I was working on a problem by professor Bradley that I wanted to calculate it in a different way. I'll show you my work later when I put all in latex. I also wanna include it in my book.
@robjohn But that $\sqrt{x+1}$ makes things look bad.
 
@Chris'ssistheartist I was looking at handling that with a series...
 
$$\sum _{n=0}^{\infty } \binom{-\frac{1}{2}}{n} x^{n-1}=\frac{1}{x\sqrt{x+1}}$$
 
10:11 AM
@Chris'ssistheartist If I ask Mma 10 Integrate[x/Sin[x]/Sqrt[1 + Sin[x]], {x, 0, Pi/2}] it gives the answer I quoted above. Mma 8 gives something with some complex polygammas
 
@robjohn Mma 10 seems to come up with a lot of improvements in terms of integration.
 
@Chris'ssistheartist well, Mma 10 gave the Hypergeometric garbage with the original statement of the question.
 
@robjohn Ah, OK
@robjohn Using that series (as you suggested) is enough to calculate the integral. There are some boring computations, and then you're done.
 
11:01 AM
@PantelisSopasakis Your hair is out of control
 
@Understand Haha! Cheers :)
@Understand that photo of mine is a bit old... now I got short hair
 
@PantelisSopasakis Hahaha my hair isn't that long, precisely so I don't have to worry about it
Oh I once had crazy hair too, but people worried I looked 'unprofessional'
 
but still it's dishevelled
no, quite the contrary... in the mathematics community a fuzzy hair style is considered perfectly professional
 
Tell me please @PantelisSopasakis, how do you approach problems, if you were to write it like an algorithm
@PantelisSopasakis Oh well in that case, I am excited :)
 
if you look like an Einstein, people tend to think you will be as clever as one
what do you mean by that?
 
11:05 AM
I mean the naive problem solving method is:

1) Read definitions
2) Attack problem

if solved, end, else go to 1)
Have you thought about precisely what your study method is? I only asked because I looked in your profile and you have masters + PHD :P
 
Do I?
Ah, yes...
Well, I have a focus or two (currently for example I'm mainly working on model predictive control and optimization methods based on proximal operators), so I read articles and books to, first of all, state an interesting problem
and then I'm looking for a solution... But I don't have a very algorithmic approach. I study, discuss, post questions on stackexchange,
try things on MATLAB to get a better understanding...
I would say
 
Discuss is what is missing for me I think, I do think that's really important. When did you gain a network to discuss with?
 
that when trying to develop new things -- is it theory or algorithms -- intuition is more important that rigor
Well, that's what stackexchange is for...
but I also have my colleagues at work
 
But stackexchange doesn't let me 'discuss' unfortunately. With your colleagues, do you ever worry what they will think of you? Do you verify things before you will say them? I have trouble with the other students because I worry they will think I am dumb
Thank you for your advice @PantelisSopasakis !
 
Hello everyone
 
11:17 AM
Hi pal
 
someone said that $L^\infty$ is connected, but normally $L^\infty$ is kind of weird, what do I need for a measure and a space such that it is connected?
 
Posing a question, admittedly, requires some effort from our side as well. And it has happened to me more than once to write down a question carefully and clearly, look at it and find the solution...
 
11:32 AM
@gideon keep working on your maths, all my biggest scores in bitwise/arithmetics/datastructs challenges are reached because of simplified mathematical equations, especially combinatorics
i like simplifying boolean expressions too
 
What is this number sequence?
1, 1, 2, 4, 10, 24, 66, 178, 508, 1464, 4320, 12886, 38992, 119030, 366740, 1138036, 3554962, 11167292, 35259290, 111825840, 356100044,...
It must have something to do with the Riemann zeta function. Possibly some kind of reverse function.
 
@Agawa001 Yes I know. Thanks for the link though.
 
If I work over an algebraically closed field $K$, and $f(x)\in K[x]$, then $f(x)$ always factorises into a product of linear polynomials, being the roots, is that correct?

I.e. $f(x)=(x-a_1)(x-a_2)\cdots(x-a_n)$?
 
12:27 PM
We say the polynomial splits and it is up to a product of linear terms and a constant. The linear polynomials are $(x-a_i)$ where $a_i$ is the root, and it is important to careful distinguish this(obvious after thinking about it for a sec) fact

$f(x)=b(x-a_1)(x-a_2)\cdots (x-a_n)$
 
12:44 PM
Hey everybody! I posted a problem about half a month ago and it is still unanswered. here Can anybody see into it and help me?
 
1:06 PM
I am trying to show $\{\frac{n^2-1}{n^2}\}$ is cauchy using definition...is this supposed to be composing in an easy way!
I feel I am doing wrong
 
1:20 PM
I was indeed being silly - it is just inverse squares when written under cauchy...
 
1:44 PM
@Understand Yes, because $K$ contains all the algebraics over itself, by definition of being "algebraically closed".
@GBeau Yes, the sequence is just $\{1-1/n^2\}$, so it is clear that you can get arbitrarily close pair of points in your sequence at some point.
 
well, right, the question was asking to show it formally using the definition (i.e., explicitly construct the cauchy criterion)
it was easy once I realized it was 1/n^2-1/k^2
yeah
:)
 
correct. it's not hard to derive that it's Cauchy once you visualize what the sequence is, though.
I really shouldn't have said "clear" up there, though. Just meant you can see what is going on.
 
Hi @DanielFischer
 
@Moses Hi.
 
woo. I just got a mail with an excellent article on reciprocity laws and galois representations
 
1:53 PM
$\sum _{n\in\mathbb{N}} x_n$ converges $\implies $ $\sum _{n\in\mathbb{N}} (x_{2n}+x_{2n+1})$ converges (this is easy since those are subseries.) My book says the converse fails - what kind of examples should I be looking at?
 
$x_n = 1$ when $n$ even and $-1$ when $n$ odd?
 
@GBeau Something where $x_n$ doesn't converge to $0$.
(Exercise: If $\sum (x_{2n} + x_{2n+1})$ converges and $x_n \to 0$, then $\sum x_n$ converges.)
 
I will work to prove this ^
 
@DanielFischer What do you think of my proof. Remember you said that the spectrum must be restricted to a subalgebra of $\mathcal{A}$ when writing $\sigma(a_{k}) = \sigma_{k}$ in the Riesz Decomposition Theorem. But I'm proving that $\sigma_{\mathcal{A}}(a_{k}) = \sigma_{k}$ holds. Let me know if you can see a problem with the proof.
 
thanks
 
1:59 PM
@Moses Small problem at the start, you need to require $\sigma_1$ and $\sigma_2$ to be closed, otherwise disjoint $\Omega_k \supset \sigma_k$ don't exist.
 
@DanielFischer Oh yes I see. So we can use that $\mathbb{C}$ is normal and hence there exists two disjoint open neighbourhoods $\Omega_{1}$ and $\Omega_{2}$.
 
@Moses But the real error is when you say $(g_k\cdot f)(\sigma(a)) = \sigma_k$. For $j \neq k$ and $z \in \sigma_j$, you have $(g_k\cdot f)(z) = g_k(z)\cdot f(z) = 0\cdot z = 0$, so $(g_k\cdot f)(\sigma(a)) = \sigma_k \cup \{0\}$.
 
2:18 PM
@DanielFischer Damn. Yeah I see, so close...
@DanielFischer How could we alter this proof to coincide with the discussion we previously had. Can we restrict the spectrum to some subalgebra of $\mathcal{A}$ so that my proof still gives a result?
 
@Moses Well, let $\mathscr{A}_k = \overline{\operatorname{Alg} (E_k,a_k)}$. Then $\sigma_{\mathcal{A}_k}(a_k) = \sigma_k$. I don't think you can use the spectral mapping theorem to see the latter, though.
 
@DanielFischer Okay so I would have to change the proof almost entirely?
 
I think so.
 
2:33 PM
@DanielFischer Could I just ask what the exact reasons are for choosing $\mathscr{A}_k = \overline{\operatorname{Alg} (E_k,a_k)}$ and if other subalgebras would also work?
 
@Moses We need a Banach algebra in which $E_k$ is the identity, and that contains $a_k$. That is the smallest such. Above that, it has the advantage of being commutative. But any closed linear subspace $\mathcal{B}$ of $\mathcal{A}$ with $E_k,a_k \in \mathcal{B}$ such that the restriction of the multiplication to $\mathcal{B}\times \mathcal{B}$ makes $\mathcal{B}$ an algebra with identity $E_k$ would work.
 
@DanielFischer Okay thanks. Just to confirm, the $\mathscr{A}_k = \overline{\operatorname{Alg} (E_k,a_k)}$ is the norm closure of the algebra formed from taking the closure of $E_{k}$ with $a$ in terms of multiplication, addition and scalar multiplication?
 
@Moses It should be $a_k$ rather than $a$ there. Apart from that, yes.
 
@robjohn do you wanna see a marvellous jewel? :-)
 
@DanielFischer Yeah typo. Thanks.
@DanielFischer Will try to redo proof.
 
2:47 PM
gah my textbook always throws in putnam problems with the easy problems
 
the jewel ^^^
 
i saw this somewhere
 
@Agawa001 I just created it. :-)
 
nearby different
 
@Agawa001 hehe, that one is a different stuff :D
I presently have enough stuff to write tons of books. However I need to do my best for my first book, and publish it in an amazing form.
 
2:57 PM
Hey guys
 
yes , good start is a good sign of success
 
if $0<C<1$, why should $\{x_n \}$ converge when $\lvert x_{n+1}-x_n\rvert\leq C \lvert x_n-x_{n-1}\rvert$? It's in the Cauchy section of my book and I don't see an obvious way to attack the problem with the other methods
 
I've been trying to figure out how (1)/(2e^(6i)) can become equal to one of these:
 
(of the section)
 
Had the option 0.5/(e^(6i)) been there it would've been trivial, but no
Can someone help me or point me in the right direction?
 
3:05 PM
@Chris'ssistheartist if you are you concerned of anything i do now, i m trying to generalize a rule of congruence
and i see that it s not less beautiful than integrals
 
@Agawa001 :D
 
@GBeau $x_n = x_0 + (x_1 - x_0) + \dotsc + (x_n - x_{n-1})$.
 
@DanielFischer This is good suggestion, I will look this now :)
 
How do I prove that if a bouned set A has a maximum M then M=sup(A)?
 
@Hjorthenify Unless you misread the value, the poser(s) of the exercise made a mistake. None of the listed values equals $\dfrac{1}{2e^{6i}}$.
 
3:09 PM
@Daniel Just realized I made a mistake. It's (1)/(4e^6i)
 
@UserX Check the definitions. Is $M$ an upper bound? Can anything smaller be an upper bound?
@Hjorthenify I expect that you then see the right value?
 
No :(
Is it option nr. 2?
 
$\dfrac{1}{4e^{6i}}$, what looks like it may be it?
 
No, so it's the LUP thus the sup. Thanks.
 
@DanielFischer I dont have a clue, sorry.. This really bothers me
@DanielFischer (1/4)*(4e^(-6i)) That?
 
3:16 PM
@DanielFischer Oh I think the geometric sum formula for $C$ gets used, I think $|x_n-x_k|$ ends up something like ~<= $(1-C)/(1-C^{n+1})$ I just need to work out specific now
or uh flip that
sorry I was just doing it in my head
 
@Hjorthenify That one isn't among the options. (Yes, it's option 2.)
 
Then I dont know.
 
@Chris'ssistheartist sure. Sorry, I am afk frequently today.
 
@robjohn $$\int_0^1 \int_0^1 \int_0^1\frac{dx \ dy \ dz}{(1+x) (1+y) (1+z) (1+x+y+z+x y+x z+yz+9 x y z)}$$
@robjohn Can your Mathematica calculate it?
 
ADG
hello everyone
i need help
 
3:21 PM
@GBeau something like. Yes, we use the geometric sum formula, you get $\lvert x_n - x_k\rvert \leqslant \frac{C^{n}}{1-C}\cdot \lvert x_1 - x_0\rvert$ for $k \geqslant n$.
 
@Chris'ssistheartist I'll have to go to that computer and check it out.
 
@Hjorthenify Option 2, $\frac{1}{4} e^{-6i}$.
 
@robjohn OK, when you have time. No hurry.
 
@DanielFischer Oh right. That was the one I meant with chat.stackexchange.com/transcript/message/24520901#24520901
Forgot to leave out the 4 :P
 
hi @DanielF, @robjohn
 
ADG
3:22 PM
if $\psi_0(x)=\frac1{\sqrt{2\pi}}\int_{-\infty}^{\infty}\phi(k)e^{ikx}{\rm d}k$
 
Hi @Ted.
 
What is the mathematical rule behind it though?
I mean I could see that was it, but I completely understand why
 
@TedShifrin Hey, Ted! Did you get any rain this weekend? We were supposed to get a good deal, but all we got was some drizzle on Sunday morning.
 
@Hjorthenify $\frac{1}{a^b} = a^{-b}$
 
ADG
what is the relation between $\phi(k)$ and $\psi_0(x)$ ??
 
3:23 PM
It's raining right now, @robjohn, just in time for me to take my car to the body shop and walk 2 miles home :P
 
@ADG Fourier transform
 
Ah, @DanielF typed faster than I did.
 
@TedShifrin Just 'cause you wasted time chatting with robjohn.
 
ADG
what is it? and how people concluded that this implies $\phi(k)=\frac1{\sqrt{2\pi}}\int_{-\infty}^{\infty}\psi_0(x)e^{-ikx}{\rm d}x$??
 
One of us resembles that remark.
It's the Fourier inversion formula, @ADG.
 
3:25 PM
@DanielFischer Yeah, and it kept me from answering, too :-p
@ADG That is the inverse Fourier transform.
 
ADG
i don't know all this but need to know just this for studying quantum physics, what could be a good source
 
@robjohn Well, doesn't matter so much who answers, as long as it gets a correct answer.
 
@ADG Any book on Fourier Analysis will go over this stuff
 
You probably should look at a math for physicists type book, @ADG, like Arfken or something.
If your math is stronger, Körner's book on Fourier Analysis is beautiful and full of fascinating applications.
OK, I'm done now :D
 
ADG
does my reputation indicate that?
 
3:27 PM
Indicate what?
 
ADG
im asking wheteher my reputation indicate my math strength?
 
Reputation doesn't indicate much of anything.
 
ADG
hmm
i think that FT is a process of converting a number into sum of periodic sine and cosine functions?
 
You need to decide if you know analysis (the theory behind calculus) or not. But my original suggestion is what most undergraduate physics majors learn their math from.
That's Fourier series, not transform.
 
ADG
*number is incorrect but it should be function
 
3:29 PM
periodic function, yes.
 
ADG
im studying real analysis too now
and linear algebra
 
Hello guys. Interrupting for a second in this conversation. i know little about Maths but had a couple of questions. All I ask is for one of you to point me to the right direction. Any fine fellows willing to help?
 
oh, you definitely should know linear algebra before you do Fourier analysis :P
Try both and see what you can read and understand.
You should ask your question, @Sampark.
 
This here is a paper of Philosophy:- biolbull.org/content/215/3/216.full. Scrolling a little down will bring you to a series of diagrams and equations. I've only studied Maths till 10th grade and thus my knowledge is limited. It does seem an awful lot like French to me.
What I wanted to know was what these equations are.
Don't explain them to me.
Just give me a brief overview and point me somewhere if you can.
 
I don't know any information theory (which is what this probably is). You need basics on logs, exponentials, and probability, it looks like. And you need to take the time to understand their notation.
Maybe @robjohn can shed more light ...
 
3:34 PM
It's some heavy stuff, eh?
And I suppose the diagrams mean little to you too?
 
Right.
But there are lots of people (not necessarily here) who understand this stuff.
 
And would be able to explain it to me?
 
I have no idea.
 
Perfectly all right! Thanks mate. :)
 
Good luck~
 
3:46 PM
Terrible network trouble to mathjax.org causing terrible lag to SE
 
Research is so great these days, pure heaven (excepting my allergy sessions).
 
@Chris'ssistheartist its autumn
everything is great in autumn
 
@Agawa001 Yeah, I know. :-)
 
@DanielFischer Of cause! Why didn't i think of that..
 
@Chris'ssistheartist except the fig tree in the backyard which gets progressivley unleaved, i see the autumn in a cheerful sens, not a sad portrait like everyone.
 
3:58 PM
@Agawa001 When I'm depressed the season doesn't matter much. Autumn is fine. :-)
 
What's up with tex rendering lately?
And hey @Chris'ssistheartist long time no see
 
@Chris'ssistheartist there s a meaninful sens of relation between automn and moral situation, automn is the outset of many things, work/studies/spousal life etc
wait, people dont usually marry in summer in your country or do they ?
 
@Agawa001 Yeap
@UserX Hey :-)
 
so it is a common sens
 
@Agawa001 It depends, I didn't check these details, I've never been really interested in it. :-)
 
4:08 PM
Depression goes way deeper than seasons unfortunately. @Chris'ssistheartist how's your research? Still dealing with integrals or did you switch to something else?
 
@UserX Yeap. I'm trying to finish my first book and publish it: a collection of integrals, series and limits (around 500 problems). :-)
@UserX You?
 
Just got back from my first day in a math uni :)
 
@UserX Oh, first day is always nice ... :-)
 
Yup we did calc 1 and got into sup/inf pretty fast. I like the pace.
 
4:18 PM
In my first Calc class we didn't ever mention sup/inf--is this an honors class or something like that? Or is that just typical at your school?
 
You have to study R on the first day right? You want to do it rigorously so you start stating axioms. How do you rigorously state the axiom of completeness? @Addem
 
We didn't study R on the first day, per se--we tend to do that in an Analysis class, where of course we discuss completeness, sup, inf, etc. But typically in Calculus things are pitched at a higher level of study, assuming these intuitive features of R.
 
Assuming for simplicity(by that I mean it has to be easy to be in calc 1)you construct reals by including it as an axiom. If you want it to be a theorem then you're in the wrong classroom.
 
oh man, these numbers go thru cycles, its like a circus of numbers
 
To be honest I couldn't really tell the difference between calc and analysis. This might sound weird but they do overlap a lot. The textbook I use for the calc 1 class is named "Differential and integral calculus; an introduction to analysis" and it was in the bibliography of the books that cover what I'll be taught in calc 1
 
4:31 PM
going to this link for me takes me to my post, then after the MathJax refresh, I am looking at someone else's post.
Does anyone else see this?
 
Well, my understanding is that Calc is where you develop your intuitions, and then in Analysis you go back over the same material but prove everything.
 
Nope, just your post. @robjohn
 
@UserX Hmm.. I've tried using both HTML-CSS and Common HTML and I keep ending up in the middle of another post.
 
@robjohn what post?
 
The middle of ADG's post
 
4:34 PM
@robjohn yes,like comments of the upper post ?
 
@robjohn Yea, after about a second it skips to the middle of ADG's post (the $L_5$ part)
 
@morphic Yes, that is exactly what I see.
 
i think it is a bug
 
More MathJax problems.
time for a meta post...
 
@robjohn maybe the style of ADG's TeX is the problem. He's using huge walls of TeX instead of cutting it up. That might cause a bug, it does in an another site that uses MathJax and I frequent.
 
4:36 PM
@UserX I will take a look at his LaTeX.
 
Did they implement the 2.6 yet? Maybe that's where the problem lies. By the way my android has(apart from the usual LaTeX rendering screw-up) no such issues.
 
@robjohn maybe because the text wasnt mathjaxed before, the cursor stays in its place when the browser is being busy unfolding the latex expression, after it the cursor points to same area of the original text (not the new uncompressed one)
 
Anyway, back to math, today my prof hinted that any set fulfilling the axioms of the reals is isomorphic to the reals. Can someone link a proof?
 
@Agawa001 In the past, after the final MathJax refresh, the window was in the correct place. It seems that something between the website and MathJax has come unglued.
 
@robjohn it would be worse on phone i guess
 
4:42 PM
@Agawa001 nope.
 
maybe mathjax is ADG-friendly :D
 
5:03 PM
I'm reading Visual Geometry and Topology and I see this notation C^(k+1)_(n+1) to mean "the number of permutations of n+1 taken k+1 at a time". But I had always thought the notation was supposed to be the reverse of that. So for instance I always thought the notation went C^n_r meant n choose r. Am I wrong or is the book using weird notation?
 
Triangle starts and row sum are A239605:
1...................................=1
1...0...............................=1
1...0...1...........................=2
1...0...3...0.......................=4
1...0...6...1...2..................=10
1...0..10...4...9...0..............=24
1...0..15..10..28...7...5..........=66
1...0..21..20..69..36..30...1.....=178

Alternating row sums are A011782:
1......................................=1
1...-0.................................=1
1...-0...1.............................=2
Can you find a generating function for these row sums?
 
Hi!!! I want to find a parametrization of the level curve: $\frac{x^2}{4}+\frac{y^2}{9}=1$.

- For $x= 2 \sin t, y= 3 \cos t$ we have $\frac{x^2}{4}+\frac{y^2}{9}=\frac{4 \sin^2 t}{4}+\frac{9 \cos^2 t}{9}= \sin^2 t+ \cos^2 t=1$

- Let $\frac{x^2}{4}+\frac{y^2}{9}=1$. Then $\frac{y^2}{9}=1-\frac{x^2}{4} \leq 1 \Rightarrow -3 \leq y \leq 3$.
We can pick a $t$ such that $y=3 \cos t $ since the latter is surjective from $\mathbb{R}$ to $[-3,3]$.
Then we have $\frac{x^2}{4}=1-\frac{9 \cos^2 t}{9}=\sin^2 t \Rightarrow x^2=4 \sin^2 t \Rightarrow x= \pm 2 \sin t$.
 
Shameless chat-bump; today my prof hinted that any set fulfilling the axioms of the reals is isomorphic to the reals. Can someone link a proof?
 
Huy
5:19 PM
what
there are non-isomorphic models of the reals
also of Peano
see Gödel's incompleteness
 
So it's incorrect?
I said hinted, because that's what he did
Maybe I'm misinterpreting
 
@robjohn Do you maybe have an idea?
 
@userx IIRC any totally ordered complete archimedian field is isomorphic to $\Bbb R$ this is probably in some notes by Pete Clark.
 
IIRC?
 
if i recall correctly
 
Huy
5:24 PM
apparently it depends on first or second order logic
for first order, there exist non isomorphic models, for second, all are isomorphic
(I'm not much into logic so I only know some theorems and examples)
 
math.uga.edu/~pete/FieldTheory.pdf For instance, Thm 194 on page 106 here.
 
@PVAL can't get my pdf viewer to show pages. Can you share the book's page span?
Oh wait you did
Nevermind I got confused over nothing
 
i have another sequence: 0, 0, 0, 1, 3, 6, 12, 25, 51, 102 it isnt even deposited in oeis
^ im serious
 

« first day (1888 days earlier)      last day (3126 days later) »