April 13, 2026
One function to rule them all?
Mathematical Minimalism
One function + the number 1 = all of math? Commenters are losing it
TLDR: A new paper claims all basic math can be built from one special function plus the number 1, sparking a firestorm. Fans love the elegant simplicity and dream of exact neural networks; skeptics argue it’s clever theory with little practical payoff given messy hardware, training, and numerical stability limits.
The internet just met its new math crush: a paper claiming you can rebuild every basic math function using only a single special function (called “eml”) and the number 1. That’s right—addition, subtraction, multiplication, division, even constants like π—built from one function plus 1. Cue chaos. The big thread is already exploding on Hacker News.
Commenters split fast. One camp is swooning over the elegance—calling it “math minimalism” and “one function to rule them all.” The other camp claps back: it’s a cute parlor trick that won’t change real-world computing. The spiciest fight is about AI: the paper hints that stacking this single function could give exact results (as exact as exponent and logarithm are), unlike the usual neural network claim that they can “approximate anything” in theory. Fans call it a game-changer; skeptics say it’s not practical—real hardware isn’t perfect, training is messy, and stability could be a nightmare.
Meanwhile, the meme brigade is out in force. “Minimalist devs be like: I only need one function and 1,” joked one. Another: “Boss: use fewer functions. Me: say no more.” There’s even debate over how hard it is to actually compute this function precisely. Verdict? Gorgeous theory, questionable impact, maximum drama—and the comments are the real show.
Key Points
- •An arXiv paper by Andrzej Odrzywolek shows all elementary functions can be derived from a single function (eml) and the constant 1.
- •The paper’s supplement provides equations to build addition, subtraction, multiplication, and division from the eml function.
- •Constants like π and functions such as square, square root, and standard circular and hyperbolic functions can be obtained from eml and 1.
- •The article emphasizes that composing eml and 1 gives exact constructions, subject to the exactness of logarithm and exponential.
- •This exactness is contrasted with the universal approximator theorem’s epsilon-delta guarantees, suggesting implications for deep neural networks.