« first day (4630 days earlier)      last day (685 days later) » 
00:00 - 23:0023:00 - 00:00

00:13
Why would it feel any more weird than a $u = xt \rightarrow du = x dt \rightarrow \frac{1}{x} du = dt$ substitution? In this case $t = u \rightarrow dt = du$
In solving problems of mathematics, I am met with a seemingly impassable wall extending indefinitely in all directions. I find windows too small to fit through, and all that appears to be an entrance is a one-way exit. I honestly weary of trying to break through. On the other side, however, is paradise, and it is the paradise of complete mathematical freedom which I seek; and I have over 3000 years of hindsight nearly untapped to get there.
That being said, if I wanted to learn math based on elementary intuition of elementary mathematics alone, where would I go?
We're talking being unpracticed in elementary algebra but nevertheless grasping only as much as the elementary arithmetic operations up to and including the exponential and logarithm. I would say I know more, but everything else I learned in Algebra 2 and Geometry I forgot.
I was prevented from being able to go beyond that anyways, not that that would have mattered seeing how much I've forgotten.
01:21
Do undirected isomorphic graphs have the same chromatic number?
 
2 hours later…
02:59
@AMDG real analysis will fix your troubled past
otherwise grab a precalc book
Great, thanks, but I need names for actually good books. Perhaps maybe ones from a century ago should be satisfactory.
we don't build up to the same things anymore. hardy published his stuff about a century ago and he introduced real analysis thinking to the english world
so if you grab stuff from a century ago chances are you won't be preparing to deal with up to date stuff
i mean the above specifically for pre-real analysis maths
what you want is a good precalc book with lots of exercises and then calculus with lots of exercises
probably ton of recommendations on mathSE
I had a wonderful Dover book by Henry Burchard Fine called College Algebra when I was in high school. I don't know if it's still available. It's old-fashioned but very thorough.
03:19
@AMDG i remember learning most of my precalc and calculus from khanacademy
YMMV, i'm not so sure it's a great idea
i.e., it works, but it might be better to use it as a supplement to an actual book
 
2 hours later…
05:04
Is the identity $\lfloor\frac m2\rfloor=\lfloor\frac{m-1}{2}\rfloor +1,$ valid for all $m\in \Bbb R$ ?
0
A: Let $F_m$ be the $mth$ Fibonacci number given by $F_1=F_2=1$ and $F_{m+2}=F_m+F_{m+1}$ for all $m\geq 1.$ Show that $\sum\binom nk=F_{m+1},$

Mike EarnestOne mistake you made is when you said $\lfloor\frac m2\rfloor=\lfloor\frac{m-1}{2}\rfloor +1$. This is only true when $m$ is even. I find it is much easier to split into cases based on whether $m$ is even or odd. If $m$ is even, so that $m=2k$ for some $k\in \mathbb N$, then $$ \begin{align} F_{m...

For reference, please refer to the above post, I made, where I used this identity.
There, some users are of the opinion that this is valid, if m is odd while some are of the opinion that the identity is valid, when m is even.
Ok, now, I am sure, that this identity is valid for m even only. But the answers are a bit confusing. If I find something not clear, I will come back again.
 
2 hours later…
07:11
Can someone explain why closed and open sets aren't opposites of each other?
I was watching this video
Why does closed not imply open?
Does open imply closed?
in general, neither implies the other (or the negation of the other).
i dunno really what to say about it, except, the relevant definitions don't correspond very well with "plain English" understanding of either term. which is fine because maybe there isn't a plain english understanding of point sets.
the thing that is generally true is, if a set is closed, its complement is guaranteed to be open, and vice versa.
which is already maybe tricky because 'complement in what?' and the openness or closedness can very much depend upon how you answer that question.
doors can be open or closed, but i don't think there's a plain english understanding of the complement of a door, let alone the complement of a door in some other, perhaps larger door.
The doors analogy suggests that the correct name for a set that is neither open nor closed should be ajar
07:35
@Ajay the concept of clopen sets might help.
Hi @AlessandroCodenotti.Thanks for your comeback.
so the courses that I wanted to take (set theory and logic) are not being floated.
they are floating courses which I have no idea about.
Sieve methods for God sake
whatever that is.
Representation theory of groups... what is that?
and Lattices...
and commutative algebra
why is there so much algebra??
Big Algebra floods the schools with money to keep it that way
@Koro The one line description is that rep theory is a way to turn questions about groups into questions in linear algebra, and the latter is much more well understood that the former, so it's a good translation
But this doesn't really make rep theory justice, it's a nice subject (that I know too little about)
you may have already seen some representation theory in another course (you mentioned haar measure in connection with something once)
07:40
Oh, thanks. I thought it would be about writing 'group presentations': <generators| relations>
but you could do a whole course just around reps of finite groups
but the course structure is disproportionate (too biased towards algebra).
wikipedia's page on representation theory is fairly good, given it's wikipedia
There is topology course too but the instructor there is one of the worst here so taking that would mean screwing up grades.
I heard once in linear algebra class that about 80% of math is about calculation, which is algebra, and the rest 20% is to find the right definition.
07:43
I think the rest of the courses are not being floated due to lethargy.
There is Riemann surfaces too.
But its syllabus is huuuuge.
it doesn't look like a semester course to me.
I guess I'll be forced to take commutative algebra... as that's one of the few subjects in the list that I'm familiar with.
it's a very terrible situation.
Pressure of syllabus kills one's creativity :)
Don't try to learn everything and this is also not possible, I guess.
in their course brochure they had mentioned so many courses. Before taking admission, I thought I'll take them. I didn't know they won't be floated.
Pick your favorite subject according to your interest and try to expand knowledge in neighbouring subjects.
I don't know which of the idiots here decides which course to float and which ones should not.
@SouravGhosh yes, so for that I take one course. Fine. But I'm yet to choose another two.
And they are all related to unfamiliar territories.
@Koro I think you are interested in Algebraic topology. Choose courses accordingly.
07:51
Note that such an issue never arose in IIT.
if they say there are n number of courses then they float all n of them.
offering more freedom to students.
I have faced the same situation.
I have chosen topology. Although, I think that the instructor who 'teaches' that is useless.
Atleast this way, I stay in the familiar territory.
And I really don't want to do Ph.D here.
Koro. Not all institutes are worst and not all professors. You can take a look at IISC's official website.The course structure there is as good as any foreign universities ( honestly I am not comparing with few std US universities).
What is the order of the group $GL_2(\Bbb Z/9)$?
one: choose any nonzero vector for the first row, and anything other than a Z/9 scalar multiple of the first row for the second... oh hrm not a field
server trouble
08:05
@SouravGhosh I'm not saying that and I'm not talking about course structure. I'm talking situations at the college, I'm in currently.
$\Pi_{k=0}^{1} 9^2-9^k $
$\Bbb Z/9$ is not a field people
yeah
the server spits back my comments and asks if i want to 'retry' sending them, then rejects them, but also displays them
koro: i'm kinda surprised anyone would do a phd there, from the sound of it
08:08
@onepotatotwopotato Correct✅
@onepotatotwopotato the answerer is a very nice professor as per what I have heard.
I've not taken any class from him as he doesn't teach at my college.
So our functional analysis teacher was 'teaching' something and except some 2 or 3, the whole class was silent. He asked 'how many of you have understood?' Only two or 3 people raised hands. He still continued.
What was the point of asking 'how many of you have understood?'?
maybe if nobody raises their hand, it's in his contract that he can go home for the day?
how else is he supposed to know when the lecture is over?
08:25
@SouravGhosh and the 'less amount of time'.
@Koro I think you have to focus on those topics that interest you.
Then try to expand your range.
yes of course
08:47
So if $G = GL_2(\Bbb Z/9)$ then $|G| = 2^4\cdot 3^5$. By the proof I linked, $g\in G$ has $3$-power order if and only if its image in $GL_2(\Bbb Z/3)$ does. Now the problem is asking: Show that a Sylow $2$-subgroup of $G$ is isomorphic to $\Bbb F_9^\times\rtimes(\Bbb Z/2)$ where the nontrivial element $1\in\Bbb Z/2$ acts on $\Bbb F_9^\times$ via $x\mapsto x^3$.
09:11
Is it possible to solve? I have no idea why suddenly $\Bbb F_9^\times$ comes out.
09:35
Is there a topological embedding $f:(X\times {0,1})\times [0,1)\rightarrow X\times \mathbb{R}$ such that $f$ has open image and contains $X\times {0,1}$?
10:14
@Koro Actually the situation is even worse here (in IIT Kgp), at your institute if you're leaned towards pure mathematics then at least you get to choose from the options, here you don't get that privilege.
Among the electives, there are only two courses on pure mathematics and you have no other options than to take those courses.
12:09
@shintuku Well books I'm really only inclined to resort to because I don't know where I can find a good teacher instead outside of a university to just learn everything optimally.
Either I find a teacher for one-on-one, or I resort to books.
Either way, I need a learning plan for getting from elementary arithmetic to the depths of discrete mathematics, real analysis, complex analysis, geometry, and I'll definitely want topology.
Having a teacher is safer; getting the wrong book will probably leave me in misery. I say "from a century ago" as a guess because the world was a bit more sane back then, and I know first hand what modern "pedagogy" so-called is, and I want none of it. I cannot fathom a multiple choice question existing a thousand years ago in a university.
Khan Academy is full of that stuff. I wasted a year completing half the algebra 1 course on there when I had already taken the course 80% of the way through; and I had already taken prealgebra in like 4th grade, but the papers were mixed up, so I was put in the wrong class; then in the right class; but then, again, said 80% through the course, I was taken back into prealgebra because the teacher was afraid I would fail the final exam.
I can't fathom what her reasoning was because that probably only happened since I didn't bother with the homework. I struggled to find the motivation to do so, especially when I already grasped the concepts.
idk, man. School is just a painful recollection and reminder that no one knows how to teach today, and having become Catholic and learned me some Filosofie, I finally know why too. When I was in some sort of precalc or geometry class, the calculus class was down the hall. I went over after class and asked for the homework and he denied it to me. Bruh. The same teacher, when I had him for geometry, questioned my ability to learn calculus by probing my knowledge of terminology instead of ideas.
12:29
*taken prealgebra in 7th grade (I remember I was going into 8th grade).
@shintuku I ended our conversation in this regard for a reason. For the sake of your reputation, I will not be stating what that reason is. Suffice it to say that the thing you have rejected infallibly refutes whatever "evidence" you might present to me on the subject.
Just for the record, such "great" philosophers as Kant said that 12 is a "synthesis a priori" concerning the sum 5 + 7 = 12. He is literally claiming that 5 + 7 is distinct in essence from the essence of 12--and the whole world seems to trust this man LOL.
^ just yet one more reason for me to distrust modern pedagogy and discipline in general, and is most probably also why my own profession is in ruins.
@TedShifrin it is still available professor
13:10
What is a Sylow $p$-subgroup of $GL_2(\Bbb Z/p^2)$?
13:27
Can anyone please help me with this: math.stackexchange.com/questions/4675125/… ?
Maybe as before, $g\in GL_2(\Bbb Z/p^2)$ has order $p^5$ (order of Sylow $p$ subgroup) if and only if $\overline{g}\in GL_2(\Bbb Z/p)$ has order $p$ (order of Sylow $p$ subgroup) under natural sujection.
Umm I guess this is too much. only $p$-power...
13:47
@SoumikMukherjee I'm not so surprised. If the courses which they are offering here at my college have nothing to do with one's interest (say, they are floating only algebra related courses), then having 'options' does not matter. At kgp, if you saw the brochure of your programme, and the brochure said courses x,y,z will be taught, then they keep those words. Now, if one goes to the college they know what subjects will be there. But here, as I said: they are not floating all courses
which were mentioned in the brochure.
So if someone reads the brochure of the programme and says ohh, I'll go here as they offer courses x and y too. But upon reaching there, they come to know that neither x nor y are being offered. Instead, they have no other choice but to choose only from algebra courses.
13:59
@Koro That's pretty harsh, same thing is happening here too, they are not floating all courses which were mentioned.
But you could go to your students affairs section and ask them to. I'm sure they'll listen because that's IIT. You could inform that there are n number of students who want to take this course, please let it be offered.
we don't have any SA section here.
IITs are very student friendly. I never faced such issues during my bachelors.
@AMDG feel free to state those reasons, i don't think my reputation is at stake here heheh
Obliged against in conscience
lots of people seem to recommend gelfand for precalc
his functions and graphs/algebra/trigonometry books
@Franklin I think that the identity is true for all $m$. What makes you think it's not true for $m$ odd? Perhaps your comment above was made before you worked stuff out in the question.
14:16
@Koro We already did, they assured us but nothing happened.
@Koro Which institute are you from?
@Koro That's true, the professors (except few) here are very student friendly as well.
I'm from ISI K.
Is there any intuition for why one requires the regularity condition in the implicit function theorem?
ISI K having no SA section is very much surprising
@SoumikMukherjee During my bachelors, during festivals (Holi, Diwali etc.), we used to go to some professors' homes also, it was so much fun :-). That's how I know professors there are really nice :-). I could even go to some professors' office to ask my doubts.
That's very nice!
14:26
So at IITs, questions are encouraged. It's a very healthy eco-system there :-). Everything there is about students and for students.
@Shinrin-Yoku what's the regularity condition
But I'm not talking about that. I expected an environment completely different from IIT here before coming here. But not floating all the electives that were mentioned in the brochure is what disappointed me.
I saw a few theorems where they assume that the curve $\gamma$ is null homotopic (for example argument principle ). Why do we need that? I'm following Rudin, and he doesn't assume null homotopic.
NVM, got it.
14:45
@TedShifrin I would pose my question to you, is there any intuition behind regularity
@shin
I mean regular point of a curve
15:14
Find winding number of p and q :)
15:44
@robjohn Yeah, you are right. I was actually talking about this idenity:$\lfloor\frac{m}{2}\rfloor=\lfloor\frac{m-1}{2}\rfloor + 1$. It is invalid for all odd $m$, say $m=3$ as pointed out in the comments of OP.
Can anyone please help me, understand the Taylor Series expansion for f(x)=cos(a+h)x ?
They are expanding f(x) about which point ?
I think there's a lack of clarity in here...
Neither do I understand, the rest part after that expansion ...🥲
16:14
If you actually read it, the text answers your question quite plainly, @Franklin. They are expanding about the point $ax$ with increment $hx$.
Hello, is a caratheodory function defined on a regtangle $\{(x,y)\in\mathbb{R}^2, x_0\leq x\leq x_0+a , |y-y_0|\leq b\}$ is bounded ?
@Shinrin-Yoku I assume you're referring to the non-vanishing determinant of the appropriate part of the derivative matrix? Yes, look at the case of $x^2+y^2=1$. Why can you not write $y=f(x)$ locally near $(\pm 1,0)$? If the tangent line (plane) is vertical, you cannot write the level set locally as a graph.
@TedShifrin Am restricting to the 2D case to have better intuition. I have seen that example, but there are others like the infinity symbol where one can have a function, but still the center is not a regular point...
I guess I am trying to say that regular point does not always correspond with vertical tangent, so its not clear how one would come up with the regulairty condition
(I also don't have good intuition for when a point is not regular when the function is from R^2 to R)
@TedShifrin Umm...Now, that you mention it, but it was not so clear, though. Also, you mean "expanding about the point $a$" instead of "expanding about the point $ax$", right ?
No, Franklin, I said what I meant. $ax$ is fixed.
@Shinrin-Yoku Yes, of course. no tangent line whatsoever there. To understand necessity, think about the converse. If it’s a local graph, then ….
16:31
@TedShifrin We need to have regularity... But why is it enough(intuitively)? Is there a nice intuition for when a point is a regular/irregular?
@Shinrin-Yoku There are some other typical examples: $y-x^2(x+1)=0$ at $(0,0)$, and $x^3-y=0$ at $(0,0)$, what about $x=f(y)$
Yes. I agree. But I want intuition for when a point is regular or not, just by eyeballing a graph
@Shinrin-Yoku Then I don’t know what intuition should mean.
what is the definition of a regular point
I already told you to prove the converse.
16:36
wikipedia is not being conclusive here
So you are saying there is no way to eyeball regualarity?
@TedShifrin
I dont understand the exact statment you are asking me to prove
Eyeball from just a function? Nope.
I am telling you to prove that if it is locally a graph, then the regularity condition must hold.
@TedShifrin 👋 hey :) how are you ?
Wow. Zut @Hippa
Bon jour
16:39
@TedShifrin is not the infinity exampe a counter example?
It's been ages since I've last been here, it's nice to see there are still some familiar faces around :D
No. It is not even a graph of any function, let alone a $C^1$ function.
@TedShifrin The thing that is bothering me, is that it's said, $f(x)=(a+h)x$. Now, if we look at the first few terms of the Taylor series expansion about ax: f(ax)=\cos(a+h)ax$ , $\frac{f'(ax)(x-ax)}{1!}=\frac{-((a+h)a\sin((a+h)ax))(x-ax)}{1!}$ and so on. These were only the 1st two terms, but they aren't alike to thing written there ? Am I missing something?
@Hippalectryon Astyx vient de temps en temps !
You’re writing nonsense, Franklin. Write this as a function of $h$.
@TedShifrin Ah sympa ! Je ne savais pas qu'il avait continué à faire des maths
16:41
Ah yes becacuse any open ball containing zero wont be a function @TedShifrin
@Shinrin-Yoku what's the definition of regularity you're using
@Hippalectryon So what are you up to, besides memes?
@shintuku Just see the wiki of implicit function thm
@Hippalectryon Oui, la physique et puis les maths.
i'm looking at it, where's regularity @Shinrin-Yoku
16:43
@TedShifrin Procrastinating instead of writing my PhD 😭
So you haven’t changed :) How’s baby Hippa?
@TedShifrin So if a curve is locally a graphm the it must be regular? Is that not intuition for regularity?
@TedShifrin Doing fine, he's finished studying and started working a few months back
@Shinrin-Yoku what's regular
16:45
@Hippalectryon Everyone’s getting older!
Aren't you supposed to be retired though ? Still working on saturdays ? :P
@Shinrin-Yoku do you even know if you're sharing the same definition of regularity
@shintuku The mild condition on that page is regualrity
@TedShifrin So by your comment it seems one can eyeball regularity, is the proof easy?
I’ve never heard it called regularity except by Shin.
@TedShifrin Is there any intuition for why the theorem should hold?
16:48
is he just asking why must the derivative exist???
what the heck is a mild condition on the partial derivatives?
@TedShifrin I am sorry, 😞 but which one do you suggest, writing as a function of $h$ ? I am definitely missing something critical !😕
We’re going around in circles. Maybe you should watch some of my lectures. And, instead of being lazy, do the exercise I gave you instead of repeating yourself endlessly.
@Shinrin-Yoku are you asking why must the function be once differentiable??
shinrin what are you even asking
@Franklin $\cos(ax+ah)$, of course.
@SouravGhosh Look like $p$ is $0$ and $q$ is $1$
16:52
I am interested in intuition for why the theorem must hold @TedShifrin
Not so much a formal proof
but what is the condition that is making you unsure
@Shinrin-Yoku
@robjohn have you seen the movie Clouds Are Not Spheres?
@Franklin Ted made a typo in his last reply. You need to find the Taylor expansion of $\cos(ax+hx)$.
What a silly title. Everyone knows clouds are cubes.
@user223626865 no.
I see it's about Mandelbrot.
17:06
@robjohn Yup, basically an interview with him.
@TedShifrin Is Balarka Sen still around by any chance ? I'm curious to know what he's become, he was an interesting guy
@Hippalectryon Still around, still interesting.
Alright, thanks :)
@Shinrin-Yoku here are some implicit functions of $x$ on the infinity loop: https://imgur.com/a/XEYcKMc
here are some implicit functions of $y$ on the infinity loop: https://imgur.com/a/mmgapVc
17:13
not in a ball containing 0,0
Wow, Balarka hasn't been here for >3 months. I thought I'd seen him more recently...
@Shinrin-Yoku no, do you see why that ball would be a problem?
obviously\
cool
How to find the area of the unit disk under the map z---->sin z?
17:15
@Yai0Phah Do you have any intuition for why a for a curve to be a local function it must be regular?
shinrin what do you mean by regular
@Hippalectryon Well into grad school now.
@SouravGhosh I responded to this days ago.
@SouravGhosh using area theorem
@PM2Ring December 10 he was in another room, but I don't know how long it's been since he was here.
@robjohn Nice easter profile pic btw
17:21
How do you obtain the first blue dot in this slide?
@Hippalectryon Thanks. Mean Egghead.
@PM2Ring December 6
@SoumikMukherjee Ohh God. f(z)=sin z is conformal with f(0) =0 and f'(0) =1>0 So area(f(D)) =\pi \sum_{n\ge 1}|a_n|$
@TedShifrin Done! I have answered similar kind of question few days ago. Probably my brain isn't working properly math.stackexchange.com/a/4662173/977780
@SouravGhosh Not right.
@TedShifrin I think you meant something like @SouravGhosh 's answer ?
2
Q: Solve the following differential equation: $\frac{d^2y}{dx^2}+a^2y=\cos ax.$

FranklinSolve the following differential equation: $\frac{d^2y}{dx^2}+a^2y=\cos ax.$ The solution given is as follows: The complementary function is $c_1\cos ax + c_2\sin ax$; the particular integral is $\frac{1}{D^2+a^2}(\cos ax)=\frac{1}{-a^2+a^2}\cos ax$; and thus the method fails. In this case, chan...

The way, he approached makes sense!
@PM2Ring Now, it's clear how they did it....you might want to check out the answer, here, this totally makes sense to me now 😀
17:34
Or do what I suggested and Taylor expand $g(h)=\cos (ax+ah)$. Works fine without addition formulas.
@TedShifrin It should be $\pi \sum_{n\ge 1}n|a_n|^2$.Still not correct?
If a in R is a root of f$(x)=x^3-3x-1$, then does Q(a)/Q contain all roots of f(x)=0?
Unlikely. There’s a discriminant condition for that .
@TedShifrin Discrimination!
@robjohn Didn’t you realize I’ve joined the GOP?
17:38
:O
>_>
<_<
Geometry Oriented Professors
@TedShifrin ohh. What's that?
Isn't there any field theory proof for this fact?
@TedShifrin Oh, the horror!
I'll see what this notion is.
Oh, I see. It has something to do w/ splitting fields.
@TedShifrin Good point. That makes more sense to me. But it looks like Franklin's book expands $\cos(ax+hx)$, so he probably got a bit confused by your suggestion.
@PM2Ring hmm... maybe that's the case...
@Shinrin-Yoku It is a sufficient condition, not necessary. Consider $(y-x)^2=0$.
So what is the intuition behind regularoy @Yai0Phah How would someone come up with that condition?
@PM2Ring I made a typo after responding so many times. Yes, we want to fix $a$, fix $x$ and consider $h$ small.
18:08
@TedShifrin Does not @Yai0Phah Contradict your comment?
@SouravGhosh I made a little plot in Sage.
No. But it is very subtle. I told you to assume the solution set is $y=f(x)$ with $f$ a $C^1$ function and deduce.
@TedShifrin So I should assume that its a theorem thats just true with no intuition/ is that correct?
We're talking about intuition, I thought. Yai0Phah is giving a famous example which shows that $g=0$ is the same as $g^2=0$ which is the same as $g^3=0$, etc.
You are really a pain.
18:13
But is there any intuition for why the theorem you mention holds? I am sorry :( @TedShifrin
@TedShifrin pain perdu?
Sorry, I'm still having my breakfast.
Question: lim_{r\to 0}{ Area f[D(π/4, r) ]}/{Area D} where f(z) =sin z?
@PM2Ring Area(sin(D(0, 1)) =π/2e according to my computation.
18:47
What is $D(\pi/4,r)$?
Have you thought of using MathJax? (ChatJax helps with that)
@robjohn disk centered at $\frac{\pi}{4}$ of radius $r$
19:04
@SoumikMukherjee Okay, then I'd say $\frac12$.
If [K:F]< infty, the Gal (K/F)< infty.
Proof: If [K:F] < infty, then K= F(a_1, ..., a_k) and K automorphisms are determined by action on $a_i$'s.
How do I conclude that these actions can only be finite?
Hi @D.C.theIII!!
0
Q: $[K:F]<\infty$, the Gal $(K/F)< \infty$.

KoroSuppose that $k$ is a field extension of field $f$. I don't understand one step in the proof of $[k:f]<\infty\implies |Gal(k/f)|<\infty.$ Proof: If $[k:f] < \infty$, then $k$ is finitely generated extension of $f$, i.e., $k= f(a_1, ..., a_k)$. I know that $k$ automorphisms are determined by actio...

19:26
@robjohn yes, I am also getting $\frac{1}{2}$
@SouravGhosh No.
koro: each a_i is the root of some polynomial. and by homomorphism-ness, any automorphism has to send a_i to a root of that polynomial. so, at most as many homomorphisms as there are maps from {a_1, ..., a_n} to a finite set. the comment shows how you can get a better bound if you otherwise know that the extension is simple.
@SouravGhosh Hint: Area of $D=\frac{1}{2i}\oint_c\bar{z}dz$
19:47
@leslietownes thanks a lot, Leslie!! I understand it now.
$\ddot\smile$
$a_i$ is the root of some polynomial because $a_i$ is alg. over F because $K$ is finite extension of F and finite is algebraic.
I'll delete the post.
for what it's worth, i don't see something that implies the primitive element theorem in your hypotheses (and we don't need the PET for finiteness, although you might need it for that bound on the degree in the other answer).
and we're not using 'automorphism,' same argument shows finitely many homomorphisms. :)
20:16
$\omega$, the cube root of unity is not contained in $Q(2^{1/3})$.
of course since $\omega$ is not a real number.
But is this the correct proof?
what do you mean, 'the' correct proof? you could spell that out in more detail, but it's fine, as a line of argument.
Or should I say that: If $\omega\in Q(2^{1/3})$, then $Q\subset Q(\omega)\subset Q(2^{1/3})\implies [Q(2^{1/3}):Q]=[Q(2^{1/3}): Q(\omega)][Q(\omega): Q]$
LHS= 3, the second term on RHS is 2 so the first term on RHS becomes non integer, contradiction.
@leslietownes ok, thanks :-). I meant 'a' correct proof.
also fine as a line of argument if you explain why that's a problem (e.g., uh, why is the second term on the RHS 2?)
if someone pickier were here, they'd also maybe just start out by noting that it would help to pin down more explicitly what you mean by "the cube root of unity." something other than 1 that cubes to 1?
oh, that's because $x^2+x+1 $ is irreducible/ Q and $\omega^2+\omega+1=0$.
yeah, that's getting exactly at what you meant by 'the cube root of unity'
20:22
the cube roots of unity: $e^{2\pi ir/3},r=0,1,2 $
is that enough?
or not? because someone might also say what $i$ is.
assuming that C has not been constructed.
well, if you're regarding all of this as a subfield of C, i don't know why you're uncomfortable using the fact that Q(2^(1/3)) is a subfield of R.
you can certainly do all of this without referring to i or complex exponentials. i was wondering if that's what you wanted. but there's no reason you'd have to avoid that.
@leslietownes yeah. The book (Morandi's) uses a similar proof that I wrote above. So I wondered they could have just said $\omega$ is not real.
you can do all of this without reference to R, also, if you want. it's just where you are comfy working.
probably the differences between various books/treatments in this area will come down to how they introduce the subject, for some people's exposition, everything is a subfield of C unless otherwise indicated, in others, you don't assume the existence of more than Q or try to do as much as possible as closely as possible to the field axioms. and maybe some authors shift focus depending on what they are working on.
Let's say T= every element of C that is transcendental over Q. Is T is a subfield of C?
I think this is still an open problem.
Solving this would tell if $e+\pi$ is transcendental or not.
It's surprising that the answer to this is not known.
is T closed under addition?
20:31
oh not necessarily. $\pi- \pi=0$ so not closed.
 
2 hours later…
22:46
$\zeta(-1)=\psi(\infty)$, therefore $-\frac 1 {12}$
Mathematicians HATE THIS one little TRICK for summing divergent sums!
Seriously, though, can any of the (retired) professors and doctors in here please recommend me a good path for learning based on what I wrote before? (read from here: chat.stackexchange.com/transcript/message/63336910#63336910 )
why do you think you need something else than precalc and then calculus
The prime counting function requires a transform of the complex-valued Zeta function.
Should be enough to see my point
Also I'm a programmer in case you forgot
Those don't exactly see a lot of use in our field
people do complex analysis already having done calculus first
but i have no clue about number theory
I'm not really sure what that has to do with what I'm saying.
i don't see why you want to skip precalc/calculus
00:00 - 23:0023:00 - 00:00

« first day (4630 days earlier)      last day (685 days later) »