iiinova — where your digital self keeps evolving

Ouf, je n'avais jamais fait le lien, mais j'ai toujours considéré le Matin des Magiciens comme le précurseur des théories du complot; Le livre est une curiosité de son époque mais porte en lui toutes les techniques des complotistes. Pas trop étonnant qu'elle le trouve génial.

000

Very soon, Trump will rename the USA United States of Super 'Rika, or USSR for short.

000

Article original: www.ledevoir.com/opinion/chro...

000

Lisounours @lise-l.bsky.social · 10/10/2026

Jean-François Lisée a fait une petite simulation intéressante sur le vote préférentiel. C'est certain que les électeurs auraient probablement voté différemment (plus besoin de vote stratégique) mais le résultat est surprenant quand-même: Au total, PQ gagne 6 sieges, PCQ en perd 1 et PLQ en perd 5.

Tableau comparatif vertical présentant les résultats simulés d'une élection au Québec dans sept circonscriptions. Le tableau est divisé en trois colonnes : « Circonscription », « Vainqueur officiel » (représentant le mode de scrutin actuel) et « Vainqueur selon un modèle préférentiel ».
Dans six des sept circonscriptions affichées (Sherbrooke, Vachon, Champlain, Montarville, La Prairie et Jean-Talon), la simulation d'un vote préférentiel change le gagnant au profit de candidats du Parti Québécois (PQ), indiqués par des cases bleues, délogeant ainsi des élus officiels de Québec solidaire (QS), du Parti libéral du Québec (PLQ) ou du Parti conservateur du Québec (PCQ). Dans la circonscription de Maurice-Richard, la simulation donne la victoire à une candidate de Québec solidaire (case orange) au détriment de la gagnante officielle du PLQ (case rouge).

100

Reposted by Fabien Niñoles

Ils étaient mi-cerf mi-escargot, et re mi-cerf derrière

Image de deux cerfs allongés devant des rouleaux de foin

62104173063

Fabien Niñoles @ninoles.bsky.social · 09/10/2026

Exemple très ironique de slop.

000

Reposted by Fabien Niñoles

bryan newbold @bnewbold.net · 07/10/2026

ah damn, RIP Margaret Hamilton. she took safety-critical software engineering seriously, and got people to the moon and back. news.mit.edu/2026/margare...

old photos of a woman (Margaret Hamilton) standing next to a tall stack of books (printouts of Apollo space mission software source code)

102143904733

Reposted by Fabien Niñoles

Daniel Burke - 🍺Innkeep @innkeep.bsky.social · 07/07/2026

Le Guin on the idea that there is a secret to getting good at writing.

I will dispose of the first myth as quickly as possible. The “secret” is skill. If you haven’t learned how to do something, the people who have may seem to be magicians, possessors of mysterious secrets. In a fairly simple art, such as making pie crust, there are certain teachable “secrets” of method that lead almost infallibly to good results; but in any complex art, such as housekeeping, piano-playing, clothes-making, or story-writing, there are so many techniques, skills, choices of method, so many variables, so many “secrets,” some teachable and some not, that you can learn them only by methodical, repeated, long-continued practice — in other words, by work.

8597123018

And through that, we have let big platforms decide which creators are worth our attention (and to live) versus those who are ignored and starved. 3/3

020

But what really concerned me is that we seem to have given up on taste and prefer to let a machine decide what's good and what's authentic. We have become passive consumers rather than appreciative spectators. 2/..

It makes new good authors even more difficult to find. You can see it in music where new "AI artists" are showing up all the time, feeding tens of new titles each year. 1/..

Un gouvernement minoritaire extrêmement régional, avec juste deux circonscriptions dans les régions fortement urbaines... J'imagine que les régions ont beaucoup de problèmes avec l'immigration et la langue française, et vont apprécier les coupures en services et infrastructures.

000

Remember someone saying that he received a heart bypass in the US in only 3 months for only $15K (probably only the surgery), pretending it would have taken years in Canada. I checked the numbers: - Urgent: 2w - Semi-Urgent: 6w - Elective: 6m All free.

000

I'm never sure if people who work in those believe in some kind of universalism or utilitarianism morale. The whole framework of this form of alignment sounds so much for me as an abdication of our own accountability to a machine.

0100

In the end, the best we can do for our future is to make today's better and preserve all our options. Nobody is better than anyone, all lives matter.

010

The second myth is the one that some people know is the best for the far future. That's like saying that we should base our current policies from some prophet of the 11th century. It is just the most stupid arrogance ever.

100

The main driver for it is the Meritocratic myth, which claims that an equation exists that can calculate the value of an individual. Michael Young invented the terms in a satire to warn about it, but people just adopt it as a synonym for freedom.

100

Fabien Niñoles @ninoles.bsky.social · 02/10/2026

The problem around most of the industrial era (including this day) is the strong belief that some humans (through race, biology, education, origin, past successes) are better than others. This leads to false questions like "Should we sacrifice X people with Y instead of W normal people?"

100

Oh, how much we can get if we can convince world powers that exploding old mines (terrestrial, marine) is a matter of national pride.

010

Reposted by Fabien Niñoles

Tim Carmody @tcarmody.bsky.social · 02/10/2026

This is the funniest thing I've seen today, possibly ever

A tweet by SarahStarberry that says "oh god read the vertical axis" on top of a chart measuring AI models' release dates and model numbers rather than any real capabilities

41120613121

Fabien Niñoles @ninoles.bsky.social · 02/10/2026

Pas plutôt à l'Ouest? L'Est, c'est le pont Champlain (vers les Cantons de l'Est). J'aime bien me rappeler que les plus beaux coucher de soleil sont à l'Oratoire, et les plus beaux levers au Belvédère.

000

I feel that just blaming AI for that is like a scapegoat and the problem will stay around longer, and more difficult to attack just like the Luddites ended up helping the industrial era (it's actually very similar!) by creating a system to protect the transformation.

000

I came across the concept of pharmakon and found it fits beautifully for AI as well as many technologies. But I don't even think this is the real problem. Tech is also an enabler, a multiplier for those who controls it, and it makes many of current techno-social problems way too big very fast.

Are you talking about the TESCREAL study? Note that those are content moderation of AI outputs, not just online. But I admit that they should have put some basic automated filtering for very bad content. I'm not sure how they were so irresponsible to let humans classify traumatic content without it.

At some point, someone would have to assign responsibility for all the problems this could create. It's just becoming so easy to say "it's the AI" but we can't continue without acknowledging the consequences on our social contract based on trust and accountability.

041

Something too many people who talk about austerity seems to not understand: you can't do more with less. Either you try to do more with what you have (aka efficiency) or you do less with less (aka cuts). But nobody has been able to do more or even the same after being cut; it's a recipe for failure.

0194