In this Twitter account you will find new open source tools, software development jokes, and stories around developing @Autoprefixer, @PostCSS, and @logux_io. Subscribe to my other accounts about fun STEM facts from Wikipedia @lostinwiki and popular linguistics @linguopunk.

You didn't know but old.reddit.com is alive

The craziest thing about Elon Musk becoming a trillionaire is that he begged Jeffrey Epstein to let him come to Little Saint James on Christmas Day

Oh no! Fable is a game changer and the US government seems to agree. The main issue isn’t even that most of us now lose access to Fable, but that they might block access to all future models in that weight class as well and not just for Anthropic.

This is pretty insane. It's the kind of thing that puts the lie to "Trump is going to let us do whatever we want" from the Andreessens of the world. They might do that, or they might randomly issue orders purporting to shut down an AI model on the basis of nothing.

this is like 9/11 for techbros

Hard to believe we went from the optimism of free the nipple to the acceptance that all of our instagrams will be deleted over the course of the rest of the year

What Europe should do right now: 1. Call all the European researchers working on AI and return them back with same salary (or they can stay but switch career). 2. Fill EU places having GPUs with money, and put those people there. 3. AI partnerships with China India.

i’m still cooking on plenty of stuff that stays true to the root of okuri traditional overengineered #designtwitter

A “scientific” journal called Rubbish has launched in China, allowing stressed-out young scientists to submit humorous papers, such as "How should Optimus Prime be insured—as a car or as a person?" webofnothing.org

Took a good bit of time to get right, but really happy with the warped edges on the new system.studio design.

in my data, people aged 50 in new relationships (under 6months) report having more sex than 25-year-olds who've been together five years

NEW: malware developers added nuclear & biological weapons text to to their spyware. Goal? To trigger LLM safety refusals... so that their spyware wouldn't be analyzed by an AI security scanner. Cleanest practical example I can think of for why over-indexing on first order safety alignment is risky. When closed (and open) models ship with aggressive refusals, they will be sprinkled with second-order blindspots that attackers will discover...and exploit. We are only in the earliest days of attackers leveraging these features, and it wouldn't surprise me if users systems that need to handle complex cybersecurity issues demand that models be less safety-blunted. In the weeds: @SocketSecurity's post also shows why intention matters in how you design a malware analysis pipeline to avoid prompt manipulation. H/T to colleagues that shared this with me socket.dev/blog/mini-shai-hu…

We also have another incredible news to share. The Zen Community has grown to an estimate of over 500.000 users this week! If you want to support Zen's development please consider donating here: — zen-browser.app/donate/

For almost the entire reign of Alexander I, the Russian Orthodox Church and public education were headed by a homosexual man Meet Prince Alexander Golitsyn: one of the most powerful and contradictory figures of the Russian Empire. He was a former freethinker who became a mystical religious minister, sponsored the first mass translation of the Bible into modern Russian, and navigated a court where his private life was an open secret. Read the full story of his incredible rise and the homophobic intrigues that brought him down in our new article: urania.institute/en/posts/ru…

Have been happily using mine for over a year and now that the hype is dead, I'm open sourcing it on my GitHub. Made a few adjustments on the model, edited my codes to more human legible form.

💻 Un développeur a créé un trou noir qui grossit sur votre écran à mesure que vous travaillez sans faire de pause 🤯 Un développeur a conçu un shader pour Ghostty qui ajoute un petit trou noir animé dans le terminal. Au début discret, il dérive lentement à l'écran et déforme le texte grâce à un effet inspiré de la lentille gravitationnelle. Plus l'utilisateur travaille longtemps sans interruption, plus le trou noir grandit jusqu'à devenir difficile à ignorer, transformant la fatigue numérique en élément visuel. Dès qu'une pause est prise, il rétrécit puis disparaît.

Lleó XIV afirma que resarà perquè millori Rodalies, però "no pot fer miracles"

Our statement on the UK government’s demand that all content on all devices sold or used in the country be scanned, on the presumption of nudity, using a dystopian combination of age verification and content scanning. This proposal will not safeguard children. It endangers us all. signal.org/blog/pdfs/2026-06…