Doomscrolling mathematics



TLDR: I forked an open-source project to make MathScroll, which shows you an never-ending list of interesting mathematics content.

Sometime last year I stumbled on a small project called WikiScroll on Hacker News which is a small website that lets you infinitely scroll through Wikipedia articles and their featured images.

While a lot less engaging than Instagram or TikTok, there’s something about the infinite-scroll feature which still makes it fun to use. The only problem is that most of the Wikipedia articles are actually quite boring, unless you’re a big fan of lists of historical batting champions and milk.

Since WikiScroll is open-source, last year I decided to fork it and create MathScroll, which is almost exactly the same except that it only shows Wikipedia pages related to mathematics and science.

I also set up a shortcut on my phone that would hopefully replace the habitual opening of Instagram or Reddit, and this worked great! Even though I was still spending on my phone, I felt marginally better about it since I was still learning something new.

But there’s only so many Wikipedia articles you can skim-read about things you don’t understand before you get bored enough that you might switch back to endlessly scrolling through Instagram or TikTok. To try and combat this, I also added the top 2,500 questions from Math StackExchange and MathOverflow which ended up being a lot more interesting than any of the Wikipedia articles!

For good measure, I also added the lecture notes for all of the mathematics courses I’m able to take at university so that I can take a peek at any of them if they look interesting.

Some highlights

These are some of the most interesting questions I’ve come across while scrolling through:

  • “Fractal behaviour along the boundary of convergence”: This is something I’ve wondered since learning about power series in [[Notes - Analysis I MT22, Power series]]U and [[Notes - Complex Analysis MT23, Power series]]U. Every power series has a so-called “radius of convergence” $R$ where if $ \vert x \vert < R$, $f(x)$ exists, and otherwise $f(x)$ does not exist. But what does $f$ look like when $x$ is very close to $R$? This question is an example of a function where the behaviour around the boundary is fractal-like.
  • Squigonometry: The “normal” unit circle is only the unit circle in the 2-norm. If you change your definition of distance between two points, you get different unit circles – for example, if you define the distance as $d((x _ 1, y _ 1), (x _ 2), (y _ 2)) = \vert x _ 2 - x _ 1 \vert + \vert y _ 2 - y _ 1 \vert $, the unit circle looks more like a square, and for other distance functions it can be a squircle, sort of half way between a square and a circle. Interestingly, it turns out you can develop something analogous to trigonometry on these shapes.

Of course, there’s a lot of the standard Wikipedia/StackExchange fluff of people arguing in the comments about closing questions for being too open or a duplicate of a question asked 12 years ago with a completely different title.

Have fun scrolling!




Related posts