4 of 5 stars to Proto: How One Ancient Language Went Global by Laura Spinney

still my favorite christmas post

tfw you have to be mary but it’s so late

≈4 dozen tamales + fried burritos for Christmas Eve #pacooks #poliscicooks

This whole thing is amazing, but tucked away in a dropdown at the end of the markdown is... quite seriously the best essay I've read on the environmental impacts of AI?

Merry Christmas from the Heiss family! Here's our 20th annual family newsletter by @nancyheiss.bsky.social, with links to all sorts of blog posts and extra features! Full PDF: files.heissatopia.com/Christmas%20...

Merry Christmas from the Heiss family! Here's our 20th annual family newsletter by @nancyheiss.bsky.social, with links to all sorts of blog posts and extra features! Full PDF: files.heissatopia.com/Christmas%20...

Confirmed: Bones is a belter

Wildly different things, tasks, techniques, subspecialties being lumped into "AI" and then being conflated with each other, doesn't help. Different types of models vs the techniques to train them vs the tasks they are supposed to accomplish, all under "AI".

Bread. ( #pacooks #poliscicooks #breadsky )

Some John Rutter for Christmas Adam

suspect a big reason why many academics and others who work in areas where getting facts RIGHT is key are disinterested in using LLMs for research: they’ve tried it, they keep noticing major errors in output, and they conclude that having to verify all that doesn’t actually save them time.

What are we doing here? www.science.org/doi/10.1126/...

Also, Hoshi Sato!

Per NY Times’s Michael Grynbaum on X, this is Sharyn Alfonsi’s email to her “60 Minutes” colleagues in full:

started Man on the Inside season 2 and it’s in a university this time??!

As a statistical hedge wizard I am really dependent on the econometricians to provide the right incantations, ritual conditions, and thaumaturgic items so I can sort of argue that my policy maybe probably works.

It’s the entire point. There’s no plan to make this meaningful for students. No plan to use it administratively. Its design and implementation signals its only real intent. It’s best and easiest use case is for systematic harassment and to scare faculty into complying in advance.

4.5 of 5 stars to There Is No Antimemetics Division: A Novel by qntm

This was fun! Josh O’Connor was so good

Time to wake up this dead man

Hey, I'm here! And I'm not the only one who's run into this!

I asked @aoc.bsky.social about why she posted this tweet, and she said “because JD Vance is a goober, man.”

Grading done, grades submitted, college daughter lands in a few hours—time to take a break!¹ ____ ¹ catch up on research stuff 😬

Hey, I'm here! And I'm not the only one who's run into this!

sombr is good

oh no

This is 100% infuriating.

Working with synthetic data is not a groundbreaking new idea or capability. Usually we do it to understand the behavior of a model or to test the capabilities of a method, and not to draw inferences about a real world population. When doing that, it is essential to know the assumptions of your DGM.

There *are* legitimate use cases for this kind of stuff! Simulation studies! Synthetic data for a preregistered analysis plan prior to an actual study! But this isn't being marketed as that. It's advertised as a cheaper, faster alternative to humans. Temu we-have-results-at-home survey results.

here’s a free version: neat_research_panel <- data.frame( id = 1:100, treatment = sample(c(TRUE, FALSE), 100, replace = TRUE), outcome = rnorm(100, mean = 50, sd = 5) )

what

🍌 Google's Nano Banana Pro is here, making spotting AI images even harder. We're at a point where anything you see online could be AI, which makes fact-checking your feed more essential than ever. 🎞 Our video: youtube.com/shorts/3M-BH... #NewsLiteracy #AILiteracy

20 (!!!) years ago today these little babies got married—that's 50% of our lives!

I might be convinced

👀👀

In Google Scholar's defense it's not listing these fake journals/articles as real, but it does list them in the full text results so you can see which real articles are citing things from them So real(?) things are picking up fake things so real people & LLMs take that as a signal that they're real

Grading and googling hallucinated citations, as one does nowadays, and now that LLMs have been around for a while, I've discovered new horrors: hallucinated journals are now appearing in Google Scholar with dozens of citations bc so many people are citing these fake things

Merriam-Webster’s human editors have chosen ‘slop’ as the 2025 Word of the Year.

The "smartphones/social media" discourse suffers from some amazing historical amnesia. There was no 2008 financial crisis and no global pandemic starting in 2020, it's all SCREENS SCREENS SCREENS. Major world events? Just the backdrop against which SCREENS happened.

Nothing will make you an education AI skeptic faster than grading some college take home assignments. Admins who haven’t been in a classroom in years push it as a learning tool. Students are laughing at them as they use it as a cheating tool. It’s all instructors are texting about with each other.

Extraordinary courage from Ahmed al Ahmed, a Muslim, 43-year-old father of two, who bravely risked his life to save his neighbors celebrating Hanukkah. Praying for his full & speedy recovery. And so deeply inspired by his example.

Another day, and another reference to these bangers. Contemporary causal inference has fundamentally changed the way I think about control variables and mediation analyses. These can show you why: doi.org/10.1177/2515... doi.org/10.1111/spc3... doi.org/10.1177/2515...

This -- a model bill to effectively decimate non-STEM research by requiring 3-3 teaching loads outside of "STEM or Americanism and western civilization" -- seems...extremely bad but also quite plausible.

oh come on

Burned out.

calling for a permanent moratorium on the word "filler" in all online reviews of tv shows (just watched the latest Pluribus and—phew—many online people are not happy)

oh no it's been Angel Studios-ified