#serious-discussion
1 messages · Page 552 of 1

eulers number?
Oh i actually didn't consider that
im too math pilled to consider e as something else
HelLO!
Hi ryc!
hey ryc
Hi
How's life in Italy?
Uhhh
It was a very good day
And now I am slightly drunk
In a not super fun way
but fine enough
My plan was to make it fun
but
my phone will not connect to the internet
Being drunk is fun af
L
But only if you're around other drunk people
Being drunk alone sucks balls
No that's perfectly fun too imo
I realized this sadly last time i was high
Like
I like being drunk so much better
Being on discord doesn't count as being alone
Gah fair
I've never been drunk alone
actually i do that while not drunk too if i'm totally alone
I would probably do the same
Yeah i do that too from time to time
It differs from person to person but oftentimes 2 drinks is the minimum to start to feel a light buzz and 3 drinks is the minimum to start feeling a proper kind of drunkenness
For me, 4 and above is where I am properly drunk
But it depends on the kind of drink too
Like 70-75 kg? I don't know exactly atm
Yeah it's not that low tbh
Less money for you to spend
I weigh 80 kg so i need a little too to feel it
I see
I weigh 92 kg so maybe that’s why 1 drink is not enough.
@proud olive explain 
red e?
bingo!

It's fun to try different drinks, some are really interesting
My recommendation is that if you have the opportunity to go to a wedding with an open bar, take advantage of it to try some interesting cocktails.
Because otherwise you have to spend like 12 bucks on something that could be gross
I found out at a wedding how good white russians and tequila sunrises are
and that i don't like whiskey sours
Oh I mean as a mod my recommendation is to not drink and take care of your body! Especially if you are under the legal drinking age in your country. Or something.
There was also an open raw bar 
What's a raw bar?
Damn jelly
0/10 mod
What I did my job perfectly!
Asking the real questions
Uhhhh...
if nami wanted it he would have added it
Maybe nami wants it but wants other people to add it?
bruh nami can't do everything here
ryc does stuff
gotta let other ppl feel accomplished smtimes too
gotta let other ppl feel accomplished smtimes too
I sure could use a vacation from these freaks
If nami were nice he wouldn't have nominated me for worst user!
no i dont
sussysullybutred on the other hand
banger emote
Sussy sully but red is at best marginally better
he's just being honest nami wouldn't lie to you
liar
true
i would happily lie to ryc
Then how am I the worse user!?
you dont get to see my pfp??
wow
Well I do on mobile
hm
Is anyone good at financial math problems? i need help

HI RYC
Random math time.
A matrix in GL(n,R) generates a Bravais lattice via the matrix acting on Z^n, and is thus effectively isomorphic to Z^n. However, the subgroup SL(n,Z) acting on Z^n is automorphic, thus for some matrix M in GL(n,R) and a matrix N in SL(n,Z), both M and MN generate the same lattice
not finished you dippy
Now, we can essentially say that the (modified to be a true lattice) dual lattice for a lattice generated by M is the lattice generated by M^-T
(the dot product between vectors the lattices is always an integer)
So lets assume we want a lattice that is self dual, so the lattice generated by M and M^-T is the same
But M and M^-T are not the same matrices, but because they generate the same lattice and due to the properties mentioned above, we can say M = M^-T * N for some N in SL(n,Z)
and from there we can say that if M generates a dual-lattice-invariant lattice, then M^T * M is an element of SL(n,Z) (and is also symmetric)
Lets say that a matrix M is crystalline if M^T * M is in SL(n,Z)
Well, let M = KA, where K is an orthogonal matrix (KK^T = I), well then M^T * M = (A^T * K^T) * K * A = A^T * A, so A is also crystalline. The set of crystalline matrices is invariant under the action of the Orthogonal matrix group S(n,R)
[The set of crystalline matrices does lack a group structure btw]
Lets add an equivalence relation to the set of crystalline matrices such that:
A ~ B if there exists a matrix C in O(n,R) OR a matrix D in SL(n,Z) such that A = CBD.
[essentially any hyper-volume preserving transformation, along with the existing same-lattice generating matrix]
Now, lets assume n ≤ 8
It's a well known fact that there is only one unimodular lattice in R^n (n < 8) up to isomorphism, the integer lattice Z^n. So, essentially, in n < 8, every crystalline matrix should be equivalent.
But in n = 8, there is also the E8 lattice. Thus, there should be TWO equivalence classes here.
I wonder if there is an actual way to prove this without going through the nuttiness of proving it using properties of lattices unrelated to matricies [including norm vectors and shit]
All of this I kinda bullshitted to myself during work today so it's almost entirely unfounded
The lattice generated by this matrix is:
{v in Z^n union (Z + 1/2)^n : sum of components is even}
but is this lattice like, isomorphic to Z^n
THis is wrong, ignore it
Also changing the name "crystalline" to "integral"
Looking up some terminology, essentially M^T * M is the Gram Matrix



Which finite groups are isomorphic to the automorphism groups of symmetric graphs?
Well all $S_n$ are isomorphic to $Aut(S_n)$ except for n = 2 and n = 6.
$S_n$ is a complete group (except for the two mentioned cases earlier). And every complete group is isomorphic to its automorphism group $Aut(S_n)$ because it has a trivial centre and trivial outer automorphism group.
In case of n = 2, the center of $S_n$ is non-trivial, and in case of n = 6, $S_n$ has outer automorphism group.
Except those, all other Symmetric groups can be the example here
Pencil/Idris
Hi det 
hewwo 
valley........
i thought u were based........
Yeah
I have some slides on this
Ooo
Its also explained in Joseph J. Rotman.
An Introduction to the Theory of Groups.
I see
Hello everyone !
Now the school year is over so I would like to focus on maths and especially on topology and differential geometry
Please, do you know which basis I need to study these topics ? Thanks !
[topology basis joke]
Not a lot of things sadly
I mean I learn quite fast and I have a good math "intuition"
have you ever taken a proof-heavy course
First year of university lets say
If I understand what you mean, yes
if you've done proofs then you don't need much else for (point-set) topology
Ow really ? Nice !
hahhahahah
for differential geometry tho
oh for diff geo yes
u do need analysis and linear algebra
diff geo is an absolute beast compared to intro topology
and a lil topology wont hurt as well
Okay ! But until which point ?
Thanks !
a first course in real analysis
I recommend topology without tears for a good intro to topology
it will help if u have some idea about continuity, convergence and all
i recc Intro to topological manifolds
it has such pretty diagrams
that wasn't a joke
Ow okay ! Such a funny name!
yeah cause it's freely available online
Yes I think it is fine for this !
true
That s a very good point
Thanks for all your good advices ! I am quite excited !
good luck out there
feel free to ask questions in advanced channels
check #get-advanced-access
Advanced ? University channels arent sufficient ?
Thank you !
and at times linear algebra stuff can just be group theory in disguise
That is quite interesting ! In France it is studied quite early
oh u are from france?
That's funny !
france is kinda cracked at maths
Yes !
not fair to compare it to other countries 
Cracked...is it good or bad ? lol
it means very good
Okay !
topology isn't, I've yet to encounter a uni that does not do analysis in first year
u are in the uk wew
go see american unis
thank u
In middle and highschool we are really not good at all
But our universities have good level !
its not like u cant do analysis in first year
every uni i want to go to also does analysis first year
(im not in the us)
who cares about america then
"often isn't" = "is done everywhere except america" now
yes.
But what do they do first in America ? Analysis for me is really the basis
they do calculus
from single variable to multivariable
and diff eqs and liner algebra
something like that
y'all do analysis before calculus??
(i should clarify not everyone has to do this, but a lot of math majors do)
no we just do calc in hs
calc is hs stuff
yea
and especially not calc 2 or 3
not many people get to calc bc
Okay I will appear quite idiot but what difference do you do between analysis and calculus ? For me analysis implies differentials and calculus...integrals maybe ?
I don't think most schools offer calc 2 other than the AP class and not a ton of people take that
In France I think we quite mix both
calculus is basically "what is the derivative of this function" and analysis is "ok let's take a closer look at what's going on"
Analysis usually means calculus but proof based
Okaaay !
it doesnt concern with calculating integrals or derivatives mostly
Thank you ! So yes in France it seems we do not separate both
damn
Or I didnt notice it which is also a possibility !
lol
Analysis is gauging into the rigorous study of functions
Two semesters of calculus is vaguely showing you what functions can do but doesn’t explain them really
Which is dumb
Ow calculus could be what we call "calcul infinitésimal" but yes, for us it's analysis
I see
Oh here’s some more basic algebra, now just arbitrarily use these methods you’ll take for granted with your basic algebra skills
So analysis is like...higher level calculus ?
It’s calculus the way it should be taught to people who are interested in mathematics
I don’t really see the benefit of taking a watered down class that doesn’t really teach you anything.
You’ll vaguely understand what a slope and an area are still after you wasted a semester or two if you ask me 🤷♂️
It’s like crazy that students ask if they should take linear algebra or multivariable calc. I tell them, why the hell is linear algebra an option here?
You SHOULD have that linear algebra foundation. Bloody hell
it's an absolute methhead arrangement that people can do multivar before lin alg
I didnt realize how important it was !
Well we deal with dimensions all the time and linear algebra is how we approach the abstraction of dimensions early on
its ok in my analysis class we saw a little bit of Rn calc but whatever
So a customer asked me if I knew anything about noneuclidean geometry and I told him I very vaguely understand Riemann Surfaces 😂
If vaguely understand counts as understanding at all.
But I sort of visually understand some things
Yes I understand what you mean ! It is quite funny and...maybe satisfying !
That stuff is probably more complicated than some QM stuff out there
You can have coordinate planes superimpose eachother like origami
That seems very beautiful
It breaks your brain at some point too I think 😂
Being human is crazy
Didn't I tell you to change the curriculum there? Smh
Can you stir fry pork loin? Almost every recipe i see calls for tenderloin
(If not like shoulder or butt)
I don’t see how that disqualifies S₆.
Actually, what about K₆?
Heya
For someone who knows how to play sudoku, why would 4 be a mistake in the highlighted box. I don't comletely understand the rules.
Because one square should not have two of the same number, there are two 4s in that top left square.
Also, each row and column shouldn't have 2 of the same number.
No problem!
kangaroo rat
a sudoku player
i didnt know people still play sudoku i thought i was the only one
Last time I said that on this server I got crucified
i would never crucify shyshy
Loll
i mean, its not exactly correct either
but it explains what analysis is pretty simply
I can also just analysis is inequality spam
and no one can say shit to me for saying that
~~Basically there exists an outer automorphism of S6 that doesn't preserve the structure of S6.
Aut(S6) is a semidirect product of S6 and C2. Hence, its order is 6! . 2.
By Lagrange's Theorem, |Out(S6)| = |Aut(S6)|/|Inn(S6)|
And Inn(S6) is isomorphic to S6, since the center of S6 is trivial (because any symmetric group of order ≥3 is non-abelian), and we know isomorphism preserves orders. Hence |Inn(S6)| = 6!
Hence |Out(S6)| = 2.6!/6! = 2, which implies that it is non-trivial, hence not complete, and implies Aut(S6) is not isomorphic to S6~~
Another great argument is that isomorphism is one-to-one and onto, and hence must preserve orders, but |Aut(S6)| = 2.6! ≠ 6! = |S6|, and hence no isomorphism exists between them.
- How is S₆’s exceptional outer automorphism relevant?
- What about K₆?
For 2, I'm not really sure about that
Hmmm okay I changed my mind, I believe the second argument is more relevant here.
Because every complete group G has Aut(G) iso to G, but converse isn't always true (Aut(D8) is iso to D8 but D8 isn't complete, a counterexample)
So for 1, that exceptional outer automorphism is not really relevant here
@hollow ginkgo you may use my second argument why Aut(S6) is not iso to S6
Let V be a vector space of (field F acting on abelian group A)
Let B be the set of all bases of A, (all with the same cardinality, the dimension), and there exists a matrix that maps each base to another base. Does this mean that the general linear group of the vector space acts transitively on the set of bases?
yes
Didn't you already ask this
Im pretty sure you did
Anyways
Linear maps that send a basis to a basis are isomorphisms
And a linear map is defined by its action on some basis
I asked if it was free. Afaik it isn't
I was going to use it at some point in the question I'm gonna try to answer in #linear-algebra
I wasn't quite sure where to put this, so I decided to try here. Suppose we wanted to find maxima or minima on a function $f:\mathbb{R}^2 \rightarrow \mathbb{R}$. Does the following procedure relate in some way to "partial derivatives"? I understand they can be used to do the same thing.
person2709505
We'll use $f(x,y)=\cos xy$ as an example. First, consider the general "cross section" of $f$, which contains the point $(x,y)$, runs in the direction $\theta$, and is parametrised by $t$. We abuse notation a bit and call the cross section $(f \circ L)(t)$, where $L(t)=(x,y)+t(\cos\theta, \sin\theta)$. We find $f \circ L$ to be $\cos(xy+t\sin\theta + yt\cos\theta + t^2 \cos\theta\sin\theta)$. Differentiating with respect to $t$, we get that $\frac{d(f\circ L)}{dt} = -\sin(xy + xt\sin\theta + yt\cos\theta + t^2 \cos\theta\sin\theta)(x\sin\theta + y\cos\theta + 2t\cos\theta\sin\theta)$. At $(x,y)$, we have that $t=0$, so we get the expression $-\sin(xy)(x\sin\theta + y\cos\theta)$. This expression, which we'll use more abuse of notation to call $m(\theta)$, gives the slope in the direction $\theta$ at the point $(x,y)$. At a max or min, we will have that $m(\theta)=0$ for all $\theta$. Clearly, any $x,y$ satisfying $xy=2n \pi$ or $xy=\pi + 2n \pi$ $n \in \mathbb{Z}$ meet our criteria. We therefore conclude that f must have either local maxima or local minima at $xy$ coterminal to $-1$ or $1$.
person2709505
Sorry that should be "coterminal to 0 or pi" in the last sentence
And how is any of that relevant?
sotrue
Because we are trying to prove that isomorphism exists between Sn and Aut(Sn)
But this is not true for n=6.
S6 is a semidirect product of S6 and C2. Hence its order is |S6|.|C2| = 6! . 2
But order of Sn is 6!
If we map Sn to Aut(Sn), we see that the order is not preserved, and hence no isomorphism exists when n=6
Why?
Why do we care about S₆ not being isomorphic to Aut(S₆)?
why wouldn't we care about that?
What does it have to do with graph automorphisms?
dunno
are we only allowed to care about things that have something to do with graph automorphisms?
it does appear to be relevant, though
in a complicated way that i can't claim i understand
That’s what my original question was about…
yeah, but you don't get to decied what the discussion covers
once the discussion starts, it's the property of the community and can go wherever the community desires
What was being discussed?
and in any case the exceptional case of S_6 is completely relevant to graph automorphisms, based on my very brief reading
How so?
In group theory, a branch of mathematics, the automorphisms and outer automorphisms of the symmetric groups and alternating groups are both standard examples of these automorphisms, and objects of study in their own right, particularly the exceptional outer automorphism of S6, the symmetric group on 6 elements.
there's a section on there that applies this to graph automorphisms
i think, not sure, still seems cogent
when reading textbooks, at what pace do yall read?
vastly depends on the book?
i mean it ofc depends on the section length and difficulty but like generally
on a good day maybe 10 pages
depends
do you spread out exercises and just reading evenly so like you read a bit then do some exercises or do yall have things like "reading days" then just "exercise days"
i go proof then skim exercises
cause some of the exercises only test for "knowledge"
particularly directly after the proof
when i find a hard problem i do it
oh, exercises take longer
i havent considered this, because i dont do them anymore
Hello
i think its good to do exercises while reading the text but its hard to judge whether you already have the tools
the proof is left.
i did say on a good day
Lol this is so true
That the pace of reading a math book isn't really fast
If I reach the exercise part of a chapter it depends how long I need to spend. Its been quite a while since I've started on Spivak Chapter 2's exercises and I still have like 2 left which I want to try as they seem quite interesting
I have like 40+ pages of latex for Chapter 2's exercises alone lol
Perhaps even 50 
(for just like 10+ qns, maybe around 13-ish)
Yeah I've been able to speed through lower level math books but once you get to the real stuff things grind to a halt. Content becomes so much more densely packed.
Really hard books I’m working thru my strategy is to try to do half the exercises per chapter and try not to be too bogged down otherwise
Cuz if you can do about half the exercises, you should know enough to move on
Like especially if the exercises seem challenging to you, you probably should do them
it's been a while since i looked at a textbook, other than to look up something from a previously used book to refresh my memory
you swear at @polar panther until it complies
be better 😎

Active roles are weighted by channel value
stop sullying me
(╯°□°)╯︵ ┻━┻
I sully YOU!


adding active/very active roles was a mistake
agree
dami

- too lazy to change it
you can if u want
Should be prime
$\frac{\mathbb{Z}}{91\mathbb{Z}}$
13 x 7 = 91
.> is that notation correct
it was just a sully in the public discussion channel, discord
oh right no mathbb
lmfao

Z/91
writing quotient spaces as actual fractions is sometimes done. the notation is technically G/H, but you will confuse no one
@sleek wing
G/H coming from the fact that it consists of left cosets and the G is on the left of H
but nobody cares
No
I am having an espresso at 11 pm
They didnt give me 3 drinks
that's the worst thing you've ever done
you aslo failed to take reponsibility
And no one can tell me what to do. Especially not you darq
What's wrong with espresso at 11 pm
what abt shin
Uhhh sleep??
what abt nami
what abt dami
Hes just usually right when i dont understand what im talking about
It's like a hot chocolate or tea before bed nice warm drink to relax
kenshin
what's wrong with you
This is why warm milk exists
kenshin has a crippling caffeine addiction ofc
But yeah hot choccy is great for before bed
.
dami being scary is just a meme
Valley you started with the most embarrassing one and you're moving down the embarrassment chain
get a grip
Admittedly I have to have a large cold brew or dopio to really feel caffeine
Dami could not possibly get me to do something
we aint even at the bottom yet boi
what about slurp
Slurp also is incapable of getting me to do things
what about alex
Unfortunately you are moving further and further away from people who historically have had any sway
valley enumerating everyone in the server
what about skid

no answer means yes btw
What if we had a real conversation
what's that
lmao
jkjk what do you want to talk abt mr bad boy
So what is everyone studying these days 
Well my espresso just got finished
So now I will head home
And not have any wifi during my walk
haskell, mostly
Or internet access whatsoever
a tragedy
Ew
gj
haskell is ok, come on
!
I am excited for tomorrow morning where I will learn about how someone is studying gravitational waves from LIGO data
I have heard many bad things about haskell though admittedly I don't know any code

where can i get help for addmath
for what?
trig
anyone I know that knows even a tiny amount of haskell seems like they want to suck its dick
this is news for me
Really my professor said he always got really annoyed with it
is he a mathematician?
Yes
He does like modeling and computation in mathematical biology
But he learned a ton of languages
Untrue.
What does ryc need to do?
ban
WEW
slurp
WEW
how's it hanging homedawg
very well 
L
most of the time
I'm rich
LMFAO
LOL
Mmm
IT'S AN ART?!
Aha
you wouldn't understand it slurp
So is stripping my friend
honestly slurp ur kinda insensitive
speaking from personal experience?
Some might even say stripping is a more complicated art
Maybe, maybe not
I strip when I go in the shower (except my swimming goggles and spongebob swimming trunks)
based
sometimes I even bring in one of those inflatable ring things
in the shower? not much
Pole dancing
it all gets wet and mushy by the time I pocket it in my spongebob swimming trunks
I'm not from poland I haven't danced with anyone from there
,w 640 Moroccan dirham to NIS
Mmm respectable!
As a flex
tbf the currency symbol is swag
Obviously I memorized the exchange rate from NIS to euro
,w 640 Moroccan dirham to USD
is that an aleph i see
No
no
₪
It’s a stylized שח
hmm
Which stands for new shekel
looks like an aleph
Ok I guess
Your mom does
yo mama lookin like an om nee cron
Wow
What is going on
what a donut
don't eat it
Is there a differential equations section for this discord
so many pings
starting to learn linear algebra!
but
the author uses a horrible notation for dot products
$u^Tv$
gmod
these aren't transposed matrices
yes they are
this notation totally makes sense if you associated 1x1 matrices with scalars
i'd prefer a \cdot but there's merit in how they present it
My fault I meant R^(n * 1).
Ryc go clean your room
oh
well I gave up on that book after I realized it was geared toward engineers
"applied linear algebra"
no ty
that explains the conflation of scalars and 1x1 matrices
if the only spaces you work in are F^n for a field F, there's no danger in doing so.
makes sense
i see that a lot, both in texts for engineers and also in content for machine learning
vectors are nx1 (or sometimes 1xn) matrices and scalars are 1-vectors and thus 1x1 matrices
conceptually it is perhaps a bit weird that associating vectors in the definition-of-a-vector-space sense with "row/column vectors" (i.e. 1xn or nx1 matrices) is totally fine, but associating 1-vectors with scalars can be problematic
like the distinction is easy enough if you know how vector spaces are defined formally
"matrices are built from vectors, whereas scalars are just a completely separate object"
but i can see why it's weird for learners
scalars "aren't really" 1-entry vectors, but vectors ARE really 1-column matrices?
¯_(ツ)_/¯
yeah
Don't do 1 vectors have seemingly same properties as scalars
not always.. for example, if $v$ is an $n \times 1$ vector and $c$ is a scalar, then you always have $cv = vc$. On the other hand, if $w$ is another $n \times 1$ vector, then $w^T w$ is a $1 \times 1$ matrix, but but you don't have $(w^T w)v = v(w^T w)$; the multiplication on the left hand side is invalid as the dimensions are incompatible
OurBelovedBungo


Okay that's actually a good argument
in practice everyone identifies 1 x 1 matrices and scalars anyways
Yea
so while this is technically right, who cares lol

(maybe "who cares lol" is a bit harsh given the original question, it is still a good point)
generally true, but the above shows that you can run into trouble if you aren't paying attention, and it's not just a silly hypothetical example, this kind of stuff shows up sometimes when doing extended calculations involving matrices and vectors (things like deriving kalman filter equations and such)
"aha, that's a scalar, so I can slide it over here to simplify..."
"oops, wtf, now the dimensions don't make sense"
So depending on context of calculations, you can "interchange" the concepts of 1 vectors and scalars
But in general, they're not the same

yeah, fair characterization, you can treat them the same in many cases but not always
3b1b made a video on the duality of the dot product with a column vector and n*1 matrices I think
Does anything actually go wrong if you treat 1x1 matrices as scalars
Like this comment. What other spaces would that be a problem in?
Well as they pointed out above, you can multiply a scalar times an nx1 vector but not a 1x1 matrix by a nx1 vector
So you have to be tricky with the notation
Well yeah so it doesn't work in the 100% formal sense but also does that matter
I mean it always matters if you're trying to prove something or if you're doing a calculation and all of a sudden you notice the dimensions don't work out
Practically it's useful to identify 1x1 matrices with scalars
But identification is not the same thing as they're always the same
Yes I'm aware they're different objects
I was more referring to Namington's comment
if the only spaces you work in are F^n for a field F, there's no danger in doing so.
Like what changes here with infinite dimensional spaces?
probably the nicest thing imo about this is using it to write a projection matrix onto the vector v, $$P=\frac{vv^T}{v^Tv}$$ it's pretty straightforward to show $P^2=P$
Merosity
interesting
I suppose I should also mention, I-P would then remove the component of a vector in the v direction, which is also handy, and you can combine more things to remove more components or to make a plane etc
I see
Does anyone here have interview experience from a European, non UK quant job?
Im wondering if it is reasonable to expect the same type of questions as one might get in the US
consider, say, ℝ over ℚ
Wait why is that a problem
the 1x1 matrix [π] obviously doesnt correspond to any scalar
How is that a matrix though
Or a 1x1 matrix
The entries of a matrix have to be in the scalar field right?
no? they just need to come from a ring
oh i see your concern
yeah theyre not matrices in the sense of being compatible with the linear map structure of the vector space
since this is an equivalence you can only make for finite dimensional vector spaces
(since you need to fix a basis)
so the direction that fails is going from scalars to matrices
How do matrices work in infinite dimensional spaces anyways
they dont
like you can talk about matrices by just considering them as arrays of numbers from some ring
I forget do linear maps have to send basis elements to finite linear combinations of basis elements?
but
they have no relation to the linear maps of the space
theyre just an extra thing you can talk about
if you wanted matrices to correspond to linear maps, your matrices would need to be infinitely large
Is that a problem though?
well, not necessarily
the existence of such "matrices" in most contexts requires the axiom of choice though
there are some special cases where it does make sense to talk about infinite dimensional matrices as maps between infinite dimensional spaces
yeah like
its a totally sensible thing to do
but its not usually as useful as finite matrices are for finite dimensional spaces
since obviously you cant really use them computationally
good luck diagonalizing an infinite matrix lmao
lol
well a bunch of LA goes down the drain with those i guess
Lartomato gave me a lil excurse about R^inf and GL_inf some days ago when we talked about infinite dimensional matrices there
Proof that the series converges to sin(x): dude just look at it, what other function can it be?
same that's what I do when I look at $x \prod_{n \ge 1} \left(1-\frac{x^2}{n^2\pi^2}\right)$
Merosity
Okay but actually though
What is a suitable "calculus way" to define sin(x)?
Like if you just define the power series and prove stuff about it, its geometric nature/importance are unclear
I guess real/im part of the exponential?
spivak's "calculus" gives a definition of the trig functions based on integration
i don't remember it off the top of my head and i don't have the pdf on my laptop
where
it is and it is god awful to work with
especially when you're a first year calculus student fresh out of highschool (where "opposite over hypotenuse" was good enough)
I think the differential equation might be the strat tbh?
you can easily get that from the power series def
which then gives some geometric structure
plus i bet you could define them in terms of polar integrals
.>
maybe
the differential equation way is cool
really good exercises to get the other properties and all
@fervent pebble the thing is
You just write down a power series and it's like
Okay cool
y tho
I think you need to start with the differential equation or say real/imaginary part of complex exponential
hm
how do you get the power series def or the exponential def from the diffeq tho?
So in calc unless you're doing complex number stuff early (but then it's weird since you're not really doing complex analysis exactly)
You prob just have to do the diff equation if you want it not to be scuffed. At least that has the oscillation-y intuition behind it
That's a good question, I know exactly what that story looks like for the exponential but hmm
i mentioned it and a rando here called it "I need to touch grass" def, forever tainting my opinion of trig & DEs /s
It's still a little bit unfortunate because I feel like differential equations are fundamentally a linear algebra idea but hmm
@tall badge sometimes you get got in life man
It's how it goes
wdym the power series is just the power series solution you can get from doing like frobenius or whatever
right ?
i forgot how it goes tbh
lol
I don't know differential equations lol
Yes I study spectral theory of the Laplacian
I don't see the problem here
understandable
the other grass
I am the grass
touch grass 😉

holy shit
"woog" has been said over 30k times on this server
not counting pings
pinged over 15k 
poor woog
thats a lotta pings
that's also not counting mod pings which i bet he used to get
she
Isomorphism has only been said around 15k times?
woog isomorphism
but im gonna guess conservatively during when she was actively modding at least 5k times?
if you want to imagine the server prior to woog, you can in part imagine ann in charge
so im gonna say total count of mentions/pings > 50k
probably more counting channels i can't access like ivory
that's not a metaphor lol
if a mod looked it up they'd prolly see more results
Well I have seen old chill posts from Ann 
17.5k woog pings/mentions
yeah that's about 2.5k more than i saw
so i imagine the casual mentions are even higher
why do people always forget woog is a girl
I think a lot people just default assume people are boys online unless told otherwise.
It's not really a fair assumption to make ofc.
That's why you go for the genius "they"
honestly on discord i would say it is
although i should be saying they anyways
proportion of men on discord is >>>> vs proportion of women on discord
How do you know
discord is mainly a gaming service, which is male-dominated
But by how much?
probably a lower proportion than discord
but a subcommunity is not necessarily representative of the greater community
actually everyone here is female
under directive of the Biden administration, we are instituting mandatory femboyization for all our users
Is it. It's been shying away from branding itself as a gaming service for years now
Like since 2018 even I think
it doesnt brand itself like that anymore but it still carries the baggage
Sure but it has gained a lot of new users who don't know of its dark past
you can say the same thing about this server
What was it's Dark Past?
g*mers

That's true nami, you don't air out your dirty laundry in public often
theres also just the whole nature of like, how online communities tend to polarize
Like how everyone eventually gets trapped in their echo chambers
a space (a website, a server, whatever) pushes out 1 demographic, either intentionally or accidentally
and then that demographic ends up in another space
and they gradually grow more and more disjoint
even within discord, this much is obvious
anime servers have a radically different culture to the rest of this platform lmao
Well it's unavoidable since you can't cater to everyone
How?
idk how to describe it


theres a sort of like, forced positivity in a lot of them
its really hard to describe
ive been in servers that straight-up banned any emotes that didnt originate on their server
including global emotes
its fucking weird

I feel like being horny is also less frowned upon in those spaces
you say that as if #「ivory-tower-emeritus」 isnt constantly horny
as long as you hide it in your fur suit, it's fine
I feel like it'd be encouraged depending on the anime
Talking about cum and being horny are not the same thing nami
I don't think ivory has been horny for at least like, a week
i think this is homophobic
and theres plenty of servers out there that are just like
either openly racist or really not good at hiding it
naturally i am not part of any of them, at least to my knowledge
Like uhh
that one
and then theres like
trans discord
which i swear is a fucking alternate universe
such a different culture to the rest of discord
or even like, general LGBTQ spaces
nothing wrong with it or anything, just a cultural observation
How does it compare to this server?
this server also has a pretty unique culture lmao
we try and make it feel less "arcadey" than most servers in general
future techno-anthropologists will study these messages
so like
less focus on roles and shit
there's a reputation system technically but no one uses it lmao
no gimmick bots or whatever
the roles that exist are almost purely functional, except the activeness roles
Yeah some people have three different roles that don't seem to do anything
and a lot of these functional roles have intentionally obscured names
One time i was inactive for a while then lost my blue name and I was devastated. I place the whole value of my existence on the color of my name
like γ and ±
(no i will not answer questions as to what these roles do)
(but they do do something)
So "Role that does nothing" does something?
that one doesnt lmao
reminds me when I left the server I never got around to requesting my gamma role back cause I like the peace and quiet

that one was just to test the role assignment system
and we kept it around because:
(a) some people really like their useless novelty roles
(b) being able to assign and unassign a role can actually be useful for debugging discord
like, we have a 10 minute wait period on new members
but every once in a while it breaks for someone and lasts forever
and they can fix it by assigning themselves a role
literally any role
Weird
discord is a weird platform.
this server is simultaneously a lot more "regimented" than most servers and a lot less "restrictive" on where stuff goes
like, except for "dont post your questions in discussion channels (unless theyre interesting)" and "memes go in #chill "
we're not SUPER picky about like
#math-discussion going briefly off-topic for a bit
People do this all the time anyways lol
if it lasts a while we'll try and turn it back around
i mean sure but its mostly new users who do it
i.e. users who havent interfaced with the server culture
That makes sense
Come back to your hole mero
It's just that because there's a constant stream of new users it's noticable
but like, ive been on servers where there are like
8 different "topics" channels
like theres a #food channel and an #anime channel or whatever
noooooooooooooooooooooooooooooooooooooooooooooooooooooooo
and if you start posting about food in #general
ok maybe later but not tonight
youll be redirected to #food
that isnt necessarily a bad system, it works for them
but we dont have anything like that
the closest thing are the "misc" channels
and #cats
Those are so random
but you can still post about those topics elsewhere
Except #❓how-to-get-help
I think the good thing is there's a kind of "directedness" that math sort of imposes on stuff that gives enough structure that works well
yeah, i mentioned this earlier
but i mean like
So Math gives the server structure?
if youre intending to just discuss something
you can just throw it in a discussion channel
I get a similar vibe from any kind of technical help server, they have this sort of intrinsic help-helper system aspect
hell, if you want to discuss analysis you can do it in like
7 different channels
not counting help channels
and each channel will have a different "target demographic"
I feel like #discussion and #chill are used for the same things a lot of the time
talkign about it in #real-complex-analysis will probably result in people trying to help you with a specific question or concept
idk how other people use this server but I have all help channels muted then a handful of math subjects I'm not that interested in muted too, so I'm only ever seeing stuff pop up that is in subjects I'm interested in
kind of how I use MSE I guess too
talking about it in ivory or advanced-lounge will probably result in people discussing work they did in that area, or maybe even extensions and generalizations and stuff
(ivory will also probably result in it being interrupted by the occasional shitpost)
like have watched tags and ignored tags
talking about it in #discussion will probably result in a low-level discussion where people ruminate on their academic experiences

also someone chimes in with "wow this looks soooo hard" every 15 minutes for some reason
as if that adds anything
it would be nice if there was some way to implement this sort of tagging system in discord to emulate this kind of experience:
lmfao
i love that MSE has a collection of "bullshit filters" tags
like how arxiv has math.gm
haha
479 questions, 6 asked this month
do i dare click
"hey MSE, solve the collatz conjecture for me"
aww poor kid, closed
it seems this person simultaneously
- did not realize that a constructive answer was a solution to the collatz conjecture
- immediately recognized that it would be a solution when informed of such
which is like, a weird mix
big yikes
there are some questions that are tagged with collatz-conjecture that make no sense though
is this actually related to collatz
RIP cv
it's probably some convoluted way they've tried boiling down how to solve it and this is their "key" step or something
What're the odds that, even if they proved this, it wouldn't mean anything cuz they made a mistake somewhere else
100%
???
like i can see how this is maybe tangentially related but
weird af to give it that tag
maybe to draw more attention to it lol
haha
im pretty sure the collatz-conjecture tag would just make people avoid it more.
I still want to find out what the gamma role does one day
like I get that's it's probably either some admin thing or access to some private channel, and in either case there's really no reason I should know
and I wouldn't particularly it's just that I am gamma
(ignoring the imposter that joined recently)
Nami is banach-alg a functional role or a shitpost role?
active and very active were mistakes tbh
they should be removed
no
yes
bad take
@strong sphinx how are 4-digit question numbers 'seemingly infinite'? :P
Every other book I've seen just restarts the question number for each section but Demidovich just keeps going. I have to respect it though.
does anyone know a proof not by diagonalization that $|\bZ| \neq |\bR|$?
13 x 7 = 91
yup
would you like to show it to me
yeah
ok so say you've got a countable list of every real number written out as their decimal expansions
okay
then I start constructing a new number 0.(stuff). The first decimal of this new number is made by taking the first decimal of the first number and adding 1 to it, if that decimal is a 9 then I just set it to 0
I then construct the second decimal by taking the second decimal of the second number and doing the same operation as before, I then do the same for the third and fourth and fifth and so on
at the end of this process I now have a new number that is by definition different in at least one place to every number on the list

so it's a number not on the list
wew
thus R is not countable
bad








