John Muschelli

Bayes Rule in a Super Bowl ad: ✅

If Jeff Bezos could afford to spend $75 million on the Melania movie & $500 million for a yacht to sail off to his $55 million wedding to give his wife a $5 million ring, please don't tell me he needed to fire one-third of the Washington Post staff.

Democracy dies in oligarchy.

I support term limits for Congress, the Senate and SCOTUS.

Dr Kareem Carr man: i wish to publish @kareem_carr Jan 21 reviewer 2: your paper is no good man: i'll do anything to improve reviewer 2: it's simple. you must read the work of the great scientist Pagliarini man: *bursts into tears* but i am Pagliarini Andre Pagliarini @apagliar Jan 21 a first: in rejecting an article I submitted to a journal, reviewer 2 noted I failed to engage the work of one Andre Pagliarini Jan 21, 2026 • 3:47 PM UTC

I just thought everyone should see this

I just woke up from a nap and somehow while I was asleep, everyone on the bus has figured out we are not going to the right place

A bar chart showing the electricity use of several daily activities with the subtitle "The 'typical query' is not a useful way to think about coding agents' energy use." The bar for a 'typical ChatGPT query' is not even visible. My median Claude Code session is somewhere between the average US household per minute and toasting bread for three minutes. My median day with Claude Code is something like running a dishwasher.

Whenever I read discourse on AI energy/water use that focuses on the "median query," I can't help but feel misled. Coding agents like Claude Code send hundreds of longer-than-median queries every session, and I run dozens of sessions a day.

On my blog: www.simonpcouch.com/blog/2026-01...

UX design so focused on optimizing for a given user task, while completely ignoring:

- goal achieved, i want to get back to where i was

- i want to immediately do goal again

- i want to switch to different goal

Way too many times is 3-4 clicks/touches just to... do more stuff with the same app.

why I have trust issues

reddit post from a presumably young person who watched sex and the city and couldn’t get over how much people smoked indoors. “wouldn’t everything smell bad?” they asked

this is a good reminder of how things that are shitty but have been accepted as normal and inevitable can actually turn out to be abnormal and evitable with some science, time, education and political will

Redacted Epstein files

Redacted Epstein files. Basically completely blank

Redacted Epstein files. Basically completely blank

Redacted Epstein files. Basically completely blank

Journal article: “the data are available on request”

The data:

Whenever I teach this, I like to remind my students that #rstats was informed by a lot of statisticians and NA means “I don’t know” so most logical tests and operators return “I don’t know”. x> 5? I don’t know. I have to know x first.

Yes, exactly. Each missing value is equal to itself and unequal to other values. In contrast, R's NA is neither equal nor unequal to anything, including itself

From a senior faculty this morning: "As you get older you become 'former'. I'm the former chair of ..."

It's a good time of year to unsubscribe. A lot of happy holiday messages are showing me what lists my emails are on

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

April 14 #ChatGPT prompt tsunami: “ fill out my tax return that maximize my refund with the lowest probability I get audited”

If students and post-docs are workers, then academia may have some of worst turnover rates in any industry.

How long until the AI Chatbots start inserting ads in the responses?

This is an ifixit email of an iPhone 6 user saying: iPhone 6, black screen problem solve With the tagline under saying was this useful?

AI teaching iPhones to fix themselves:

Still on Rstudio but I have it on machine. Does it simply go via the diff or does it read the whole repo? We’ve been discussing Claude and codex but they don’t seem to respect ignore directories (e.g. data)

Does AI write commit messages yet?

Jobs with biggest % decline in job postings include lots of creative jobs. And *biostatisticians* 😬

These are thought to be most impacted by AI. Great job everyone.

bloomberry.com/blog/i-analy...

“There’s more open imaging data about listening to music than menopause and contraceptives effect on the brain” #nexus @russpoldrack.org

Trump has only ever really cared about one issue: Email storage.

Best case: The President of the United States knew about a pedophilia ring and did nothing to stop it.

Medium case: He also used his office to protect the perpetrators.

Worst case: Because he was one of the perpetrators.

This is one of the most remarkable academic debacles I've ever seen.

A large RCT got published in BMJ. There are currently 44 Pubpeer comments, mostly about the data, including...well. Read for yourself.
pubpeer.com/publications...

Birds Aren't Real - Wikipedia

Birds aren’t real en.wikipedia.org/wiki/Birds_A...

I used to feel some type of way when people would write `library(tidyverse)` in #rstats scripts. I've written a lot of packages so it seems like a dependency hell (which it is for pkgs). Although introduces many dependencies for the script, I've come around to making scripts quick to make.