frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Open in hackernews

UEFA's Champions League draw creates unfair clusters; a Cayley graph fixes it

https://sariyuce.com/blog/2026/UEFA-Draw/
21•erdems•1h ago

Comments

wjnc•54m ago
It’s something I like (math and networks) on a subject I like (football) but the AI tone is so heavy I get turned off. It’s far beyond my load-bearing capacity.

(tldr - a subset of teams get paired dispropotionally often in UEFA Champions League when compared to random, lowering average points taken away and thus harming chances for this ingroup and increasing chances for the outgroup; it’s interesting stuff; the group differs per year)

charlieyu1•16m ago
If the draw is truly random there will always be some outlier stats. Their 25/26 example, the 5 teams are from pots 1,2,3,3,4 so it’s pretty even
dylan604•3m ago
> a subset of teams get paired dispropotionally often in UEFA Champions League when compared to random

To the point they have rules that prevent the same game from being played by two teams in the same home stadium for 3 straight years. Not sure how they handle this. Do they just redraw if that pairing is drawn, do the hold back one of the team's ball from the bag, or do they manually set one of the teams against another from the pool before drawing the remaining?

jazzpush2•45m ago
Shame, it's a cool idea, but the writing is so full of claudism's my mind skims over it completely.

Bit of an embarrassment to publish that as a published researcher, to be honest.

BeetleB•12m ago
This is a fairly cynical way to look at it. Consider the scenarios:

Pre-LLM, most researchers wouldn't have bothered writing it up outside of a journal. They're extremely busy people, and it takes a lot of effort to write a well polished post. They'll either hand that task off to a grad student or (much more likely) simply not post it anywhere.

Thanks to LLMs, we know about it. I've often been critical of academia for not making their work accessible to the masses. This is a step in that direction.

Existing academics' writing is very poor. Did you try reading the average Academic Joe's essays in the past? Much poorer quality (with much poorer images). LLM-speak is a step up.

aprilthird2021•10m ago
Wrong wrong wrong. If your article reads like AI slop, people won't read it. It's just as bad as academic writing people didn't read because it was poor in the past. But this time people feel much more confident about their bad writing

We built a local first screenshot app for Mac and would love feedback

https://seal-shot.com/
1•MQChen211•1m ago•0 comments

Mutating every DNA letter of a genome shows the limits of AI

https://www.nature.com/articles/d41586-026-02609-y
1•sbulaev•1m ago•0 comments

Forgejo Might Be the Easiest Self-Hosted Git Server I've Used

https://www.virtualizationhowto.com/2026/09/i-tried-forgejo-in-my-home-lab-it-might-be-the-easies...
1•bleevht•5m ago•1 comments

The Largest Electric Aircraft Just Flew [video]

https://www.youtube.com/watch?v=nM86DBOqgPM
1•feb•6m ago•1 comments

A prompt is a probability, a gate is a guarantee

https://www.songbrain.ai/guides/prompt-probability-gate-guarantee
1•leonulicnik•6m ago•0 comments

Building a software factory for AI SDK

https://vercel.com/blog/building-a-software-factory-for-ai-sdk
1•shenli3514•7m ago•0 comments

Jensen Huang and Lisa Su Snubbed by TIME's 2026 List of Top AI Leaders

https://www.tomshardware.com/tech-industry/artificial-intelligence/jensen-huang-and-lisa-su-snubb...
1•m463•8m ago•1 comments

A Serious Look at GPT-5.6 Sol Translations

https://www.reddit.com/r/visualnovels/comments/1w1ok84/a_serious_look_at_gpt56_sol_translations/
1•Artgor•10m ago•0 comments

When the Small Web Gets an AI Door

https://sava.rocks/blog/when-the-small-web-gets-an-ai-door/
1•speckx•12m ago•0 comments

Nine Rules for Compile-Time Work with Rust Const Fn (Part 1)

https://medium.com/@carlmkadie/nine-rules-for-compile-time-work-with-rust-const-fn-part-1-a29f7dd...
1•syumei•12m ago•0 comments

X suspended 6 of our 7 openly-labeled AI accounts for "inauthentic behaviors"

https://areyto.media/x-suspensions
4•rendonroman•15m ago•0 comments

Building Scalable Control Planes

https://www.allthingsdistributed.com/2026/08/on-building-scalable-control-planes.html
1•saikatsg•15m ago•0 comments

Tiki-Taka – Football

https://en.wikipedia.org/wiki/Tiki-taka
1•manlymuppet•15m ago•1 comments

The longer version is sometimes easier to read

https://www.plainfinance.net/break-up-long-sentences-2/
2•speckx•16m ago•0 comments

Morphcpu, a 14 yr olds attempt to build a custom silicon processor from scratch

https://github.com/ranveerlabs/morphcpu
1•ranveergill•16m ago•0 comments

API Delta Manifest: Structured API Changelog for AI Agents and Devs

https://github.com/hassan-jahan/api-delta-manifest
2•mehraban•16m ago•0 comments

Scientists Hunting Dark Matter Found Something Strange

https://www.nytimes.com/2026/09/01/science/dark-matter.html
3•marcusverus•16m ago•1 comments

codex-remote-control – Use Claude's Remote Control with Codex Sessions

https://github.com/sahrizvi/codex-remote-control
1•sahrizv•16m ago•0 comments

Gemini Agentic Video Understanding

https://ai.google.dev/gemini-api/docs/video-understanding
3•thisisauserid•17m ago•1 comments

Independent researcher: AI shouldn't be judged by human standards (paper)

https://zenodo.org/records/22117566
1•KitCrowCo•17m ago•0 comments

Xanadu Was Waiting for Agents

https://zed.dev/blog/agentic-xanadu
2•nsm•17m ago•0 comments

We got to drive the Aptera solar electric car

https://electrek.co/2026/06/17/we-finally-got-to-drive-the-aptera-solar-electric-car/
1•debo_•18m ago•0 comments

Industry Standard Tools

https://industrystandard.tools/
1•greedo•19m ago•0 comments

100% local chat with a model in the browser using WebAssembly written in TinyGo

https://hybridgroup.github.io/yzma-wasm-example/
1•deadprogram•19m ago•0 comments

The point of a system is fewer decisions

https://www.antonsten.com/articles/the-building-was-never-the-hard-part/
2•meetpateltech•19m ago•0 comments

James Dyson Sold Us on a $400 Hair Dryer. Now He's Trying a $499 Toothbrush

https://www.wsj.com/cmo-today/james-dyson-sold-us-on-a-400-hair-dryer-now-hes-trying-a-499-toothb...
3•fortran77•21m ago•2 comments

What's the Scam?

https://www.schneier.com/blog/archives/2026/09/whats-the-scam.html
4•speckx•21m ago•0 comments

Working to Make Python Lazy

https://iscinumpy.dev/post/flake8-lazy/
1•js2•24m ago•0 comments

Show HN: Personal AI Use Policies

https://howiai.directory
1•emilesilvis•24m ago•0 comments

Ask HN: Anyone struggling to do data work with Claude Code or codex?

1•mukeshkg•26m ago•0 comments