#serious-discussion
1 messages · Page 296 of 1
Thanks, that bugged me
It also bugged me when Richard Behiel, in his newest video, cancels Psi on the right side of the equation; excuse me is this an integral domain? How do I know that manipulation was valid, it feels illegal
this is wrong, the complex plane and the extended complex plane are different, only the latter is a topological sphere
i have theorem i will get it in a sec
If you can find a homeomorphism between a compact space (sphere) and the Euclidean plane, I will be deeply impressed.
im not trying to find homeomorphism im trying to find the theorem(i hope it will work) not 100 percent sure tbh
extended complex plane.
What do you mean by that?
one sec let me type it out
complex plane homemorphic to riemann sphere(stereo projection) and riemann sphere homeomorphic to sphere
thus complex plane homeomorphic to sphere
its the transitive property
complex plane homemorphic to riemann sphere(stereo projection)
What do you mean by that?
homeomorphisms are an equivalence relation
I know this; how do you understand "the complex plane" here? And how do you understand "riemann sphere"?
the complex plane with its usual topology
And the Riemann sphere?
both with their usual topologies
What is the Riemann sphere?
does the argument make sense?
i couldnt find the theorem i wanted to
so i just did the transitivity thing (easy way out)
It doesn't, because the complex plane is not compact, and the Riemann sphere (which is homeomorphic to the extended complex plane) is compact.
A compact space cannot be homeomorphic to a noncompact space
oh wait yeah your right mb
This is the stereographic projection I think you're talking about. It maps P on the sphere to s(P) on the complex plane by drawing a line from the north pole N through P. Notice that you can't map N itself to any point on the plane; you either need to remove N, or add a point to the complex plane, which we call the point at infinity
oh i get it now, they are only homeomorphic with the extended complex plane right?
rieman sphere homeomorphic with extended complex plane
not complex plane?
Yep, the extended complex plane is homeomorphic to the sphere
oh ok that explains it
I prefer to think of it as R^2 is homeo to the sphere minus a point
i havent done math in a real long time lol
extended r2
not r2
right?
but the sphere minus the point is the rieman sphere then right?
The better way to think about it is that R^2 has S^2 as the one point compactification, so when you add on the structure of C to R^2, it turns S^2 into the Riemann Sphere
@zealous garden ur bio says I should ping u
hi guys i need help
write random number n
print all the purmutation of n
like n =3
Based, thank you
!help
To ask for mathematics help on this server, please open your own help channel or help thread. See #❓how-to-get-help for instructions.
whatever question you have its probably an unsolved problem in number theory tbh lol
Np I'm tryna be nice today
After all my problems got t solved and I almaot failed math and my friends betraying me but ok
Wordle 1,498 4/6
🟨⬛⬛⬛🟨
⬛🟩🟩⬛⬛
⬛🟩🟩🟩🟩
🟩🟩🟩🟩🟩
Yo
gng I want to learn the secrets of hte universe, what should I self study?
Astrophysics
And pure math
yeah thats what I thought. Ive mastered single variable calculs and got a B in differential equations. I havent done much physics since high school. An I ready or should I do the book fundamentals of physics first?
Astro study the very small and the very big and all in between and math allows representation of disparing concepts
Fundamental physics
ok ill probably do that
Then go reading on math as the tools pop up
and I did calculus 3 but im not proud of that experience, but I know hte basics tahts for sure
Start with kinrmatics and go climbing up
Physics allow physical corolation
Is better to learn
alr im thinking do the book called fundamentals of physics, then an introduction to modern astrophysics
let me check
Fymann are also good
yes
I mean, why not just do a classical mech book, a electrodynamics book, a quantum book, and then anything after that
Good also, but fundamental physics is good to show how basis is set
More how we define physics, good start to show what is fundamentals and the machanics of how physics experimental and theoretical is done
What is 0 decided by 1
could work
0/1 = 0; 0 choose 1 = 0
K tbx
Hey guys, I built something for y'all and I would love your feedback...
I guess I cant attach a pic, we built a fully navigable database of math. Currently, we have linear algebra and we're working on adding calculus by the end of Septemeber
It's a full graph of theorems, definitions, axioms, etc
seems like a for-profit app and that you joined solely to post this link, I'll let you know unsolicited advertising is against our #rules
with eventual paid tiers, from what I gathered
honestly it would be fine if you had a link in your bio or if you actually had a demo you wanted feedback on
We do have a demo, that's what I'm trying to share
Yes, the paid tiers are geared towards professors/researchers
It's in my bio now if anyone is interested
okay, sorry if my reaction seems a bit harsh we just tend to enforce a strict policy on this given that we're a big server that gets a lot of spam
fwiw the concept seems interesting
Thank you! No worries, I understand how annoying spam and promos are. We've been trying to filter out spam from our signups as well
Wtf
yeah I'm kind of curious how was it useful 
does your life include a M/M/1 queue or something
I once used markov chains to estimate DPS in World of Warcraft
What is this
fancy way to say you have a queue, clients arrive according to a Poisson process and they also leave in exponentially distributed intervals https://en.wikipedia.org/wiki/M/M/1_queue
continuous time Markov chain
Oh lmao no my markov chain was not this complicated
Granted if you want to use it for something big it will be more complicated
Arguably with continuous time this isn't really a Markov Chain? At least when I hear "Markov Chain", I think a finite number of states with transition probabilities between them.
tfw dynamical systems
I should study dynamical systems it has been intriguing me recently
countably many states is a common thing to assume as well
Although the Wikipedia article seems to refer to the MM1 queue as a "continuous time Markov Chain", so I will not fight over terminology
Random walk in Z? That's my wife!
a continuous time markov chain can still have finite number of states, though it's defined in a more complicated way than the discrete "depends only on the latest state"
and you can still represent it by a matrix in the finite-state case
read Deninger's papers

meme or serious?
if you want to learn some spooky arithmetic geometry and also learn all these analogies with dynamical systems yeah it's worth reading
there is plenty of stuff from Deninger's papers which is less schizo and is actually useful in practice at least
there is a lot of related work by Connes and his coauthors on this sort of thing too
At first glance I don't see any reason not to read more about these stuff
a nice perspective is like, if you have a variety over a finite field F_q then the Frobenius gives you a discrete dynamical system and the closed points are periodic orbits
something like this should be true over Spec(Z) or over Arakelov compactifications but it's more schizo
In this paper, we present a geometric generalization of class field theory, demonstrating how adelic constructions, central to the spectral realization of zeros of L-functions and the geometric framework for explicit formulas in number theory, naturally extend the classical theory. This generalization transitions from the idele class group, whic...
this is the closest I've seen to actually talking about these things rigorously
peak
the adele class space A_Q/Q* that Connes talks about is the same one which Connes uses to give some very natural spectral interpretations of L-functions
there is a scaling action by positive reals on this space, the periodic orbits have length log(p) for each prime p, which is part of what Deninger's conjectures demand of Spec(Z) seen from a dynamical point of view
lmao
gives pi_1 vibes
yeah the paper I linked really nicely explains how this sort of thing makes the knots and primes analogy very explicit
like it's the closest thing you can get to actually viewing primes in Spec(Z) as "knots in a 3-manifold"
the main catch is this has to be some sort of non-commutative space
this seems like a nice thing to give a talk about
take one copy of this picture on the left for each prime p and then glue all the generic orbits together, so you're left with one periodic orbit of length log(p) for each prime p, and a single generic orbit
you can't do this sort of gluing in a reasonable way without working with noncommutative spaces
also quite nice because the paper spells out how you can recover class field theory and view the monodromy of periodic orbits in this space in terms of linking numbers

it is quite robust especially when considered in the context of Connes' other work using these same spaces to do spectral theory around L-functions
it's a little frustrating that this sort of NCG approach is sort of orthogonal to algebraic/analytic geometry but it's clearly useful in a lot of ways
maybe when you're handed a construction like this which seems to have all the right properties you shouldn't complain just because it's not what you expected
ultra knows way more about NCG than I do in case you do want to dig into this stuff and ask questions
this is nice 😌
What does this even mean.
I feel like it's kind of impossible to draw a reasonable picture for the second thing
the first thing is very classical though
I guess the closest thing that comes to mind for the second thing are various pictures of chaotic attractors
those are also very classical examples in dynamics
They do look like that, but are they really that in some sense?
I mean the graph looks similar.
should be taken with a large grain of salt not least because we have no clue how to write down these required arithmetic dynamical systems
So, they are / maybe "arithmetic dynamical systems?"
in the finite field case yes these things exist with all the right properties, in the number field case sort of yes sort of no
there are extremely strong analogies that constrain what should be true about these things and there are some constructions which realize some aspects of these naturally and realize some other aspects in a very unsatisfactory ad-hoc way, with no good way of addressing certain analytic difficulties
like along with these sorts of dynamical realizations of arithmetic schemes there should be some corresponding cohomology theories which are in close analogy with previously studied leafwise cohomology for foliated spaces
this diagram is so satisfying to look at
you should have a dynamical Lefschetz trace formula for these which recovers zeta functions of arithmetic schemes just as the usual Grothendieck-Lefschetz trace formula recovers zeta functions for schemes over finite fields
it's just that the kinds of dynamical systems that necessarily show up in the number field setting are much more complicated than the discrete dynamical systems that play the analogous role over finite fields
The three classes of being.
I wish I understood more of these terms, but this is interesting.
part of what is appealing about this sort of idea is that it gives a completely uniform cohomological description of the local factors of L-functions
over non-Archimedean places typically you can write down the corresponding local factors by taking the inverse characteristic polynomial of the Frobenius acting on suitable cohomology theories for varieties defined over finite fields or their algebraic closures
over Archimedean places it would seem like there is nothing like an "Archimedean Frobenius" which plays an analogous role and which recovers the usual Gamma-factors in completed L-functions

something like this does exist, at least through some slightly ad-hoc constructions, but it's not the sort of discrete dynamical thing you would get by iterating Frobenius over finite fields, instead you genuinely have a continuous flow with infinitesimal generator and you can look at regularized determinants of this
Where's the cohomology
yeah good question, nobody knows any good constructions of this
like we know what it should be locally at all places but you can't just glue together the expected answers over each place and expect to be able to extract global arithmetic information like this
similar to how we know what the Fargues-Fontaine curves look like locally at all places but nobody knows how to define a "global Fargues-Fontaine curve" which specializes to all of these upon localization
but like here's a fun example to contemplate, consider the following R-vector spaces, for each place v of Q:
for v=p a finite place take R_p to be the R-vector space of R-valued finite Fourier series (in y) on R/log(p)Z
for v=\infty the infinite place take R_\infty to be the R-vector space R[exp(-2y)]
(Side note it's crazy to go from writing in Hebrew and Gandhari today now trying to understand this.)
all of these carry a natural R-action s^t by translation (s^t f)(y)=f(y+t) with infinitesimal generator \theta=d/dy, whose eigenvalues are the poles of the corresponding local L-factor \zeta_v(s), and one has the following regularized determinant formula:
$\zeta_v(s)=\mathrm{det}_\infty(\tfrac{1}{2\pi}(s-\theta)|R_v)^{-1}\text{ for every }v$
nGroupoid
here det_\infty is the zeta regularized determinant exp(-\zeta'_\theta(0)) where \zeta_\theta(s)=\sum_\alpha \alpha^{-s} is the spectral zeta function of \theta
you can similarly rig up such formulas which recover the local L-factors for basically any variety over a number field, the R_v's should be some shadow of some globally defined (and almost always infinite dimensional) cohomology theory which nobody knows how to define in an especially natural way
assuming such a theory exists it seems like it should involve a lot of really hard analysis to say anything interesting about on top of any already difficult arithmetic geometry
Thanks for the explaination.
compare this to the Ruelle zeta function: if you take the primitive orbits of the flow to be the periodic orbits of length log(p) for each prime p then Ruelle zeta = Riemann zeta, this is where a lot of the analogies come from
https://ncatlab.org/nlab/show/Ruelle+zeta+function
I recognize the form here is an SSP (sum sum product), but the extact motivation is still unclear.
hyperbolic manifolds section, it does seem useful, I feel like I have seen holonomy and length products elsewhere.
these Ruelle/Selberg zeta functions tend to show up a lot in quantum field theory around various partition function computations among other things
they show up naturally when working with various sorts of dynamical systems that's the main context in which these things are defined
the way that Selberg originally encountered these things was in parallel with the Selberg trace formula on hyperbolic surfaces where you can produce nice formulas and asymptotics for counting geodesics
this is the same sort of question as asking about the spectrum of the Laplacian for various hyperbolic surfaces
the Selberg zeta function of a Riemann surface encodes the length spectrum and this is an isospectral invariant, that's one good reason to care about such things
you can get explicit formulas relating the zeros of the Selberg zeta corresponding to Laplace eigenvalues and counting prime geodesics up to a given length bound, just as you get explicit formulas relating the zeros of the Riemann zeta function and counting primes up to a given bound
pretty much all the things you can do with one are in complete parallel with the other, including various conjectures in both settings really looking like the same statements in different contexts
Strange, I wonder where else they show up.
these things are very closely connected to analytic torsion/Reidemeister torsion and eta invariants, those also show up a lot in physics around QFT and encode a lot of subtle information coming from index theory among other things
Honestly I wish I knew QFT more from the math side.
Although, this is really really neat, and a lot of things flying around at once.
It's really cool you know this, and it's really cool that you're like explaining it out.

The main reason I know about QFT is LQFT and trying to program that. Along with regge calculus, etc. I know mostly about Zeta stuff from the p-adics and Lattice theta functions being used in string theory. And me trying to understand why lattices show up there. I recognize a lot of the formulas/equation types(?) from this conversation. As I said I did see the SSP which I know from LQFT actions. (LSS being a programer sucks.) But over all a lot of terms, equations, and mathematical notation sort of flies over my head.
Also, Modular stuff sort of goes with trees, and just generally modular forms / theta functions, etc.
Lastly I guess know a lot of this stuff does blend together at different points.
hello um anybody familiar with the Diophantine equations? I have a simple question: the satisfactibility problem (SAT) is turing computable so is computably enumerable (simple proof), then by Matiyasevich's theorem it is diophantine correct?
Do you ever feel like you arent good enough for math? My uni entrance exam results came in couple days ago and i can only enter unis (specifically the good/top ones in my country) for a science major and i am not even good enough for engineering stuff unless i apply for a more average university , maybe this is just an asian system sorta thing where they do exams super duper hard just to eliminate people and it doesnt show ur true worth but there were still people who did better than me and who were good enough despite the fact that i also studied very hard too (i feel like i would have felt this way less if i didnt study very hard)
maybe i am indeed not good enough for math/physics stuff even though i genuinely enjoy studying them a lot idk
Do you ever feel like you arent good enough for math?
all the time. you might be surprised how many people feel this way 
is it really good enough though if u are really passionate about something but not necessarily talented at it? as much as i dislike beiliving in talent
the whole uni exam thing made me realize no matter how hard i try or how much i love the thing i am not good enough
a single exam is not representative of your capacity to do mathematics
ye thats true but it really feels like it when u worked for it a lot
like whats the difference
Between me and the people who did better than me?
Where did your marks get cut?

If you work on the areas where your marks got cut, then you should be scoring very well in the next exam
…
this is embarrassing to admit but there was a reading compherension part
on the exam
and uh
that didnt go very well-
but people who are better than me did both good on that and also math/science part stuffy
so therefore
I am still not good enough
because surely there are people who can do good on everything
It will take time to progress, don't give up and keep practicing in areas you are weak at

that's my thoughts too. and i still do think so. but people here have started to convince me that perhaps there is a reason no mathematicisn knows everything
Math is girlboss

do you ever feel like you arent good enough for math?
Yeah but I don't really care. If you have the passion for it then you wouldn't mind getting stuck on problems (because you will do all the time). You need to be stubborn enough and things will click eventually
Do you ever feel like you arent good enough for math?
It seems like for an eternity, then again I program mostly.
My uni entrance exam results came in couple days ago and i can only enter unis (specifically the good/top ones in my country)
Apprently I did above average on my entrance stuff, and I never studied.
But I'm not sure America counts in this way.
I also don't know if Uni here would be the same thing.
Everybody here in chat* rn knows more than me.
probably false
ye but i would like to be atleast enough for it
like i am pretty ok with not knowing everything
to be honest im not even sure if it has a truth value
Honestly knowing everything would be very boring or like being very smart
Okay then you just put in work
but i still wanna be enough for the thing i like D:
Wowi
I did and it wasnt enough
For what?
I mean passing literally means you're okay for it.
It's more a "must be this tall to ride" rather than "noooo you will suck."
I don't think this is true. Given enough time and dedication you'll get the hang of some topic
Idk by top universities’s standards?D:
It doesn't work for this case
Then study?
There are other factors you need to consider
I did though
like from the exam’s standards i get into 2-3% but the good unis dont really care for anything below 1-1,5% it seems like;_;
Did you pass?
wdym pass?
The exam?
doesnt everyone who get into the exam automatically pass it though
By pass u mean get into the place i want
n o
Can you retake?
ye but after one year
I genuinely have no idea what the people who did better than me did differently
And i really dislike the idea of
t a l e n t
like i dont wanna beilive that they are just naturally better than me
mannn
Well u have to ....I had to and it sucks
D:
Yea );
Everything can be learned, or at least gotten to.
If you want to get into the "top" go into a "mediocre" one and transfer afterwards idk
that was my plan too
or well like go into a science field in the top ones
but i feel like
I am being too delusional again
Like having too much hopes on that
if i work hard enough
i can get there
Even though i already tried the working hard thing
But if i cant do it i will probably crash down very badly D:
Idk I really can't give good advice on these stuff. They tend to involve nepotism and P2W methods so if you have these ig you can get in
But the advices I gave should work for math
Working hard and study never goes in vain u will always have some result
or any field of study really
This I have heard of a lot in the US.
You will do well, and I hope you get where you're going.
Okay, explain something here, is the a_{i1} ... a_{il} like adding them together separately, adding over them, or multiplying?
Thank u-
slop
Ask Aluffi, Chen or Marcolli
It's a mistake
The amount of times I see slop... It's not "weird" anymore
Its ryoever
I'm 16. I love math. But I somewhat suck at it. I started studying graph theory and fell in love with it. Now it's getting harder and harder to understand certain ideas cause I have zero background in pure math. Do I continue this or do I start over from the basics?
Start with basics I'd say honestly u need some kind of guidance for maths not teacher but a guide who'd show u the right path to study
Ok then. Thanks!
the optimal path to mathematical mastery
you should probably know the basics, but learning math is not really "linear" in the sense that you learn all math up to certain level before moving up to more advanced stuff
after some point there's too much to learn to cover every branch of math so you have to backtrack whenever you see something new
That point of yours actually makes so much sense. Thankyou!
@rain patio go to #❓how-to-get-help
as you said yourself, you need to make a private channel
everybody lies
Hi @neat lintel, welcome to the mathcord! I noticed you posted your question in several other peoples' help channels and people quite politely asked you not to do that. That's spam! Please don't do that!
I know it can be very frustrating when people don't immediately come and help, but everyone here is a volunteer, not on payroll.
but what if he is hungry and he needs to eat spam for hunger purposes
Hey everyone, I'm a French student and I was wondering if there's an official mathematics curriculum for grades 9 to 12 in US. I'd really appreciate book references or any resources that could help me understand in depth what is taught during those years.
I tried looking it up online, but every website seems to say something different.
t'es en quelle classe frérot ?
en vrai il y a pas mal de trucs ici, tu dois pouvoir trouver de quoi te satisfaire même si tu es au lycée
I don't think there's a definitive list but the kind of common list of courses (optimistically) would be alg 1, alg 2, geometry, trig, precalc, maybe calc 1, but very often not. It depends heavily on the individual school though.
When I went through the US public education system a very long time ago the requirement to finish high school (12th grade) was just geometry with a C or better I think. I remember taking alg 2 and dropping it to work part time.
Later when I went back to school at the community college level it seemed most students finishing high school aiming towards stem degrees finished as far into calc as they could.
So probably the mandatory amount is lower than that.
there are different state standards and varying textbooks, unlike France
it looks like US Integrated Math is more like what's taught in France
the sequence Dooter said (except Trig and Precalc are usually the same course) is common
Hi Im planning on posting an answer book for exercises in curtis abstract linear algebra on a blog or something, mostly for fun, but also I was wondering if colleges would care much about that.
I know thats pretty basic but if I can show at least some profficiency in proof writing I feel like that would at least make me stand out a little.
If it makes a difference Im coming out of highschool not cc going for a physics degree.
I have an encryption method/program i've been working on and i want to ask if people can help determine if it's secure. could that just be discussed here or would a help channel be better?
can you show it?
i'm probably bad at this but i wanna see it
yeah, here is the most recent version: https://github.com/Ernesti04/general_projects/blob/main/necklace_data_encoding_v3.py
you can either run the program and it'll ask you for stuff or this version has command line arguments as an option. you do -d to decode and -e to encode, -o followed by the output type if you only want one, then do -t followed by either the text to encode in quotes or the encrypted text to decode
it's still a work in progress so the code is a bit messy but i tried to label a lot of it, please let me know if there's any bugs
Wait, does it do encryption or encoding? There doesn't seem to be any keys/passcodes involved
i don't have a method of exchanging keys so they're all at the top of the program with a b c and d
Oh, I see
Just to be clear: encoding is very different from encryption: encryption involves keys: you need a key to encrypt the plaintext, and you need the same key to be able to decrypt it again. Encoding on the other hand is just a reversible transformation of the data, like base64 encoding or HTML encoding
ok, i can run statistical tests on it tommorow
.remindme 17h do this
Created reminder do this (#serious-discussion message) for <t:1753706330:f>
You seem to use the words encoding and encryption interchangeably, so I thought I'd point this out
okay, i say encoding because if the program is left unchanged then it is. if you change the keys then it is encrypted because someone else has to have the same set to decrypt
it's very confusing though so thank you for pointing that out. I'll try to get better with it
okay, thank you!
@sharp cedar sorry, i'm on my phone so its hard to read the code, can you explain how it works?
block cipher or stream cipher?
it's weird is what it is lol
i think stream cypher fits better(?) it converts each character but it can also treat each word differently if you split it (poorly named standard mode in the program)
but it converts each character into a binary necklace based on a b and c with the alphabet being shuffled with d. the higher the values for a b and c the harder it is to brute force
necklace?
i'm pretty sure that just one or the other would basically be a cypher, but both of them together (at least as far as i can tell) makes it pretty secure
can it encrypt arbitrary binary data?
is the substitution the same for all letters?
i think it's a combinatorics thing? basically a set of characters with an alphabet and a given length where none of them can be cyclically rotated to match another
it can encrypt binary, it treats it as individual characters when doing it however. each letter is substituted on its own but will match, so they are randomly rotated. this makes it so you can't get the original value without knowing a
technically a is the key length, it't the length of the necklaces
you could theoretically factor the length to get it so non-prime values are used and a random set of junk bits is added to the end to detur that
d is used to seed the random function so whatever that accepts is the bounds
b needs to be low enough that there are enough possible necklaces in a set of length a to fit all alphabet and spacer characters
c needs to be small enough to accomplish the same thing
so b and c are bound by a, a higher a gives b and c a higher bound
I think in practice d is the actual key; a, b and c seem too restricted
a b and c together is basically a key while d is its own key
abc determines the necklaces that will be generated
It's not a good idea to have two keys, one of which is only "basically" a key. Should you keep a, b, c secret? If b and c is almost always between 1 and 10, how much extra security does it provide to keep them secret?
Almost all cryptosystems have only one key, it's more practical, and usually more secure, because it's easier to protect one key than two
i mean
if a, b, and c are all between 1 and 10 that's slightly less than 10 bits of information
you could do say, 8 rounds of this
with round key a, b, c, d generated from a 128-bit key
and also the 1-round version is breakable
for every possible a, b, c, do frequency analysis to recover d
my thought is that abc is for configuration and d is what's used as an actual key
plus the values aren't that limited. with a, b, c = 24, 31, 3 i get
xcb210674h884j7f04h28h29h8d11h8ff0406g8fk121iec3g233gc31h1
this actually works for an n-round version as well, so this is insecure if i understand it correctly
you need to add data-dependant stuff
You could think of a, b, c as parameters, like the exponent e in RSA. It's not secret, it's just something you can adjust
yeah, that's a much better way to look at it i think
Btw, I think this line makes the encryption non-deterministic: https://github.com/Ernesti04/general_projects/blob/fff7a91dd673ae00fb7bb1cb8abfc7300b676290/necklace_data_encoding_v3.py#L349
Since you're using random.randint after this point
you could make d operate on 2-byte sections
and iterate it like 32 times with key-dependant a, b, c, d for each round
that would be much harder to break
that's done so the necklaces can be rotated randomly and not be determined by any seed. if a b and c are known then it makes no difference but it makes it harder to determine otherwise
what do you mean by that?
you have some big key
and you generate a, b, c, d by taking various bits from it
for each round
or, you could set a = 128, do 1 round, but do it in CBC mode, and then do like 32 outer rounds of that
Hmm, so encrypting the same text multiple times results in different ciphertexts, but it can be decrypted to the same plaintext?
yes, exactly that. and so multiples of the same character are harder to identify
oh, interesting
okay, i'll look into it!
frequency analysis is still possible with this kind of cipher however
so d only operates on individual characters and not groups of them?
right now d just shuffles the alphabet once
actually, you need to do it in reverse CBC every other round for security
That's pretty weird, not sure how much effect it has on the security 🤔
actually, this still isn't secure, you need to do some kind of data-dependant stuff
Actually maybe not so weird, I think it's basically the same idea as using a random IV for a block cipher
true
Btw, I hope this is more like a fun project and not something you're planning to use for important stuff. Creating a secure cryptosystem is really hard, and nobody here really has the expertise to say for sure whether a cryptosystem is secure or not
oh yeah, i'm a cyber operations student and thought it'd be fun to try and make something like this. i'm just wanting to see if i did a good job of making something new that's secure enough, not something to replace anything lol
I see, that's cool
I have one last suggestion: maybe put some of the code in the main if-statement into separate functions encrypt() and decrypt(), so it's easier to read, plus it makes it possible to use the encryption directly from python
yeah! this isn't a final version yet (i want to add encrypting/decrypting files as well) and i planned on doing things like that to make the code neater!
to people who want to try and break it, here is a sentence with abcd all modified
x5a1661k529w4e71C2121scf81l4261D757oa421r2121y4622s81a2w529r345z4f12m8v5b82u85m529o345vd78B69dp307F4942k8563za611z8224k1
do you have any suggestions?
also here's another program i've made on the same concept if anyone wants a visualization or cool looking graphics
https://github.com/Ernesti04/general_projects/blob/main/improved_circle_plotting.py
Dih
try using letters to index others
e.g. for each letter, set it to the ((previous letter's index + this letter's index) mod length of alphabet)th letter
so like each letter in the text to encrypt becomes a different one based on the one before it? how would that be reversed? get the first letter and store its index, get the next letter and reverse it by the index of the previous letter? would it be better if they cascaded or only the original index of the previous letter matters?
?
cascading is better
you do it backwards + subtraction for reversing it
okay, thank you!
that sounds like a really good addition so I'll work on adding something like that next time I'm working on it!
what book is this? :o
My first complex analysis book
wowie! i miss complex analysis and i need a book to continue my knowledge and review :(( (wont ask here bc theres a channel for that)
<@&286206848099549185>
you gonna say something or just ping helpers?
No
I need help
I alrdy made a forum
for future reference you should ping helpers in there and not here
have a good sleep!!!!
helooo 
hello!!!
hru??
i'm alright :] about to head to bed, hby?
Its morning here 
Good night

xDDDDD
@vivid halo
Hey so, I was wonder about that equation you were talking about the other day. I have been looking The Ruelle Zeta Function over more, and read a bit about the motivation for his Selberg trace formulas. It has me thinking what exactly do they mean by f: M → M. I see the Selberg trace formulas looks at "sums of eigenvalues of the Laplace operator" could f be the laplace operator then too? What exactly could f be?
This is rather different geometrically speaking to what is considered for Selberg zeta functions
f is certainly not the Laplace operator
Dih
The second section on that nlab page for compact hyperbolic manifolds is a bit closer to what would show up in Selberg’s trace formula
That only feels a little better in terms of understanding it more.
These Selberg trace formulas are an identity between information involving the eigenvalues of the Laplacian on the spectral side, and information involving orbital integrals and geodesics on the geometric side
A lot of what Selberg was originally up to was extracting interesting spectral information from geometric information like this
This becomes hard outside the case of compact quotient because both sides of the trace formula will diverge and require regularization
Man physics people do wild things, honestly.
This is what number theory people do too 
Honestly wild enough, I like the thought in it.
I found that the otherday.
And then this massive survey by Arthur summarizing various aspects of his generalizations of this https://www.claymath.org/library/cw/arthur/pdf/62.pdf
Arthur’s trace formula is probably the biggest cornerstone to most of the Langlands program in the classical arithmetic setting
A lot of modern number theory (e.g. the proof of FLT) would not exist without some form of these kinds of trace formulas
Tr go burr.
Trace but everything is infinite dimensional
Honestly I need to learn more things.
All of this thought doesn't seem too bad, just more of a time sink.
Besides recursive functions (I love them, but can't understand them at that level).
What else have you been up to?
One's you're reading or writing, and like on what exactly?
Writing
Writing what?
Three or so papers (had to split them up because it got way too long for a single paper that tries to do everything)
Surveying and organizing various examples around Beilinson’s conjectures on special values of L-functions, mainly around those aspects which do not fit into the usual conjectural framework but where there is clearly room for generalizations
Upshot is there are more L-functions which show up in nature but which lack clear explanation
in nature
Mean's what exactly in this context?
Also lots of parallels with partition functions in physics which have guided the thinking around this
By nature I mean they show up even in pretty low hanging examples that one inevitably runs into when playing around with mixed motives
But also like half the people I talk to about this stuff locally (a few collaborators) are string theorists and a lot of the examples I have were things which showed up in their work
Okay, so for an example the Ising Model, does it show up there somewhere?
No they mostly work around computations of string amplitudes particularly in higher genus
You encounter things like modular graph forms and iterated integrals of Eisenstein series or other kinds of automorphic Greens functions for example
Well, that might be why I haven't used them before.
There are quite a few examples around genus 1 amplitudes which involve iterated integrals of modular forms which are very much beyond what the usual conjectures about L-functions can tell you about
Those are what got me interested at the start of the year
Or like here is a more accessible example related to Chern Simons
If you have an arithmetic hyperbolic 3-manifold (e.g. a knot complement) then its volume can be computed in terms of special values of dilogarithms and the value at s=2 of the zeta function of its trace field
I know about modular forms because of lattices, and messing with them.
But there is a natural complexification of the volume which recovers both the usual volume and also the Chern-Simons invariant
Why would integrating them be interesting?
Well this is how you produce period integrals which are known to relate to special values of L-functions
Like the L-function of a modular form is literally just integrating from 0 to i\infty as a Mellin transform
Huh.
There are much more general L-functions relating to various generalizations of modular forms but the relation to period integrals is nowhere close to this simple
So then \int_{0}^{\infty} \theta(\tau)x^{n} dx?
Well I'm looking for simple, so I can like grab and work up a bit.
Vaguely like this yes
Like how would you write it out?
here \theta_f(t)=f(it)
(this is when f has constant term 0, if you have something like an Eisenstein series you have to remove the constant term otherwise things diverge)
in general you are usually given something that behaves like a theta function \theta(s) and you can write this as a sum \theta^0(s)+\theta^\infty(s) where \theta^\infty(s) has exponential decay at infinity, then take the Mellin transform to get an L-function
but you can also iterate these Mellin transforms and this is where you start to get some really new things going on
a favorite example involves the double L-values of Eisenstein series \Lambda(G_4,G_10;s_1,s_2)
if you let f=\Delta be the cusp form of weight 12 for SL_2(Z) (this is the simplest modular form which is not an Eisenstein series) then you can write the first really interesting L-value L(\Delta,12) as a very particular linear combination of \Lambda(G_4,G_10;2,5) and \Lambda(G_4,G_10;3,4) for example
but there is also an accompanying period integral "c(\Delta,12)" related to \Lambda(\Delta,12) which does not fit into the usual conjectures about special values of L-functions, nevertheless you can again write it in terms of \Lambda(G_4,G_10;3,5)
so in this case you get a very interesting transfer of information between these double integrals of Eisenstein series and single-integrals of cusp forms, as well as a "new L-value"
these "new L-values" are confusing for the same reason that the "complexified volume" of 3-manifolds is confusing and subtle
it is a little tricky to formulate the right kinds of conjectures because these these types of "new L-values" will live in something like R/Z rather than R, there is some extra indeterminacy that you need to account for compared to the usual conjectures about special values of L-functions
a good analogy is that the special values of L-functions tend to have conjectures of the form "|Period|^2 ~ L-value" whereas now you want to say something about the argument of the period rather than its norm, which has additional ambiguities or anomalies associated with it
in physics you typically cannot observe phase angles like this but there are still some things you can say about them especially when keeping good track of anomalies and so on
Are these functions related to the gamma function?
I know that might seem out of pocket, but they look similar to me.
you typically see terms involving the gamma function as coming from the Archimedean place sure
Also, again off topic a bit, is there like an inverse real gamma function? And not an approximation?
what do you mean
Well I have been dealing with functions of logs.
And I found that it would be nice if there was an inverse gamma for x >= 1.
Yep, it would help me get exact numbers for a problem I have.
certainly this exists yes although it cannot be single-valued and I don't know of any good integral expressions for this
And I could also solve another problem because of it.
Well I know of an approximation, which gets better with bigger numbers.
I also know I can make the approximation better, just by adding some terms.
So, I feel like it should be possible, but it might look weird.
I don't want all of it though.
Like a sqrt(x) isn't the whole parabola if that makes sense.
even just a single branch isn't going to have such nice formulas
I know, I have seen the aproximation ones.
if approximations are enough for what you need then yes this can be improved
Well, I just need what I posted in chill.
but yeah I've never encountered this function before
Yeah, it feels like it should be more used, it would solve a few things.
I did say I found an inverse for the log integrated Kepler's equation, but it involves the inverse gamma like the one I have in chill.
But if I want an exact inverse, it needs this function to be exact. Not just approximated.
this is probably the best sort of exact formula you can hope for I would imagine
reciprocal gamma has a much cleaner expansion and then you can extract some kinda awful expression in terms of polygammas from this if you like
works for any choice of branch you want
there's probably other formulas which might be more or less appropriate for what you need idk
Again I stumbled across it, and sort of was like huh and kept going.
yeah lmao I feel that
I ran into some strange stuff with elliptic q-gamma the other day 
it's always funny to me when you run into the inevitable section on these special functions pages like "here it is as a Meijer G-function" like wow thanks that's useless
Are all your books in physical format? Or just your favourites? o.O
Most of my books are non physical
The physical ones are either really old (I got them when I started getting interested in serious math) or my favorites
so ca is special :3
complex analysis is very cool
love it
I’ve been working through a lot of it in #1368757937954099342
Lee is really good at exposition and quite a careful writer, imo
he does talk a lot, but I don’t mind that 
Yes :3
Indeed
very gud

Woke up, showered, ate chilli. Then I failed at solving my exam assignment and prepared some questions for the upcoming "extra class". Wbu?
So short answrr: oki?

I'm reading over NT
I kinda lost interest in my research project this summer 
only one month left

but it was a good experience
Do you heave a deadline?
yeah it's basically a job
why
yeah
but math

Btw ehy did you mute discussy 1
also it seems like my advisor won't really write a recommendation letter
:(

Why not
idk
You're not lazy right?
no?
yeah idk
well if they don't want to write one it's not the end of the world
I will choose a different prof

yes
New field? 
the one im doing currently is combo rep theory
Mhm
yeah idk
I know a prof I'm really familiar with
but they don't take undergrads for research

Wait
I thought you were in masters
Now I am really confused
Or was that vero?

Idk tbh
@neon garnet cleooo I’ve made up my mind
I think I’ll start jogging / walking everyday
It has sooo many benefits it’s crazy
Edit: I did some research ima go with running even ehheheheheheh
@jaunty ibex
what does remind you of CA
oh lmao

$$\frac{d}{dx}\left(\ln\left(x!\right)\right)=\gamma - \sum_{n=1}^{\infty}\zeta\left(n+1\right)\left(-x\right)^{n}$$
Cyberist
Why, did I have to bump into zeta, ughh.
@thorn wren You were talking about weird publishing methods. If and when I will totally randomly post some dumb desmos charts in here of some cool stuff I found recently.
Don’t start running every day if you’re not used to it
Walking every day is fine of course
Going from not running to every day running is asking for injuries
Unless I’m reading this wrong and you’re experienced
+1
This might sound really stupid, but it is meant more curious. Injuries from running? I mean yes you can get sore legs and/ or fall on the ground. Can you really get "injured" from running? I mean you're muscle won't magically rip apart, no?
yo chat
Overuse injuries in knee/shins are very common in running
In the shin?
How does that work out
Shin Splints
it does what
That’s the name
oh
…
lmao
oh lmao
i have that
when i play basketball
._.
though i haven't really thought about the bones etc.
that's good to know
Yeah
Yeah that’s how you wanna do it
at what point do you "go up" in training time?
like what's a good rule of thumb
for jogging/ running ofc
Don’t really know I already did it from a pretty young age
hm
It depends on how well your body is adapted already
not gud
You’ll find information on this if you search for it actively
Im just a random dude on discord
You can find professionals talking about this
?
Biking for example
They combine nicely as well
oooh
Yea I’m gonna start like 3-2 times a week and I won’t RUN run more like jog
Yea so I won’t start sprinting everyday without proper training 
Oh you like running?
Running feels so good to me for some reason
Well I haven’t started yet…
But yea generally I do enjoy running
When I do run
But I haven’t taken it as a habit or anything
I’m trying to do so tho
👍

altho the graph im tryna interpolate kinda looks like this
which idk if thats gonna be well approximated with a cubic
but we'll see
yayyyy
that's such a great decision nona!!
i love running. i love how i feel when doing it. i learned recently that your diaphram is a muscle that's important to train too for running a lot.
i'm super curious to hear your journey!! running is def a topic that interests me. i feel excited for working up to a long, pain free running experience, for me c:
i hear running 👀
@jaunty ibex verooooo we miss you girl 💖
hope you're welll
Running feels so good for some reason
she'll be back when deltoid remembers the password
pff 
Yayyyy so Pilates and running sounds like a nice calming combo
I wanna like acc be able to run without running out of breath loll
@latent edge so disorganized tsk tsk /j 
omg pilates is so peak
omg yessss
i love pilates
Sameee
Even tho I still shake
Through almost the whole thing
As if I was electrified bahhhahha
it's so funny. like, my knee hurts too much to run more than a few minutes, so i feel limited by my musculoskeletal system more than my cardiovascular system for running right now. i hadnt expected to experience that before 
thats fine, tbh the most important thing is having good form and really activating that mind body connection so u activate the right muscles
Oh well ig ima experience both lollll
I’m starting tmrw
super truee
also i learned about the TVA??
lower back on the floor!! (if mat pilates
)
Yea quality over quantity
you go girll!! 
dont forget ur pelvic tilts
Yayy
its js so much better, also slow movements burn more LMAOO
yep yepp
deep core deep core

Although I can’t start running right away bc my body isn’t used to it and it’s also dangerous
So I’ll start with walking and then maybe switch to running
Idk I need to read more about this matter
For realll
you can explore this too! [1]
they value preventing injury a lot
i could do that lowk.
Ooh tysmmm
mhm!!
always start slow and low
Mhm I will
i know some ppl start and immediately go "i wanna run 10kms"
nah thats way too much

Fr
u have to build up to it
i nearly C4'd my heart and after four years of recovery i was able to slowly complete a 10km recently
Oh that cool be a cool goal
yayyy congrats ^^
im not into running but i do cheer u on fr
i believe in you folks!
mhm fr. my friend ran one recently. so inspirational
That's impressive ngl
ngl, thankfully i'm still young. if i was older probably would have been harder
Let's rename discussion 2 into running
kek
Running is very healthy for your heart!
you are correct!.... but not for my condition
i have to always be mindful of high blood pressures
@unborn meteor asked to be reminded <t:1753645130:R>, do this
Woww impressive
took a lot of physical training under medical guidance tbh
uhh do you mean literally C4'ed
cause that's kinda hard to do
it's a euphemism ofc
ah
but close to it
in a sense, what ended up happening could be described as a kind of c4-ing
i think its a recommendation to get 150 mins of steady state cardio a week
im tryna figure out how to get it in
Maybe try to split it up?
Start off by doing like a 30 min walk 5x/week while browsing something on ur phone, better than nothing
. Eventually you get the urge to try jogging a little in between and then it just gets more and more addicting and faster from there. That's what I did but with 1 hour walks.
i try to do 10k steps a day but its kinda hard
just in terms of time
hmmm 10k steps a day probably takes longer than 150mins of steady state cardio a week? Idk.
Maybe you could get a walking workstation? Admittedly I've never tried to do math on one, seems kind of hard.
you can have a walk by the park for 20mins at night so you can cross the 10k steps mark i used to do this, about 6k steps before walking the park
Same, except for the "like" part
its a gigantic public server
and i know the US government joins those in mass
so im both being spied on and am available to be seen by an audience of hundreds of thousands
whenever i do anything ever
What yall know about the pretty lines and colors in lambda
oh hi kirbs
lmfao
Hoii
This really wasn't the direction I was expecting this to go.
explain
Of all the reasons I could come up with to despise this server (and there are many), "being spied on by the government" would honestly be quite low on the list.
That’s fair
dude the colors mean nothing
it just helps keep track of wth is going on
But they pretty :(
ok
glam up your factorial function 💅
Your name is funny
Oh yeah
@junior turtle corporate needs you to find the difference between these two pictures
one is "with deliberate mistakes" and the other "clean and correct"
I dont understand lol
continuation of a discussion from #advanced-lounge
I asked chatgpt to generate a diagram like the one here https://math.stackexchange.com/questions/2245314/complex-keyhole-contour-integral
both are wrong 
im going to fucking explode
Not the wisest thing to say if you think the US government might be listening
nuh uh
lmao
is that a raven with rolly chair legs
pukeko
hello guys. I wanna study master degree major applaid mathamatics but which bacholovr subject important for this specialty/
What's up?


Hi zan
also fine, can't wait for my vacation in august 
I'll be at an anime convention this weekend
Afterwards, I'll visit my home country for 1 month since I haven't been there for almost 3 years

bnuuy indeed
oh yeah, I finished my phd if you didn't know already
Thanks! 
Prof. Zan 



Are you happy with your current job?
Yes
I was having a bad time in my masters and I thought it was really not worth all this mental health problems
I see
I think what really matters is that you found a satisfying job that you can live with
Yeah I'm happy with my current job. I just thought of doing masters as a break from my job but it was not worth it apparently


Gn peeps
Pinging you in here since I don't think advanced needs any more LLM convo @old oak
But, I recently saw some video on social media talking about how LLM ads present the weirdest, most alien use cases for 'AI'
"Hey meta, pick up [friend] from airport"
(how does the friend know what car to look for, which airport, etc. - just silly as presented)
Granted, there are some approaches using multiple "agents" to try to be able to do this work flow, but I think it rings true that these tech company try to manufacture these nonsensical uses
I don't know at what point we decided as a collective that we wanted to delegate our actual human lives to AI instead of just the boring stuff
Oh, another goofy one was some AI glasses commercial where the scenario was being in a museum and being interested in a display, but the display mysteriously has no info placard, so you use your AI vision to learn some definitely accurate facts about whatever's in front of you
Who is the target audience??
To be honest I don't need any more LLM convo, but I agree with all your points.
Fair lmao
And I'm not just saying this to get out of the conversation, it's almost all dystopia, bullshit, or dystopian bullshit, and I'm just so tired
For all my genuine interest in machine learning and AI, if it was possible to ban it all tomorrow, erase all the models and make it impossible to train more, I'd consider it a net positive.
based
Yeah I think we just need a full stop as well
At least its presence lets me easily filter out the companies that use it in their ads, etc.
AI stopped being funny when they started using it for technical interviews
yep
@haughty loom hi!!!!
oh no
@neat lintel conversation moved to here to not make mods mad
nevermind they're leaving
lets not continue it somewhere else
alright
The experience of going to a museum when you dont speak the local language
I honestly hadn't thought of that tbf
anyone here
hi
hi
hi
hi
bye
hi
To ask for mathematics help on this server, please open your own help channel or help thread. See #❓how-to-get-help for instructions.
welcome to the mathcord! see #❓how-to-get-help to open a help channel 
Yes? I'm fine with pings, but other helpers may not; just a notice.
Also, I don't see your help channel yet 
are you asking for study tips then?
or book recommendations #book-recommendations
yeah, #book-recommendations might be the best place to go
if you don't have a specific math question, then don't open one
happy to help!
@limber haven welcome to the mathcord! 
welcome!!
welcome! :)




Do u like snorlax
i do like snorlax!!
snorlax is my favorite
that's a great favorite!!! my fav is chandelure :]]]
hi, I'm new to this channel. I'm starting A-level maths and additional maths after the summer and I have a somewhat descent knowledge of maths after taking additional maths last year. I would be happy to help with anything I am able to but my knowledge is currently limited so there may be some things I need help with or don't understand
if u need help with math feel free to open a help channel in #❓how-to-get-help
and if u want to help ppl, there is a "Math Help" category where it says "Occupied" u could help people there
welcome to mathcord btw 
thank you
I LOVE ADO
im listening to her colab with imagine dragons
based
its so good dude
indeed
yall
hello
nothing much
just learning abt making money using crypto
I just joined this group 😄
thank you
Welcome to the server @neon gyro
unc why are u here rn
YSST KNOWS WHICH PERSON IS WHICH

hi
It's over for you vro.
No.






