Patrick Blanchenay

Any sufficiently advanced file-drawer is indistinguishable from p-hacking

It's hard to maintain websites. /s

I’m not worried about AI taking away all the jobs. As long as someone needs to check the “I am not a robot” box, humans will be needed.

At the turn of the millennium, 2.2 billion people in the world lived in extreme poverty. In international statistics, this means they survived on less than $3 per day (in today’s money).

In the two decades that followed, this number more than halved. You can see this decline in the chart.

In turn, the journal builds a reputation for offering high quality peer-review, and being hard to publish, therefore offering high quality work.
You can finetune the cap so that it's only binding for mass producers.

Wouldn't it be easy for prestigious journals (as a signal of prestige) to actually cap the number of submissions by a given author over, say, a year? Authors have an incentive to only submit their best paper there, which require them to think self critically instead of mass-producing.

Pretty sure this is someone who SHOULD have used AI.

Instead, it sounds like what you get when you just use next word suggestion on the Android keyboard.

Very complete.

If the costs of refereeing are also going down.... Someone should create the Journull.

One potential upside as the cost of writing a paper goes to zero: null results might finally get written up

Can't make an evolutionary omelet without breaking a few eggs.

Just a pointer that Copilot can actually remove passwords on certain Excel files. Worth trying to avoid the whole unzip, clean, rezip.

Authored? Produced? Generated? Published?

Over the Top: when Stallone turns the hat backwards

How I feel turning on "extended thinking"

Aid organisations in disarray.

i love data, me too meme

Wheat Field with the Alpilles Foothills in the Background - 1888
https://botfrens.com/collections/46/contents/14510

Meth cooking is a healthcare failure. #breakingbad

Santa climbs in a window, sees a letter written to him from a small child who is sleeping. The letter reads “observe the counter factual”. After reading this Santa tries to shoot the child.

🎅

Screenshot of the "Does that use a lot of energy?" online app

Hannah Ritchie has built a fun little tool where you can compare energy usage of various products and activities.

This is super helpful imho, because it's so hard to develop intuitions even just about the scales involved here.

hannahritchie.substack.com/p/does-that-...

Kirill, whoever you are, I owe you one.

I heard there was a secret code
That purged effects of survey mode
But you don’t really care for bias, do ya

The actual solution is to take causal inference seriously in research designs; and weigh the possibility of misinterpretation strongly in the decision to accept a study or not.

I mean sure, but this is not enough; many still interpret associational and predictive language as causal.
If you write an article showing that "blueberry consumption can predict incidence of cardiovascular diseases", news outlets and Instagram influencers WILL tell you to eat more blueberries.

Email from LinkedIn titled "View Princeton University jobs and your next steps".

I hope you all find someone in life who believes in you the same way LinkedIn believes in me.

It's not lying if it doesn't what is true or not. It's just "trying its luck".

In French, Claude Code is pronounced the opposite way as it is in English (closed "o" sound first, open "o" second). (Unless you're in the North, where both "o" are closed, or in the South where both "o" are open.)