Paper defines theorem interestingness as proof length divided by statement length
tak3sh8 · x · 2026-10-11
A recent paper defines the "intrinsic interestingness" of a theorem as the ratio between the length of its proof and the length of its statement — the longer the proof packed into a short statement, the more interesting.
Commenters quickly noted the punchline: by that definition, famously short statements with enormous proofs are in fact "extremely interesting" 😅. The thread also riffs on David Chapman's question about what it would mean culturally to discover that most of math is, by some measure, stupid and ugly.
More from Fun
- Meme: "The voices in my head: Claude and sol 6.1" — Gauri_the_great · 2026-10-12
- A superintelligence in everyone's pocket — used to decode 'hey' vs 'heyyy' — signulll · 2026-10-12
- David Senra didn't come out of nowhere: his Founders podcast started 10 years ago — sahilypatel · 2026-10-12
- Yoav Artzi clashes over AI drug discovery: 'understanding' maps to FDA approval — tallinzen · 2026-10-12
- Mathematician jokes OpenAI's 3 sign errors—an odd number—break math's delicate balance — basedjensen · 2026-10-12
- Musk jokes the best AI sandbox is a Delta flight — no internet access at all — elonmusk · 2026-10-12