"[…] books are all that matter. Podcasts, films, and TikToks are good at attracting ears and eyes, but in the realm of ideas, they punch below their weight."

https://www.experimental-history.com/p/text-is-king

Today's terrible python code is sponsored by: me!

I managed to get very different numerical results depending on whether I add a space to an unrelated string or not.

Anyway, convex optimization is fun.

Do we have such a thing as a nerd book canon? Is there a definitive list of books every nerd has read?

Maybe it's simplifying things too much, when we talk about "nerdiness" in such a general way, but like for sure every train fanatic has also read Lord of the Rings at some point of this dabbling into different super niche fields of interest, right?

#reading #nerdculture

For all who know their way around linear algebra and machine learning basics, Gregory Gundersen posted an interesting blog post about the history of Large Language Models. I found his thoughts rather refreshing.
He finds that the scientific breakthroughs happened before GPT3 and how scalability is frustratingly superior to complex model architecture.

https://gregorygundersen.com/blog/2025/10/01/large-language-models/

„Erklären ist leichter als Nachdenken.“

Ursula K. Le Guin, Immer nach Hause, S. 403