« first day (2327 days earlier)      last day (2993 days later) » 

00:00
mmh, is a proof about nonexisting counterexamples a proof about the statement?
No, because such a proof would be flawed. There are linear functions in the above sense which are not continuous anywhere.
@mick let's just say it this way: a conjecture lasting 100 years in the industrial age is not easily provable
@BalarkaSen So, this statement is not always true?
@MaryStar What you asked is true and well.
I'm replying to Null. What he claims is false.
Ah ok.
00:05
For your question, you know continuity near $0$. Pick an arbitrary point $a$ - if you want to prove it's continuous near $a$, what you want to do is to "translate" the function to $0$.
I recommend writing down the definition of continuity on this one.
So, for an arbitrary $a$ we want to show that $\lim_{x\rightarrow a}f(x)=f(a)$, or not?
What do you mean by ""translate" the function to $0$"? Do we have to set $y=x-a$ ?
Yep, do that.
Setting $y=x-a\Rightarrow x=y+a$ we have that when $x\rightarrow a$ then $y\rightarrow 0$.
So, $\lim_{x\rightarrow a) f(x)=\lim_{y\rightarrow 0}f(y+a)=\lim_{y\rightarrow 0}(f(y)+f(a))=
So, $\lim_{x\rightarrow a) f(x)=\lim_{y\rightarrow 0}f(y+a)=\lim_{y\rightarrow 0}f(y)+f(a)=0+f(a)=f(a)$.

Is everything correct? Could I improve something?
I can't read that. You'd come upon $\lim_{y \to 0} f(y)= 0 = f(0)$, which you know is true by continuity near $0$.
Rigorously you'd do all this through epsilons and deltas, but it's ok in principle.
no, that's continuity at $0$, not near.
(Not that I've read the stuff before that.)
00:14
Sure.
I'm using bad terminology.
Sorry you're still awake ...
test $lim:x\to\infty$
This time its my biological clock, not the thing on my face
test2: $\lim_{x\to\infty}$
and then there's '\lim\limits_{x\to\infty}', @Null.
00:17
I always forget that
@TedShifrin i know, just teste the shorter one ;)
I guess I finally understand the convergence theorem of spectral sequences
@BalarkaSen Ah ok. Thank you!!
actually I think it would be very helpfull if this where a macro in my tex :d
if i have velocity vector of 140 degrees at 1.1ms^1 why would the calculation be |v|sin(theta) i + |v|cos(theta) j

I've only ever known it to be cos for i and sin for j not the other way around =/
00:21
like \lx, done
From where is the angle measured, @WDUK?
north
That's why.
Usually, you measure from the positive $x$-axis?
You might need to watch out for signs, too.
by positive axis you mean 90 degrees from north
i dont quite understand how to visualise the logic for it
Yeah, don't you measure clockwise from north, as opposed to counterclockwise, too?
Draw pictures!
00:23
my other two vectors did not require the cos and sin to swapped
i did draw a picture i calcualted it by 180 - 140 = 40 degrees for theta
North is $j$, east is $i$. Oh, so no minus sign.
No, no, you can't just substitute like that.
$\theta$ is measured from the positive $y$-axis, going clockwise.
yeh i am doing it clockwise but i had to get the acute angle for my right angled triangle so i did subtraction that way
all my other vectors i calculated using cos for i and sin for j but this one is swapped which i am not getting why
But you've confused yourself on the meaning of $\theta$ when you do that. The trig functions of the reference angle need to be adjusted with different signs depending on the quadrant.
I've explained it's because $\theta=0$ corresponds to the positive $y$-axis, not the positive $x$-axis.
Draw a triangle with $\theta=40º$ and figure out the components of the vectors.
well i calculated it to be 0.8i - 0.7j which would be south east in direction
Which problem are you talking about?
00:30
well cosine relates to the x-axis right , so my thinking here is if i relates to east then cosine should be used for i aswell at least thats where i am getting confused
I'm going to tell you to reread all my comments. I'm not going to say the same things a fourth time.
i did read it just not getting it at all
00:44
never mind ill ask on a forum ill understand it eventually :P
@TedShifrin im trying to think of what tools i have to show collinearity
besides desargues
01:13
hi @ZachHauk
oh hey @Adeek
what's up?
nothing much finished all exams
just have to finish this project tomorrow
and mark final exams
then I am ready for break
I was proctoring on calculus 1 students today
very boring
heh
my sister's in calc right now, and so i ask her questions relating to her current unit
(she doesn't always know them!)
don't be a pain in the butt, @Zach :)
lol
01:18
@TedShifrin but she thinks limits are just the result of "plugging in"!
hey @TedShifrin
(a common belief among students at this level)
well, Zach, all functions are continuous ... high school calculus is not a rigorous course
@TedShifrin hi Ted, yes a bit
I think they should teach analysis in high school.
01:19
who is they?
I rue the day that AP calculus was born.
@TedShifrin AP calculus should be analysis
@Zach: Surely, she won't plug in for $\lim\limits_{x\to 1}\dfrac{x^2-1}{x-1}$.
No, Karim, don't be a twit.
mmh, why is a proof about the nonexistence of a counterexample not a proof about the statement? I don't mean that the counterexample is incomputable or inexpressable, I literally mean it is nonexistent.
of course, if you plugged in there, her math teacher would send her to
L'hospital
01:21
haha
no, L'Hôpital comes way later.
i kid
I want l'Hôpital to be banned when it's the definition of the derivative. Totally banned.
Mathematics is so hard!
Back to your conic, @Zach.
01:22
@TedShifrin I agree, even tho I'd be happy to use it
ok, ok ;)
@TedShifrin by the way, i own rudin's; granted, i haven't read it yet, but nonetheless i own it
anyways
conic time
I don't think you should start with rudin @ZachHauk
I can't love a book that has literally zero pictures.
try maybe abbot
Spivak is the obvious answer.
01:24
nah abbot is better
i own spivak as well
Rudin is like starting to read with shakespeare
oh wait, im not sure which spivak book i have
let me check
I mean Calculus.
abbot will give you intuition to many stuff in analysis.
01:25
For the rigorous multivariable course, I of course prefer a book by an idiot.
@TedShifrin your book is wonderful for multivariable calculus totally beats rudin and other books
@TedShifrin ah, sorry i dont own that.
What do you have, Zach? Calculus on Manifolds? Forget it :P
i thought i had a spivak book
i guess i dont
@Ted naw :P
Back to conics.
01:27
i need a giant poster of desargues on my wall just so i dont have to keep looking back at it when i forget it :P
But he's not the only theorem in the world.
i manage to remember Pappus quite fondly
because its cool :P
Apparently not fondly enough.
I like conics
quadratic forms $k^3\to k$ are in bijective correspondence with symmetric bilinear forms on $k^3$
as long as k is characteristic $\ne 2$
LOL ... I was just typing about char 2.
01:39
hehe
Is it true that the hypervolume of an n-dimensional unit hypersphere goes to zero when n increases?
I've heard that it is true for hyperspheres with unit diameter but I don't know if it is true for hyperspheres with unit radius
@AliCaglayan why not?
If it's true for diameter, it's more true for radius, @DHMO.
@TedShifrin I don't think so...
01:41
You get an extra factor of $(1/2)^{n-1}$, which, I believe, goes to $0$.
You're saying that if (a_n) goes to 0, then (2^n a_n) also goes to 0
You are upside-down, aren't you?
unit diameter means radius 0.5
Oh, sorry, you're right. Sorry.
Actually I changed my mind
01:42
it's fine
Regardless, I know the result in terms of radius.
@TedShifrin you can use desargues for exercise 11! :)
@TedShifrin do you have any proof/source?
How is the conic then relevant, @Zach?
oh wait
let me check the hypotheses of desargues
01:43
Duh :P
The volume of the unit ball in $n$-space is well-known, @DHMO.
In geometry, a ball is a region in space comprising all points within a fixed distance from a fixed point. An n-ball is a ball in n-dimensional Euclidean space. The volume of an n-ball is an important constant that occurs in formulas throughout mathematics. == Formulas == === The volume === The n-dimensional volume of a Euclidean ball of radius R in n-dimensional Euclidean space is: V n ( R ) = π ...
The "hypervolume" of the unit sphere is given by differentiating that times $r^n$ at $r=1$.
It really doesn't help to stick that in here without just linking.
@TedShifrin whaaaat
well that statement contradicts with my link
How so?
Well, because your formula always gives an integer?
01:46
No, it does not.
what does "that" refer to?
the volume of the unit ball in dimension $n$ ... as in your title?
wait, I meant the ball not the sphere
I'm in the stage of "what's the difference between a sphere and a ball"
A sphere is the boundary of a ball
The $(n-1)$-sphere is the boundary of the $n$-ball.
01:48
I know
You know? Then why did you just say you didn't?
I mean I still mix them up
Now, from the link given, it looks like that a factor of 2^n is not enough to counteract the factorial, meaning that it would still tend to 0
Right.
unless youre in $L_\infty$
Huh?
We are working with the usual $2$-norm on $\Bbb R^n$ as $n\to\infty$. You're confusing things.
01:50
the Lp norm with p -> infty
what is the name for the empty letter? ""<- this is what i mean
@Null what?
null character?
yeah something like that
hi tern!
What does it mean that R^n has room for exp(n) almost-orthogonal vectors?
01:53
I have no idea.
@ZachHauk mmh, I'm not quite sure if "null-character" is the right word. I mean with what charakter do we replace when we replace by nothing?
a space?
@Null we are not replacing it by any character
it is not one character
@Null we are replacing by the null character
or, appending to an array of characters
113
A: Colloquial catchy statements encoding serious mathematics

Armin Straub A drunk man will find his way home, but a drunk bird may get lost forever. This encodes the fact that a 2-dimensional random walk is recurrent (appropriately defined for either the discrete or continuous case) whereas in higher dimensions random walks are not. More details can be found for ...

@Null it also does not make sense to ask "what character is 'ab'?"
it is not one character; it is two characters
01:57
@DHMO I see your argument ;)
also lol at the link :D
> All primes are odd except 2, which is the oddest of all.
3
@TedShifrin would you recommend a good text on representation theory ? I am taking reading course on it next semester will use the familiar book.
Fulton and Harris
That's a good book, I think.
I am no expert on this material.
okay cool.
"If there are fewer pigeons than holes, there must be a pigeon with at least two holes in it." — Akiva Weinberger Sep 2 '15 at 20:58
02:06
I guess I can stick to it I found it also not bad
Jim Humphreys has various books, too.
James and Lieback
okay I will check them out thanks.
it is always good to have more than one resource.
Sometimes too many resources lead to more waste of time and confusion. It's a balance act.
yeah I think maximum like 2 resources on the subject is good.
02:07
Proof: The internet
Yeah
I'm glad there was no internet when I was a student, @Ali :)
your lucky @TedShifrin I found that I waste a lot of time
Of course, it's changed the whole complexion of research.
I know you waste a lot of time, Karim.
For good and for bad in all fields
02:08
on the internet. Next semester when I am studying I am planning to shut down the internet completely.
The best source is ...? feel free to complete yourselves
I do @TedShifrin
Good, Karim ... so next semester we shan't see you here!? :P
yeah I will only come on saturday to ask questions.
I mean grad school isn't like undergrad I need to portion my time properly.
You need to be more self-reliant and ...
02:10
yeah
I agree
finish my sentences?
smacks @Ali
will this be on the exam?
Everything will be on your exam.
do we have to go to all the classes
is the homework going to be checked?
02:13
My teacher multiplies triangles INSTEAD of square matrices ! Anyone familiar with that ?
and what @TedShifrin :P ?
@TedShifrin can you give me a hint on exercise 11? am i supposed to use cross ratios?
@AliCaglayan Yes. Not doing it is punishable by death.
btw @TedShifrin I downloaded EisenBud commutative algebra.
> if it can be proved that it can't be proved that two plus two is five, then it can be proved that two plus two is five.
02:14
It looks very nice
What does this mean?
@DHMO if we can show that we cannot prove $2+2=5$ (but we can prove it, obviously), then we can conclude $2+2=5$
@ZachHauk why?
@ZachHauk no
@Ramanujan ... what?
i was just explaining what the quote meant
02:16
IIRC you can never show in a given language that a statement is unprovable
need to pass to a stronger language
@MikeMiller but what does the second part mean?
what second part?
@Ramanujan yes.
it can be proved that $2+2 \ne 5$
it cannot be proved that $2+2=5$
and it cannot be proved that it cannot be proved that $2+2=5$
So you men 2+2=5?
02:18
ah by unprovable here they don't mean independent
but w/e
@MikeMiller I'm just having trouble with the second part of the quote, which states that if it can be proved that it cannot be proved that 2+2=5, then it can be proved that 2+2=5
mike the sassy assasin
it's not inspiring enough for me to figure out tbh
and I hope that it is not just a vacuous truth
yay i got it :)
using compass and straightedge geometry, is it possible to draw a line parallel to a certain line intersecting a given point?
02:35
392
Q: Polynomial bijection from $\mathbb Q\times\mathbb Q$ to $\mathbb Q$?

Z.H.Is there any polynomial $f(x,y)\in{\mathbb Q}[x,y]{}\ $ such that $f:\mathbb{Q}\times\mathbb{Q} \rightarrow\mathbb{Q}$ is a bijection?

Can anyone with high reps tell me what the deleted answers contain?
nobody here has sufficient rep on mathoverflow, but i can tell you they contain incorrect proofs
No, @Zach, you're supposed to use a relevant theorem.
@TedShifrin i used desargues dual
@MikeMiller I can tell also
I just like to see the failed attempts
@Zach: I find that hard to believe.
02:36
shall i share my solution? it uses that "line parallel to another line given a point it intersects" though
@TedShifrin this is for a different exercise
Exercise 16 to be specific
Oh, 16 is cool and tricky.
would you like me to share? :)
that article by bjorn poonen in the comments of that post is really cute
But I'm not impressed that you used that theorem, since the problem told you to, @Zach. :P
:P
well, its one of the only theorems that discuss concurrency
so it wouldn't be that much harder to figure out without the hint
02:39
@ZachHauk Are you in the US?
OK, you can send me that one, @Zach.
oh i also got an answer to 19 as well
@MikeMiller yes
Can I ask whereabouts?
Northern New Jersey. Bergen County
@TedShifrin ok, ill go ahead and type it up
nice area
02:40
OK, @Zach. And you should get 11. Rescan the chapter.
does it have to do with cross-ratios?
@MikeMiller indeed.
if you don't mind me asking, how about yourself?
If you're going to try to reinvent the wheel, yes, @Zach.
@ZachHauk Mine's in my profile so easier to find. I'm a grad student at UCLA. California native.
Best state in the union.
02:55
@TedShifrin would you like me to make a tikz diagram?
Do what you need to to make a good explanation/solution, @Zach :)
ok :P
03:07
absolute convergence is a stronger property, and convergence alone doesn't implies it. Simply example $\sum\frac{(-1)^n}{n}$.
^completly unrelated to what's the topic
03:20
@TedShifrin OK, i sent you an email consisting of a drawing in my notebook and a typed up proof
03:43
@Null $$-\sum_{n=1}^\infty\frac1{2n(2n-1)}$$
more unrelated
hello
Hi nerds
3
@robjohn mmh, why the minus before it?
03:47
@Null Because it is negative as written. If you start at $n=1$
Who you calling a "nerd"? @Ramanujan
@Pissedofflayman to those who are
good answer ;-)
@Ramanujan I wonder who the real Ramanujan would consider a "nerd"?
@Ramanujan what is your full name?
well my exercise boils then down to: "does there exist a field where $(0+\varepsilon)^2>(0+\varepsilon)$ for some $1>\varepsilon>0$."

« first day (2327 days earlier)      last day (2993 days later) »