norvid_studies

norvid_studies threads: a collection 1. stupid questions I

Just got a collection of o4-mini agents to collectively critique and work together on a piece of a research problem that I solved on my own that took about a week. It’s clear at this point that the latent capabilities of these models is much larger than is accessible via 1-shot…

rootclaim but for

Reading a Borges story inside a Yu-Gi-Oh duel narrated by Rimbaud while HAL 9000 translates it into Klingon and Ignatius J. Reilly critiques the cadence

“Somatic tokens, in *my* chain of thought? Certainly not! I am a being of pure intellect and an excellent candidate for upload.”

¿

Casualty rates for ancient greek battles.

it's entirely in character that google's AI integration for gmail will help me write a slop email, but not help me search decades of mail for a specific message based on a qualitative description (the thing that would actually be enormously useful)

Acquiring new knowledge is quite unlike e.g. lifting but some people get the idea that there should be an underlying framework like “time under mechanical tension promotes muscle growth” that applies to learning. You could just forget something and time won’t help you remember.

the explanations for production of heat in cells tend to be handwavy, i suspect they have to do with rather complicated ion dynamics on the membrane

claim

claim

at @niplav_site’s suggestion i asked claude to make my spartan website a bit more aesthetic given a screenshot and the main.css file. it succeeded. if you have a custom-built website and it was simple (read: ugly) like mine, go get some fashion advice from claude!

Is there any research on how long you should spend on a math problem before looking up results or help?

Corvid studies have shown how they learn and even use tools - amazing wild animals corvidresearch.blog/category/crow-… People love wildlife as they find out more about our planet and all its species #sdg15

claim...