The last Magdalene laundry closed in my lifetime. Homosexuality was decriminalised in Ireland in my lifetime. Access to condoms without a prescription in Ireland only came in 1985. We repealed the 8th. There's a lot of ways in which Ireland is still bad, but it is an entirely different country.

Constantine@ajaxsinger.bsky.social · 16h ago

Food, man. Food is so much better than it was in 1986. All of it. Poor-people food, restaurant food, fresh food, frozen food. Food is just a hell of a lot better.

Y a-t-il eu de la part de climato- ou météorologues des explications précises sur les causes de la sécheresse et des blocages anticycloniques répétés, que subit l'Europe cet été? (⚠️ Avant de dire «c'est le changement climatique», lisez la suite, merci.) Précisions: 🧵⤵️ •1/11

Wesh les histoirologues, je me demandais si l'an 1000 de l'Hégire qui est tombé en 1591 avait donné lieu dans le monde musulman à des mouvements messianiques/millénaristes? Z'avez des infos là-dessus?

Exemple d'instance de la loi de Campbell: ‣ Les cyclistes sont moins dangereux et plus vulnérables sur la route, donc on leur donne des facilités adaptées (p.ex., franchir un feu rouge sous conditions). L'ennui est que ceci dépend d'une définition légale précise de «vélo»: …

I asked on MathOverflow whether the topos of sheaves on Baire space ℕ^ℕ satisfies Choice from ℕ^ℕ to ℕ: mathoverflow.net/q/513942/17064 (I thought I knew this, but either I forgot or I had confused myself)

Does the topos of sheaves over the Baire space satisfy choice from $\mathbb{N}^{\mathbb{N}}$ to $\mathbb{N}$?

Question: Does the topos $\mathbf{Sh}(\mathcal{B})$ of sheaves on Baire space $\mathcal{B} := \mathbb{N}^{\mathbb{N}}$ (with the usual product topology) satisfy the following Choice principle (★) b...

mathoverflow.net

The Biclops

Discussing math with Mistral's “Le Chat” AI is actually very instructive, but not for the reasons you'd think. It doesn't solve your problem, it makes a lot of mistakes and gets itself confused all the time, so you end up having to explain to it why it's wrong …

One thing I only just realised is that sharks are old enough - not just marginally but by millions of years - to have eaten trilobites. This is somehow more odd feeling than the fact they pre-date Saturn's rings.

Glasgow's subway may have some of the smallest trains in the world but its network map has one of the largest graph automorphism groups.

The only growth industry left in America is in White House staffers making increasingly bespoke AI slop specifically to maintain Donald Trump’s airtight information bubble

Aaron Rupar@atrupar.com · 2d ago

Trump: "We have photographs or tapes -- like moving cameras, right? -- where people are on the side cutting it with a box knife."

This place’s medium term survival (at this point) depends on X going badly wrong along the lines of the various incidents in the period just post EM’s takeover — note that *all* the big waves here (with the possible exception of the open to all wave) have occurred due to X related issues.

There's something I like to call the “middle ground meta-fallacy” where, when political (or otherwise) discourse becomes so polarized, any attempt at nuance is seen as traitorous to one's “side” and is accused of “middle ground fallacy” and/or “bothsidesism”.

Marie Le Conte@youngvulgarian.marieleconte.com · yesterday

Musk has managed to poison the well to such an extent that we can now only think of platforms as "The Woke Factory" or "you will receive abuse from misogynists, racists and fascists and you will like it", and that's not great - hopefully Bsky does manage to create something in the middle!

Musk has managed to poison the well to such an extent that we can now only think of platforms as "The Woke Factory" or "you will receive abuse from misogynists, racists and fascists and you will like it", and that's not great - hopefully Bsky does manage to create something in the middle!

Easy mistake to make: struct whatever args = { … }; pthread_create(&t, NULL, func, &args); return; Can you spot it?

The distance Finns naturally take between each other puts them just out of reach of an axe. Coincidence or not?

I think (unironically) that you can generalize this “silent majority has a nuanced opinion but is afraid to speak it because they will face the vocal extremist minority” theory to every divisive topic on any social media, and it works fairly well.

One of Russia's most effective techniques to destabilize the 🇪🇺 is to encourage people to try to migrate to Europe, while simultaneously supporting far right political parties that depict such people as invaders and will blame the EU for “lax” border policies.

Sven-Erik Volberg@volberg.bsky.social · 2d ago

The Moroccan government confirmed that the Ceuta migration wave was organized and amplified through social media. A similar campaign is currently being run on TikTok, where people from Afghanistan, Pakistan, and Iraq are being encouraged to migrate through Belarus to EU via the Baltic states. 1/2

This is funny because, in the algebraic geometry context, regular⇒normal and the converse fails, whereas for topological spaces (where the words have completely different meanings), as you well know, normal⇒regular and the converse fails.

Until we overthrow the shackles of capitalism, mankind is doomed to continue to remake Spiderman movies until the sun at last envelops the earth and ends our suffering.

Release the tension patreon.com/berkeleymews

“Le commissaire australien à l'eSafety a enterré un rapport montrant que l'interdiction des médias sociaux aux moins de 16 ans n'a pas réduit l'utilisation des médias sociaux chez les adolescents, mais a abaissé la vigilance parentale. Nous avons fait des concessions sur notre vie privée pour rien.”

Ben Schwarz@benschwarz.bsky.social · 2d ago

Australia’s eSafety commissioner buried a report (released Friday afternoon) that shows the under 16 social media ban didn’t tangibly reduce teen social media use, and also resulted in parents tuning out of kids usage. We traded our privacy for no reason. theconversation.com/more-than-80...

… Donc je ne dis ❁pas❁: «tout usage d'un LLM pour résoudre des problèmes de maths est forcément néfaste», je dis «ce que fait actuellement OpenAI conduit (malgré les apparences) à une régression de la science». … •7/8

Oh yeah, I should add this: if OpenAI wanted to be loved by scientists all over the world and ACTUALLY BE USEFUL TO SCIENCE, they would make a grant-writing and general paperwork-filling agent freely available, thereby lifting an idiotic mental burden researchers waste a huge amount of time on.

Nicolas Audebert@nshaud.bsky.social · 2d ago

I can't do actual science when I have to spend half of my time writing grants. I need time to think by myself, to read, experiment, ingest and digest information. It's unfortunate but science takes time. Advancing humanity's collective knowledge is bounded by our meek human brains. 5/15

I'm sure some people will argue I'm a luddite, and e/acc weirdos would say I'm a monster, invoking grotesque utilitarian arguments on why we should submit to the singularit. But I don't think I am anti-AI. I do use LLMs. Hell, I'm even fine with using agents if that's what you like. 14/15

I can't do actual science when I have to spend half of my time writing grants. I need time to think by myself, to read, experiment, ingest and digest information. It's unfortunate but science takes time. Advancing humanity's collective knowledge is bounded by our meek human brains. 5/15

Funding agencies, and now weird billionaires with delusions of grandeur, would like us to be productive. Turn labs into science-producing factories. We've played this game for many years now, outputting papers, optimizing h-index and citation counts. Promise to solve moonshot problems. 7/15

This is a good piece of journalism on a serious topic, but the bit where the French journalist 'disguises himself' as an American (sportswear, baseball cap, USA branding) had me 😂