« first day (5308 days earlier)      last day (5 days later) » 

03:03
@BenSteffan Did someone say Borromean?
Nov 8, 2022 at 23:22, by Akiva Weinberger
user image
 
5 hours later…
07:57
@Soham Saha.
@vigilantvino02 hi
Hallo! Can you just help me how to answer @SohamSaha
Answer what exactly? @vigilantvino02
@PM2Ring Cool sage code! :)
Any kind of answer. I have read your answer and i want to answer in the same way as how you did. Can you tell me how to answer any kind of questions which follows the policies of MSE? @SohamSaha
This covers all points I think
08:06
One thing.. I shouldn't copy the answer from GPT and paste here right? YOu must have seen my previous answer, where i didn't copy the answer from any AI yet others say i did. What should i do? @SohamSaha
Others can say and do whatever they want. Being true to yourself should be enough. This is the current MSE stance on Gen-AI content: math.stackexchange.com/help/gen-ai-policy
I just use webs like GPT, Grammarly, edurev and so on to get the reference for my answer and to also check if i did in the correct method so that the asker won't follow the wrong path.. Is this wrong? I won't copy the content from that web, i just frame it in my own words.
but that leads to my bad reps right.. They think i copy and vote mine down.
Mention which parts are genAI content, and ensure that the core elements of your post are not AI generated. I for myself am strongly against AI content, as they are really error prone in mathematical context
But that’s just my opinion
Even reworded content should be referenced to the appropriate sources
08:25
Oh okay bro! Thanks for your input.. I will try to give my level best in my next Maths answer!
@vigilantvino02 You should be ok. Just don't post text generated by AI. However, some recent grammar checking tools can use GenAI to rewrite your text. It's probably not a good idea to use such text, even if the program isn't actually changing your ideas.
Such tools make you sound like GenAI. And people may not believe you if you say, "Oh, it's just a translation / grammar tool!".
Oh. So, i will just try to gather my inputs from these webs and make the answers with my own words.. Thanks @PM2Ring and @SohamSaha
You can put a note in your answer like: "I used ChatGPT while doing research, but this answer was written by me, and cleaned up using a non-AI version of Grammarly".
Oh okay sir.. I am afraid that saying this might make people believe i lie and this might make me earn bad reps among us..
@SohamSaha Thanks! Since I wrote that code, I figured out a superior way to create an elliptical torus:
But Chat won't let me post it. Hmmm
Here's a temporary short link. sagecell.sagemath.org/?q=geeeya
08:36
Wow @PM2Ring. Cool mate!
:-)
 
2 hours later…
10:38
More fun with parametric surfaces: a coiled catenary tube: sagecell.sagemath.org/?q=arkqpm
@PM2Ring ok so you're now orienting the circles at a plane perpendicular to the tangent to the ellipse right?
The 'stretch artifacts' are gone, looks even better than the prev
@SohamSaha Correct.
@PM2Ring this reminds me of one of my questions
Yeah. I originally thought it was impossible to get rid of those stretch marks. Until I figured out that trick of aligning the circles.
@PM2Ring Cool trick
10:49
I did some experiments with coiled coils. But you can't go very deep (with good resolution) without needing a lot of points.
@PM2Ring would need to do something that uses the self-repeating pattern somehow maybe... Something like oimo.io/works/life
Try zooming in, it just goes on and on and on...
A simple toroidal coil.
@SohamSaha Ah, the good old Unit Cell. There's now a more efficient version of that. But the original was mind-blowing when it first burst onto the scene.
Catenary coiled coil. Obviously related to the previous thingy. :) sagecell.sagemath.org/?q=yulqve
The OTCA Metapixel was in 2005, or so the Wiki says. I wasn't around at that time :)
I used to play with Conway's Life quite a bit. I didn't make any major discoveries, but I did find a few useful patterns.
@PM2Ring And I have no idea how to even start finding useful patterns. Do you just start with a random soup and see what happens?
11:01
And I did build the first Life pattern that calculates Fibonacci numbers in binary. conwaylife.com/forums/viewtopic.php?t=81#p281
@SohamSaha It's very rare to find useful stuff from random soups, but it does happen.
Nov 24, 2024 at 4:28, by one potato two potato
"Every closed, oriented 3-manifold $M$ contains a link $L\subset M$ such that $M-L$ is homeomorphic to a finite sheeted covering space of the Whitehead link complement"
@PM2Ring that's seriously cool
@PM2Ring so how does one search?
one of the motivations why (I think) people consider Whitehead link besides for toy example
Are there any articles about finding interesting patterns?
Just a few years a new infinite growth pattern emerged from a random soup survey. The Conwaylife forum people have been doing a distributed exploration of 16×16 random starting patterns. Of course, it will take eons to completely investigate all 2^256 starting patterns. :D
@SohamSaha Thanks!
11:06
But I only know the Borromean ring as a toy example for something.
<Mental note: Marking PM2Ring as GoL expert. Would pester for further projects>
@SohamSaha One way is to collide gliders or space ships into various "still life" stable patterns, or into small period oscillators.
Ok...
@PM2Ring Gotta go now. It was nice talking with you. Bye :D
11:25
Consider the definition of the exponential function as a power series, namely $$E(z)=\sum_{n=0}^\infty\frac{z^n}{n!}.\tag1$$One can derive $$E(z)E(w)=E(z+w),\quad E(z)E(-z)=E(0)=1,\tag2$$and then Rudin claims that $E(z)\neq 0$ for all $z$. How does this follow from $(1)$ and $(2)$?
Ah wait. After some thinking, I guess it is because if $E(z)=0$ for some $z$, then $0=E(z)E(-z)=1$, contradiction.
 
1 hour later…
12:44
fun fact: If $M$ is a closed oriented surface with a constant negative curvature w.r.t a metric $g$, then $L_{X_1}g = L_{X_2}g$ for vector fields $X_1,X_2$ on $M$ implies $X_1 = X_2$.
@onepotatotwopotato You and I have very different notions of "fun".
13:34
@onepotatotwopotato it's the simplest example of a non-trivial link with linking number zero
@psie yes
 
3 hours later…
16:37
"the Bousfield lattice is not a lattice" ah yes why would it be
I have a linear transformation $L: R^n \to R^n, y = Lx $ such that $y_i = x_i \forall i \ne h,k$ while $y_h = x_h+c x_k, c \ne 0$ I have a rectangular closed cuboid $I = [a_1,b_1]\times ... \times [a_n,b_n]$ and I gotta prove that $m(L(I)) = m(I)$
@BenSteffan me having a moment yesterday where I realized the elementary acyclic cofibrations were not cofibrations
I tried the simple case $h=1,k=2$ and I got that $$L(I) = m_2(A) \times \prod_{3}^{n} [a_i,b_i]$$
where A is the rectangle $[a_1,b_1] \times [a_2,b_2]$, whose height remains unchanged
(well, they were injective cofibrations, but the paper specifically needed to consider the projective model structure)
16:51
so I have $m(L(I)) = m_2(L(A)) \cdot \prod_{i=3}^{n}(b_i-a_i)$ which is not the result I wanted to see :p
If I have an exact functor F: Sp \to D, where Sp is the \infty-category of spectra, and D is stable, complete and cocomplete, there exists a regular cardinal \kappa such that Sp is \kappa-accessible, F: Sp_{\kappa} \to D is k-accessible, where Sp_{\kappa} denotes the \kappa-compact objects, and taking Ind-completion recovers F? This seems not true in this generality but I have seen (from my understanding) it claimed somewhere
Like if F is not exact why should it be \kappa-continuous for any \kappa? Here \kappa-continuous means that it commutes with \kappa-filtered colimits
17:12
ok I got it lol, the area of $L(A) $ is simply $(b_2-a_2)\cdot (b_1-a_1)$ so everything's fine
 
3 hours later…
20:26
In Rudin's PMA, he says that there exists a unique $t\in[0,2\pi]$ such that $e^{it}=z$, where $|z|=1$. Then, when proving the algebraic completeness of the complex field, he claims there exists real $\theta$ such that $e^{ik\theta}b_k=-|b_k|$, where $b_k$ is a nonzero complex coefficient of a certain polynomial. I simply do not understand why such a $\theta$ exists?
$k$ could be any natural number between $1$ and $n$, for some natural $n$.
Well, write $b_k$ in polar form...
...and use that $e^{it}$ is periodic.
Also you're misquoting Rudin, for your first statement.
yes, I have a typo :) ok 👍 sometimes I get so rigid in Rudin's definitions, I forget a complex number has a polar form
 
2 hours later…
22:40
@Jakobian I never realized that I've always seen the $K_\sigma$ notation exclusively in descriptive set theory books/papers and not really in topology ones. Funny

« first day (5308 days earlier)      last day (5 days later) »