math

Doing a bit of conjuring today

scala> type Something = Nothing => Nothing
// defined alias type Something = Nothing => Nothing

scala> summon[Something =:= Nothing]
-- [E172] Type Error: ----------------------------------------------------------
1 |summon[Something =:= Nothing]
| ^
| Cannot prove that Something =:= Nothing.
1 error found

#dev #tech #software #math #scala #TypeTheory

A lab mate shared this write up of Don Knuth using LLMs to solve a math problem: https://www-cs-faculty.stanford.edu/~knuth/papers/claude-cycles.pdf

It's clear that using Claude did help them arrive at some new understanding here, which is wonderful. I'm happy for them.

However, I'm upset by how much they personify Claude and attribute the solution to "him."

From this narrative, it's clear that the humans were very actively involved from beginning to end. Claude was a helpful tool, but it did not solve this problem on its own. What role did it actually play? How was it like or unlike a human collaborator on this problem?

It did generate a crucial insight, but where did that come from? Was it plagiarized from some unknown source? Did it "just emerge" from text completion and interpolation in latent space? Do we need some other explanation for Claude's apparent creativity?

These folks don't care. They just wanted a solution, which they attribute to Claude, and leave it at that. I think that's a serious problem.

#llm #ai #science #math

Did you ever like the groove of a music track, starting to dance to it but then you stumble because the beat seems to reset itself randomly? I've got a bunch of those tracks this time in

RADIO IRRTUM! !!

so happy dance lurching :)

--- ↓↓↓ Deutsch Version below unten ↓↓↓ ---

The next issue of #RadioIrrtum! will arrive this Saturday, 2026/04/11, 8:00pm UTC+2/Berlin time/CET. On Alex Berlin at DAB+ K12D in #Berlin/Brandenburg and worldwide @ https://www.alex-berlin.de/radio-livestream .

A German language #radioShow exploring the sonic vastness of the #underground. Emitted transmissions may raise attention for new unheard #music. Don't worry about the #GermanLanguage aspect - this is about #music!

So what's in it?

- #AlternativePop
- #offline Music
- cluttered #Beats
- super #slowBass with DnB elements
- #UKGarage
- #fingerdrumming
- #ByteBeat from #Math formulas
- #microtonal #MathPunk
- #Punk of various flavors (Post~, Garage~)
- #livingroom music from Brazil and Lebanon

On top I will pronounce words wrong at an embarrassing amount, & play stuff that isn't available at #Spotify, which simply shows how much this service sucks.

Besides unusual rhythmic structures there will be examples of microtonal music (explained even!) – if you followed the music news the last weeks you may already know which project will be featured.
Featured are #FediverseMusicians as well, like @ScottHuehnerkrisp or @herrkaschke .

But a very special one among those #fediartists will be @stephan.
This is probably your ONLY chance to have a listen to one piece of his offline album "Strategies against Algorithms" if you don't already own it. It will only be played during the aired show, it won't be contained within the archive version! You know: That's what "offline" is all about.

I'll post playlist + manuscript later here. Having questions? I'll gladly answer. Just ask. CU all on Saturday! :)


--- ↑↑↑ English version above, the Deutsche welche jetzt ↓↓↓ ---

Radio Irrtum! kommt näher → Samstag, 11.04., 20:00 isses soweit auf Alex Berlin (DAB+ K.12D / Stream URL oben).

Also die seltsam funke(ln)de Radiosendung, in der planetenumfassende Weiten des musikalischen Untergrunds erforscht werden.

Mit:
- Alternative Pop
- offline Musik
- Bass / Garage
- Fingergetrommel
- ByteBeat aus Mathematik
- microtonaler Mathepunk
- mehr Punk
- Wohnzimmerpop

Viele Sachen davon gibz nich auf Spotify, und Fediartists sind auch dabei (s.o.); vor allem Schall und Stille: Da wird nur in der Ausstrahlung was vom aktuellen spektakulären (!) Offline-Album gespielt; aus meinem Sendearchiv ist es rausgeschnitten. Offline muss Offline bleiben!

Details in englisch oben, Playlist/Manuskript nach der Show hier drunter. Fragen einfach stellen – ich antworte!

Bis zur Sendung… vielleicht? :)

#SchallUndStille #fediArtists #musicRecommendation #MathInMusic. Radio Irrtum! stands for: #NoAi #NoAiInMusic

RE: https://stefanbohacek.online/@primestamp/116800542675238476

Giving this one another go.

Very niche, and still a bit noisy with an hourly posting frequency (nowhere near as noisy as when it posted every single time it found a timestamp).

So, uh, please enjoy, I suppose!

#math #primes #PrimeNumbers

did you know that Fibonacci numbers (where each number is the sum of the previous two) can convert miles to kilometres?

1, 2, 3, 5, 8, 13, 21, 34, 55, 89, 144, ...

pick any number in the sequence as your miles, and the next number in the sequence approximates the number of kilometres

this works because the ratio of consecutive Fibonacci numbers converges to 1.6180334 (the golden ratio φ), while the number of kilometres in a mile is 1.609344

to see how close this is, consider the approximation compared to the actual kilometres

miles km actual km

1 2 1.61
2 3 3.22
3 5 4.83
5 8 8.05
8 13 12.87
13 21 20.92
21 34 33.79
34 55 54.71
55 89 88.50
89 144 143.20

#math

"While Buckmaster and Alpöge were finalizing their own results, they learned that “information about our progress had been passed to OpenAI.” When they contacted OpenAI, they were told that OpenAI had already achieved a full proof of the central problem. But when they asked follow-up questions about when OpenAI had begun its research into the problem and how much human input was involved, the answers became more evasive.

“It emerged that an entire team had been working on the problem,” Buckmaster said, “and that an insane amount of compute had been used…. Eventually, it was agreed that [the first prompt] had been sent in the past few days, after information about our work had reached OpenAI.”

If true, that would suggest the OpenAI team had become convinced that Buckmaster and Alpöge’s approach was the right one, and decided to use its material advantage in computing resources to reach a formal proof first."

https://techcrunch.com/2026/09/08/openai-fought-dirty-on-career-making-math-problem-says-nyu-mathematician/

#AI #GenerativeAI #LLMs #OpenAI #Astra #GPT #Math #Mathematics