Every month the Quietus sends me the strangest damn music and keeps me up to date on topics of international interest from Eric Bristow to Zamrock. It's so great that I can forgive that there's also podcasts. So get yourself a subscription, and maybe one for a friend – if you really like them.

It's not 'social media', it's a perpetual emotion machine.

A gentle summary of the last twenty-odd years of Pearl-style causal inference with DAGs, perhaps most helpful if you're just arriving from old-school philosophy of causation. I like this genre. And if you do too, I can also recommend Weinberger et al.'s forthcoming piece in BJPS 1/2

"for all the points of the compass, there's only one direction and time is its only measure" Rosencrantz, Guildenstern, and Tom Stoppard are dead. www.theguardian.com/stage/2025/n...

The arc of the moral universe is two metaphors in a trench quote.

I have news. Wonderful news. New paper on #disinformation. The title is the thesis. Think of disinformation as content that functions to optimise the decisions of the audience for the disinformant. This doesn't require deception or content that induces false belief. philpapers.org/rec/LITDIF

In some impoverished and amoral sense, if you're sure your military will never fight an equal or more capable opponent then maybe you don't need to worry about war crimes. But if that's something you're sure of, you are also a fool. And a danger to your country.

#rstats It is with profound sadness I heard that my long-time friend and colleague, John Fox passed away this week. He was the author of {car}, {effects}, {Rcmdr}, ... and numerous influential books. I will miss him greatly. www.john-fox.ca

This belongs in a gallery

The revenge of the Marshmallow Experiment This seems like it might be a clean confirmation of the theory that 'irrational' preference reversals in temporal choice problems are due to marginalising over socially-calibrated uncertainty about whether the later event will actually occur.

Tariff-brain (US National Parks edition) Note to Americans: These things start with foreigners but don't stop with them.

Whether or not you think replication failures show that cognitive dissonance (or dunning-kruger or grit or...) is not a thing, the big question is: where do 'effects' belong in psychology? Some think they're just practically-derived motivating phenomena, like fire and rusting for thermodynamics 1/2

Monster: ...but you get to choose what. Me [thinking furiously]: Aha! Feathers. I'd like to be hit by a ton of feathers, please. Monster: You don't understand how weights works, do you?

Some of these Harry Potter sequels look surprisingly good

Apropos nothing in particular, I just got a nice new T-shirt (from the fine folk at @kyivindependent.com)

"we are not fighting fossil fuels, we are fighting the use of fossils for fuel" well, ok then...

"Beeindruckende Karte, aber eigentlich hätte ich gern das, was die da drüben haben"

May as well admit what you noticed already: I have absolutely no idea where to put commas, and more German is just making it worse.

The really annoying thing about "No, you don't understand, my LLM can summarize the state of the field for me" vs "No, you don't understand, LLMs can't summarize anything, they just predict the next token" is that each side can tell themselves it's the other side making the category error.

So you think to yourself, "Ok, I'll just go check in on the old place" and what do you find? A Home Secretary figuring out how to make asylum applications revenue neutral by confiscating wedding rings, and whatever the fuck this is:

Periodic reminder to authors that when people say they were inspired by what you wrote, you really never know what they took away from it.

"I'll just age faster if I leave this longitudinal analysis and you can't make me. I'm never dropping out. Just close the R session and walk away."

To be fair to Aristotle, some people do have radiators for brains.

I have time and respect for the religious to the extent words and beliefs turn into deeds. This here? Deeds. So, respect.

We are phasing out the terms “extant” and “extinct” from our exhibits and will now refer to species as either “Darwinners" or “Darlosers.”

This picture of why EU leadership is particularly susceptible to AI hype rings true to me. Extending @epopppp.bsky.social's view (in HTLE) of motivating governance metaphors, it feels like a successful bid to move elites from technical or cost-benefit thinking to a management picture.

*Some great looking pull requests appear from a developer in Virginia*

Y = [T=1]Y⁽¹⁾ + [T=0]Y⁽⁰⁾ I don't make the rules

So, this character just turns up in my building, jet black and rhythmically rocking the bird over our mailboxes. Sometimes I wish Berlin could stop being quite so Berlin for a moment.

Conjecture: If the Vietnam War draft is any guide, most of those exiting the US will go to Canada and an order of magnitude fewer to Europe. In this case, there's also US private industry which will probably soak up most wavering early career folk.

Kein besserer Start in den Morgen als mit einem starken ZEIT-Artikel meines großartigen Hertie-Kollegen Christian Gläßel und @ascharpf.bsky.social: www.zeit.de/politik/ausl... (Freebie-Link)

Someone alert Elon. This are the laws Bladerunner would have written.

The secret to not worrying about the ever increasing volume of research papers is the ability to evaluate a methods section for obvious horseshit. Turns out, not reading those papers frees up a lot of time that you can spend worrying about the appalling research waste.

Despite having written a whole PhD thesis on word embeddings as a psychological model of word representation, I would still rather it had the Box property: false but useful.

The Appointment with AI in Samarra

The slow but happy convergence of experiments & design inference and modeling & simulation strategy continues. Apropos, I'm beginning to see why @stephensenn.bsky.social is so enthusiastic about John Nelder (general balance & design inference) and also John Nelder (GLMs & hierarchical models).

You can wear a suit, you can be in a complicated relationship with a suit, or the suit can wear you.

See, this is what happens if you feed them

1 Minute zum Kaufen 6 Tage bis zur Lieferung 9 Monate bis zur Genehmigung durch die Eigentümergemeinschaft Wir können alles. Außer Hochgeschwindigkeit

Es gibt eine Schlange, und wir stehen drin.

There's another indirectly malign effect here. If you know what happened when the 'ban the box' campaign was successful and 'the box' was banned (Agan and Star, 2018), then you know what to expect when visitors think about who exactly those American soldiers were, who helped liberate Europe in WW2.

This is a useful interview for European students of English politics. Not only does it hit most of the tropes you'll run into in conversation, but it'll give a good idea of what is common knowledge there, what is surprising news, and the arguments that connect them. It's a mental landscape painting.

Just checking in on the Brits, and... oh good Lord

Ok, the anniversary was back in October and it's for a bank, but c'mon I'm looking at this in the former East and... it's pretty funny.

One thing I took away from Hacking's Representing & Intervening is that many theories of How Science Works don't survive contact with the historical record. Apropos here's an account of Franklin's role in discovering the structure of DNA. A lot going on that's hard to compress into a talking point.

They finally created it: the quintuple-barrel question.

"Home Office data suggests that recent immigrants are more likely to be working than British nationals and make higher wages than them." ✍️ Lauren Gilbert argues that migrants to the UK are net fiscal contributors, adding much more to the economy than they take out, ukandeu.ac.uk/lower-migrat...

Forecast intervals from my time series model be like

According to Bagehot, governments have efficient parts to do the governing and vibes parts to maintain – through stirring, aspirational though maybe not so rational claims and actions – an emotional connection to people who aren't otherwise much involved in politics.

At some point you might want to write your own geom_* for ggplot2. Here's an instruction booklet: evamaerey.github.io/ggplot2-exte... #rstats