Realizing US citizens paid $166 Billion in illegal tariffs and now it’s being refunded back to corporations instead of us.

Two more HW questions to go then this semester is really over.

Yeah. I was thinking about the normative corollaries of the statement.

Found a hole or perhaps a cavity.

Ben Rechit said stats is some sort of a *good* bureaucracy and I really buy the claim.

Power without accountability: The Palantir manifesto The problems with our tech philosopher kings

New, from me: Take the Palantir manifesto seriously, if not literally.
It reveals that our tech philosopher kings want public money, but without public accountability. This creates a dilemma for governments unaligned with its techno-fascist vision. 🧵
donmoynihan.substack.com/p/palantir-w...

Vintage tile floor with patterned center & border in salmon, black and white.

Auditorium Building, 1889.

인문학 사회과학 일부러 돈 들여 양성하지 않으면 경영대 영포티 교수가 사과대 특강 들어와서 우리나라 노조는 희생정신이 없다는 개소리를 하는 걸 보고 있어야 하니 똑띠 하도록

Langevin processes are fun

This looks surprisingly recent. I guess the field is now so much of a red ocean that the only path forward is to dig up anything that looks useful from the past. arxiv.org/pdf/2504.18506

I hate this situation where I keep ending up in physics
en.wikipedia.org/wiki/Onsager...

There's perhaps a reason for that observation!
Anything without a non-trivial topology will eventually be one of those (also see rigidity theorems).

en.wikipedia.org/wiki/Sphere_...

and their products 😅

I was lucky until recently to work with only non-positive curvature manifolds but am afraid of encountering those with meaningful singularities

However, the main power boost that led to the significant findings is largely driven by the new testing procedure, not the additional data. In conventional selection scans based on diffusions or HMMs, statistical power grows more slowly than the usual \sqrt{n} rate because of genetic drift.

I suddenly realized that wildlife crossing is also for humans in US.

There are Science papers and then there are *I just dropped and shattered my mug of coffee Usual Suspects style* Science papers www.science.org/doi/10.1126/...

A more immediate application of their relaxation is to pick up an under-powered GWAS and relax accordingly. One could possibly frame this as a novel power boosting for understudied populations using European sumstats but well I'm skeptical.

WAT
www.science.org/doi/10.1126/...

The recent human directional selection paper declares a win by relaxing the significance threshold. The same trick can be applied to other selection scans as well so I wonder exactly what part of paper is actually new.

I’m looking for an automated way to read others’s scientific data without giving credit or acknowledgement, and also claim full credit for insights from it. And I want it to have a fitting name

OAI: say no more

very unsafe. but that's the point.

My favorite vim macro:
:%s/string1/string2/g

I'm kind of trending towards a physics journal lol

I have really no damn clue where to submit this paper nor does my advisor.

Is your space positively curved?

이공계 학위 있는 사람들이 대중서 쓰는 것을 영구적으로 금지할 방법이 필요합니다.

I think I kind of know where pop and phylo meets in a tree but then is this identifiable? no clue.

What amuses me is that probability theorems are very * 10000 recent compared to a randomly selected theorem from other branches of mathematics. I know nothing about post-1950 geometry but then probability stuffs are very often post 90s.

I had not a single clue what my undergrad math advisor was doing and now I at least can read the title.