@peterdkramer

Ordinarily Well, as reviewed in @TheAtlantic, theatln.tc/2qk6H1w: "an ambitious, persuasive, & important book"

Re-upping from last November: bit.ly/46Qk9kh

I've been remiss in not posting the new landing page for the audio book of my novel Death of the Great Man. More timely than ever... and now, at half price. audiobooks.com/promotions/pro…

@JamesGleick @KathaPollitt @Neko_Akage @kevincarey1 @goldberg_dan @LisaBirnbach @NormOrnstein @NYTimesWordplay @borkow1 @SEKramer @EWFIII @oliverbutler @RadicalSister15 @notinwordlist @nytimes @parisreview @thegridkid @sv1121 @OED @WSJ @JohnFetterman @washingtonpost @nypost @guardian @NewYorker @paulbloomatyale @jknotjk @kfaisteele @taffyakner @LauraMLippman @kvetchnewyork [Spoiler alert Feb 7 2026]: We've seen these letters 3x times before, & today is the first time that 2 pangrams are accepted. I wish that twitter/X had kept this long spelling-bee-related thread afloat—so that we could see whether contributors had suggested the second pangram.

@JamesGleick @DavidJManson @EllenFremedon @KathaPollitt @BCDreyer @Neko_Akage @kevincarey1 @goldberg_dan @LisaBirnbach @Digby_C @NormOrnstein @NYTimesWordplay @borkow1 @sekramer @EWFIII @oliverbutler @RadicalSister15 @wiffleclaw @notinwordlist @MerriamWebster @nytimes @parisreview @thegridkid @sv1121 @OED @WSJ @tinclock @JohnFetterman @washingtonpost @nypost @guardian @NewYorker @paulbloomatyale @mjohnharrison @jknotjk @kfaisteele @BcDryer @alicia_furiosa @taffyakner @DianeEnglander @LauraMLippman @kvetchnewyork Maybe the puzzle is just finding its level? As it gets more popular it needs to get simpler? I can see making a different choice a few months later. Or allowing one different, less standard word when a puzzle repeats. We're not worse off without donator & mandator & nondormant…

this is a family twitter but there is a certain very notable swear word missing from today’s word list #spellingbee

[note: newyorker.com/magazine/2022/…]

@parul_sehgal In which skepticism of the overused trauma plot becomes a meme. youtube.com/watch?v=SxReJx…

@DanielJDrucker Promising, but as the authors suggest, plenty of room for confounding by indication, where here "indication" includes not just MDs' reasons for prescribing but also patients' social class, sophistication of the prescribing doctors, etc.

Patients with #diabetes and #cancer who received GLP-1 RA experienced superior survival outcomes and reduced rates of hospitalization compared to patients receiving metformin academic.oup.com/jcem/advance-a…

@mkonnikova @RachelAviv FWIW: Anatole (not Anatoly) Broyard. Broyard's wonderful memoir, Kafka Was the Rage, is also slightly deceptive in one regard, but as it was published posthumously... hard to assign responsibility

@RachelAviv Link to piece: open.substack.com/pub/mariakonni…

Oliver Sacks was a hero of mine. I couldn't let @RachelAviv's brilliant piece pass without comment. My latest, on why Sacks's fictions are so much worse than those of your garden-variety journalist fabuslit. Link below.

Republicans’ promise to preserve Medicare didn’t last long. Their cuts just had to be well hidden.

@thegridkid I'm little slow on the uptake. Today's puzzle has a seasonal theme, one that justifies that S. Interesting how the single vowel and the central B moderate the S's power to multiply accepted responses.

@thegridkid Thought today must be some sort of anniversary. The second S.

Day 4 of OHHHHHH WAITTTTTT HAPPY SOLVING SSSSgetit

@RecoveryDoctor But this is a study where you can override randomization, and some participants chose meds, no? (Even *with* randomization, exercise studies can suffer from dropout bias—if people who favor exercise —& are more reslient—leave the no-exercise arm.)

@PeterDKramer Literally only people who are willing to exercise will sign up for a study of exercise. No one who is unwilling to exercise is going to even sign up in the first place.

You are being disingenuous for posting it because I know you know better than this. This is not really a "study". People got to choose whether to run or get antidepressants. THEY WERE NOT RANDOMIZED. Even the authors point this out: "Limitations A minority of the participants was willing to be randomized; the running therapy was larger due to greater preference for this intervention." It's a study of people's choices, not the so-called "intervention". And it's not the SSRI that worsens things, it's inherent in the participants. Hard to believe that the journal even published it.

@RecoveryDoctor Of course there is also the obvious confound—that people who are more ill (or too ill to contemplate exercise) choose medication treatment.

@RecoveryDoctor The link: pubmed.ncbi.nlm.nih.gov/18678794/

Republicans are saying that our opposition to Trump nominees is “unprecedented.” Seriously? Let’s take a look at some of the stuff that is really “unprecedented”:

@LHSummers Perhaps not the most serious, but serious all the same. Trump had a consistent policy of trying to take control of civic institutions that might act, or traditionally do act, independent of the Presidency.

Firing the head of a key government agency because you don’t like the numbers they report, which come from surveys using long established procedures, is what happens in authoritarian countries not democratic ones. This is surely not the most serious threat to our democracy that @realDonaldTrump poses, but it is the one closest to what we economists do. I don’t see how serious members of the economics profession can stay in his administration.

@DrDominicNg Also, the NEJM cases are summaries prepared to elicit discussions from (narrowly) relevant experts, not general physicians.The cases are complex ones that mostly worked their way up the chain to specialists—who did get answers (albeit on rare occasion, post-mortem).

@DrDominicNg The challenge has strange aspects. If AI (or a doc) asks for a test result and there is none (it's not in the NEJM text), it knows that it's looking in a wrong direction. In practice, if you ask for a test result, you get one—and it may lead you on a wild goose chase.

Microsoft claims their new AI framework diagnoses 4x better than doctors. I'm a medical doctor and I actually read the paper. Here's my perspective on why this is both impressive AND misleading ... 🧵

@StephenKing theatlantic.com/books/archive/…

Deep down in his heart, I believe Trump knows he’s an incompetent asshole.

@StephMachado The Francis Scott Key Bridge in Baltimore has about the same projected completion day, but of course, that project is much more complex.

BREAKING: RI state leaders say the Washington Bridge in I-195 west will be rebuilt by November of 2028. They just tapped Walsh Construction Company to build it bostonglobe.com/2025/06/06/met…