(co)Brantley (co)Vose

Two strange polygons.

Four people standing in front of a conference poster. From left to right: Yicheng Lin, Mohamed Musa, Brantley Vose, Alex Theis.

About 30 years ago some mathematicians came up with some theoretical drum shapes that solved a long-standing problem. Some undergrads and I tried to make them!

Wrote a short blog post about it:
prismika.github.io/2026/03/01/w...

Got sweaty palms watching the end.

I think I gasped out loud. The slide was meant to overwhelm you. I think that's fine. It perfectly communicated what he wanted to say and added to the story, even if you weren't expected to read the expression term-by-term.

I do think slides like this can be done well. I once saw a talk where the speaker was an excellent storyteller. He described the process of reducing his problem to showing that a certain polynomial was non-negative. He then showed the polynomial, which surprisingly covered the entire slide!

A crude modification of a famous tweet by @screaminbutcalm. The modified text reads: writing a convex relaxation: Haha fuck yeah!! Yes!! Analyzing my convex relaxation: Well this fucking sucks. What the fuck.

Playing Cards with Bayes' Theorem Wizard 330 is a home-brewed variant of Wizard, a trick-taking card game. (Think Spades or Euchre. That sort of game.) While strategizing...

Whipped up a Python script to help me play cards.

brantleyfightsfunctors.blogspot.com/2025/12/play...

Nice! Though to be fair, we *could* say that n=1 is the base case (f'=f') and the difficulty gets pushed to the inductive step.

I propose that the *true* center is the mean of every center currently listed in the Encyclopedia of Triangle Centers.

kisonecat.github.io/triangle-cen... lets you view a ton of triangle centers from faculty.evansville.edu/ck6/encyclop... all at once

So what you're saying should be possible unless I'm missing something.

My gut says the triangle is already determined by the information given without the semicircle. The fact that the point on the hypotenuse nearest to the opposite vertex is one third of the way down the hypotenuse should determine the triangle up to similarity, the the 2's provide scale.

I bet there are a lot of things that people think are American but are actually just movie quirks.

A video game:

Final Fantasy (there were in fact 15 more Fantasies)

perc fireworks!

Someday a confused programmer is going to visit this thread to see why their scraper is looping infinitely, and I hope they read my comment.

Holding rocks from two celestial bodies, climbing rocks on a third.

Someone once told me that the reason so many mathematicians are climbers is that, even in our hobbies, we prefer to face the walls instead of each other.

N-SPHERES YouTube video by Jerobeam Fenderson

If you haven't seen it before, you might like oscilloscope music! It's music specifically designed so that its oscilloscope visuals go with the music.
www.youtube.com/watch?v=BDER...

Sick. Any good tricks for making things sound so punchy?

A conversation with ChatGPT. A user says "Do you know why I keep winning?" The model responds "Well, you do get to hear what I’m going to play before you make your choice—so you have a bit of an unfair advantage. Essentially, you’re countering my move with the perfect response each time. If we both had to choose simultaneously, there’d be more of an element of chance. But since you know my move first, it’s almost impossible for you to lose!"

This is a fun test! o1 passes, but only after I directly ask.

Will this discussion be posted anywhere?

Haha that's delightful!

I hadn't thought of it this way. I think of productivity tools (like coding assistants or debuggers or whatever) as ways to save me from things I don't want to be doing. You seem to be thinking of them as economic gadgets that are benefiting (primarily) your employer, not you. That's interesting.

Yeah, maybe the term "boring tasks" is conflating "things I'd rather not spend my time on" with "low-stress tasks".

I enjoy occasional mundane tasks. I think they're psychologically important. At the same time, if a tool can save me from tasks I hate doing, I'll take it.

I might have misunderstood your original post. I thought you were saying LLMs don't make programmers more productive, but maybe you were actually saying that productivity at work is not a proper goal?

Maybe, but I think that's a rude thing to say.

An intentionally low-effort meme. A running skeleton is surrounded by white text that says: "JUST WALK OUT you can leave!!! work social thing movies home class dentist clothes shoppi too fancy weed store cops if your quick friend ships IF IT SUCKS... HIT DA BRICKS!! real winners quit"

I saw this years ago and it is the least serious meme that has ever changed my life philosophy.

Yeah, I'm surprised this turned into a discussion around the claim "Coding assistants are bad because you should be doing more boring tasks at work." There are objections to LLM coding assistants that are worth discussing, but this one is pretty odd.

A scatterplot of various stories, with one axis describing the date of publication, and the other describing the year in which the story is set, whether past or future.

I have to mention XKCD's incredible chart on this topic. xkcd.com/1491/