Let me recommend grabbing a cup of tea & reading this really thoughtful piece from the inimitable @hogg.bsky.social on moderating AI/LLM use in astrophysics. 🔭
"the real question we face is not ... how we do astrophysics. It is the question of why we do astrophysics."
📖: arxiv.org/pdf/2602.10181
1. This weekend, a federal judge permanently blocked funding threats to providers and hospitals that provide trans youth care.
The judge also blocked "any similar policy."
Hospitals have NO excuse and must return to providing care under many blue state laws.
Subscribe to support our journalism.
After 150 years of boring wooden Galton boards, I finally fixed it!
Now with lasers, bazookas, bombs, pinball bumpers, and extremely serious sci-fi hamsters.
🎯 EXTREME GALTON 👇
sachaepskamp.com/extreme_galton
I am excited beyond description to lift the veil on what we have been working on in 2026:
Please meet ggsql! A new extension of the SQL language for creating visualisations using the grammar of graphics. Read all about it in the blog post or visit the website at ggsql.org
The preprint is now available: osf.io/preprints/ps...
Thanks to the PsyArXiv moderators for their speedy review!
Freely available to any who want to read about the many ways in which the oft-cited literature on desistance spectacularly fails to support the claim of widespread desistance
Check out our preprint: "What Pilot Studies Can (and Cannot) Do for Validity in Psychological Research"
Great job @yashvin.bsky.social and @mbneff.bsky.social for leading!
doi.org/10.31234/osf...
The desistance rates generated from these studies are not functionally usable for informing policy or practice. From a quantitative perspective, the data are incoherent. From a contextual perspective, the underlying assumptions are indefensible (21)
This literature is so small and imprecise that one can essentially pick whatever number they want and find it within a prediction interval; this is not what scientists expect from literatures with strong evidential signal. (12)
Table 2 from the article demonstrates a wide range of possible desistance and persistence rates
We were astounded to find that meta-analytic prediction intervals from original sample of 11 articles supported ridiculously wide ranges of desistance (37% - 100%) and persistence (0% - 80%). The range for both expands to 0% - 100% (and beyond) when the new articles are included. (11)
I am so happy that this has finally seen the light of day. On the one hand, it seems an obvious answer to an obvious question, but it needed to be done.
And I'm never gonna start harping on the bit about the Spice Girls. Ever.
Literatures with strong evidential signal are not this flimsy or fragile!
If you are familiar with the claim that most transgender youth will "desist', please read, and consider sharing.
@catherinesjwall.com, @ghawinriver.bsky.social and I are very proud of this work.
Dunno what to tell you Hadley: the leaders of our military have been pretty transparent about our genetic engineering aspirations since at least the late 90s...
Racoons are a good start.
@pingfanhu.bsky.social has an amazing 3-part blog series on agentic engineering workflows:
Part 1 - Claude code basics: pingfanhu.com/blog/2026-03...
Part 2 - Workflow and examples: pingfanhu.com/blog/2026-04...
Part 3 - Software that ties it together: pingfanhu.com/blog/2026-04...
New paper, out this week in PLOS One, suggests that most close relationship self-report measures are primarily capturing relationship quality 🧵
journals.plos.org/plosone/arti...
My book Courtroom Science and Trans Youth is out TODAY, just in time for Trans Day of Visibility!
It’s about how courts treat science and scientific expertise in family law cases involving trans children, and how we need to do better and stop medicalizing and gatekeeping them.
apparently we can ban abortions but we can't ban conversion therapy 🙃
happy trans day of visibility to all of the incredible trans people in my life. the world is better with you in it and I'm never going to stop fighting for you, because your fight is my fight. we're in this together.
Care grows when we choose to share it—it doesn’t shrink. #TheAcademicActivist #TransgenderDayOfVisibility
carlosrd.substack.com/p/transgende...
Don't watch the new HP series and (further) enrich a known billionaire-bigot: www.theverge.com/report/90181...
Don't watch a sportsball production that sells out on its stated principles and values: www.olympics.com/ioc/principles
Put your money/time/energy into people/orgs that are for everyone.
This thread made me circle back to our Table 2.... Exposure for Specific Phobias just leaps off the page as a behemoth Tx, with impressive credibility across basically all the metrics we looked at, at the time.
FWIW, I think our work was a better take on credibility (vs. efficacy, though we look at that a bit too). And that's just 1-2 ways of thinking about scrutinizing the evidence base of a tx. Others, like inclusivity, accessibility of communication, also hold tx's back: www.nature.com/articles/s44...
The "BBQ Sandwich" I ordered at a DQ when visiting my sister in Illinois, sometime in my teens. Had never ordered food there before/didn't know that was a thing, so was thrown and "panic-ordered". It was soft and saucy, and to this day I haven't a clue what, exactly, were the BBQ'd contents.
Millennials looking at this like…
+1 for this workflow. Coding burden is low (and frankly, very automate-able if that's your preference). Fully reproducible; updated version won't replace older if anything errors out. Transferable skill/technique, whether synthesis, R pkg development, generic website, etc.
Rodent's Revenge
Resharing again! I'm looking for a postdoc with experience or interests in longitudinal data collection OR analysis: indiana.peopleadmin.com/postings/32132
a black and white edit that reads "the future includes trans people". around the text is a floral pattern. the image is grainy
the future is for trans people
The United States slaughtered school children in Iran, aided by AI.
