kash patel’s general demeanor reminds me of an animal raised for meat

thinking about how knowledgable maintainers have to pay off the expertise debt created by LLM-assisted code hackaday.com/2026/01/26/t...

From @acyn.bsky.social (posted on X) "Reporter: Do any of you have a favorite animal? Child: My favorite one is a gold snake that can move. It has gold eyes, and it has a super-duper tail… Reporter: Mr. Mamdani, the second question for you. Mamdani: Yes. It’s also the golden snake."

very normal news updates

Happy Ratboys album release day 🐭 If you've never heard this band, are feeling down (who isn't?), or miss that sensation of a spark burning inside of you, give this album a spin. It's one of those records that keeps on giving the more you play it

It’s called R

Hey, so, I don’t do civil litigation, is it generally considered good when the judge does this? storage.courtlistener.com/recap/gov.us...

working on one-leg double unders

you can’t have a classifier with less than 50% accuracy: groundhog edition www.tiktok.com/t/ZThPjSAe7/

This reminds me a lot of one of my favorite Steven Gould quotes: I am, somehow, less interested in the weight and convolutions of Einstein's brain than in the near certainty that people of equal talent have lived and died in cotton fields and sweatshop

S A D F A C E E M O J I

embraced the monad of it all audit_wrap(data) : put values in the bubble audit_peek(x, fn, ...) : log fn(data) but don't update data audit_poke(x, fn, ...) : log fn(data) and set data <- fn(data) audit_unwrap(x) : pull data out of bubble

i bought the album anata by @jcc-pf.bsky.social on bandcamp joshuachuquimiacrampton.bandcamp.com/album/anata

i once made a higher-order function l(). it takes data x and function, calls f(x, ...) stores the expression of f, stores the results of f(x, ...) but returns x. i made it to run and log a series of validation functions. does this have a (cool) name? i think it's a tee function that logs?

in real terms, #7 would give states leverage to put an end to surges, #8 would give states the ability to outright stop construction of new detention facilities, #1 and #6 would slow the rate at which DHS could train and deploy officers, #2 and #10 would likely shrink the pool of recruits

You get the sense these Jeffries and Schumer saw the ICE murders and mostly felt embarrassed for how upset their constituents were going to get

it’s genuinely reassuring to me that this is the most miserable and paranoid person in the entire world

That video about Chicago style pizza except instead of saying “the cheese is under the sauce”, they’re saying “just round the data”. I am very normal. youtube.com/shorts/o6XCp...

This picture is going to stick with me for a while.

inspired by CLAUDE.md, I’ve started putting markdown files named after coworkers into work code repos so I can remind them to stop doing shit to the codebase that annoys me for some reason they’re all mad at me now, which means ill be adding commands to JEREMY.md for an attitude adjustment

We compared intelligibility scores from in-person listeners vs. listeners recruited on Prolific. In-person listeners had higher scores. (No surprise.) Screening out listeners with catch trials helped improve reliability. We had to recruit 360 participants to get 300 that met screening criteria.

dplyr 1.2.0 is out now and we are SO excited! - `filter_out()` for dropping rows - `recode_values()`, `replace_values()`, and `replace_when()` that join `case_when()` as a complete family of recoding/replacing tools These are huge quality of life wins for #rstats! tidyverse.org/blog/2026/02...

Oops. repo! github.com/eitsupi/arf

there's a certain genre of guy (and it is mostly men) who's good at X technical thing but assumes that means they're good at ALL technical things "data grifter" is a good term for this when it's someone who's good at building ML models but confidently makes pretty basic causal inference mistakes

what now

I see many folks generate hypotheses exactly in this manner (w or w/o the AI) and design whole studies around something they came up with as they scroll the TL, and it's treated as standard scientific practice. We should be able to do better than this crackpot when it comes to hypothesis generation.

New Epstein files say that Steven Pinker was paid $10,000 to help Epstein’s defense (even though Pinker previously said he wasn't paid).

we developmentalists are a small minority of scientists, but had the NIH reorganization gone through, the entire child development wing of NIH (that is, the NICHD) would have basically been decimated. so yeah, this counts as a win in my book

the rating scale was a slider with tick marks on the 10s

anyone want to come over? we can stand across from each other and do this

What real heroism looks like is you drive past a school and it’s 7am and it’s -3 out and there are four dads posted at every door scanning every passing car to be sure it’s not someone coming to take the children. And you drive by at noon and they’re there. And you drive by at 3pm and they’re there

# tweak to avoid running both branches `%?%` <- function(test, yes) { t <- function() test y <- function() yes list(test = t, yes = y) } `%:%` <- function(lhs, no) { if (lhs$t()) lhs$y() else no } if (TRUE) 1 else 0 if (FALSE) stop() else 0 TRUE %?% 1 %:% stop() FALSE %?% stop() %:% 0

zooming in was more of a treat than I anticipated

Don't you f**king dare.

🚨 GAMs have moved on—so it’s time for an update. On March 3, 2026 (17:00–19:00 CET) I’ll be livestreaming an updated introduction to Generalized Additive Models in R 📺 YouTube livestream link: youtube.com/live/A9U8e1K... #RStats #mgcv #GAMs #gratia #statistics 🧪

they should make a movie series about how it’s irresponsible to use technology to mess with the natural order of things for profit

Stop playing "diagnostics ping-pong" with your users 🏓 Wrote about why every R package wrapping external tools needs a sitrep() function — with real examples from meetupr and freesurfer. #RStats https://drmo.site/F5gVjR

Frivolous post: "Parallel strands between learning crochet and learning programming" 🧶 masalmon.eu/2026/01/26/a...

I for one detest groundhogs

the names in stopifnot() become error messages for failed assertions, so... basename(dirs) |> is.element(data$parent_dir) |> all() |> stopifnot("Every folder name should appear in the table" = _)

I’m being told this is real.

this is your mission. our pedophile president needs you to wake up at 5 am and don 30 lbs of tactical gear you bought off amazon to sit in a chevy tahoe in the parking lot of the elementary school where you kidnapped a five year old kid to intimidate his teachers. good luck soldier

Unfinished. #LiamRamos

From the Epstein documents: The FBI having to Google where Wisconsin is???

He’s been through so much. They shouldn’t have made him fly the plane too.

This is maybe the most bone-dry Chotiner interview I've seen in a while

here's my entry!! #Comics4Liam

a 30-point swing isn't a flip, it's a Sunset Flip Power Bomb

“organizm” he’s so dumb