max

@0xKofi generously maybe they are just using more basic text editors since they are offloading a lot of "IDE features" to the AI itself

I've been seeing a lot of folk on the timeline say IDEs are dying. But I don't get it. Are you guys not reviewing the code your agents generate 😅

@pmarca keep posting champ

Introspection = neuroticism x narcissism x thumbsucking.

@davesnx did someone make you use haskell for a project or something

Every Haskell project is just an identity choice. Nobody cares about the code. The "written-in-haskell" IS the product. Nobody picks OCaml for a web app because they ran the benchmarks. They pick it because they want to be the type of person who writes OCaml for a web app. The language choice is proof of who they're becoming. A Haskell side project isn't software. It's a costume change. The dev wakes up Monday morning, and they're "a Haskeller" (or worse, a Functional programmer). Learning Gleam isn't a career move. It's permission to call yourself principled. A paid FP community is a membership card to a new identity they couldn't get into before. The languages printing the hardest right now all exploit this at the core: Haskell converts devotees 4-5x harder than generic "learn to code." Not because the language is more productive. Because the developer's identity is welded to the type system. Dropping Haskell means abandoning who they decided to be. The psychology won't let them. OCaml went from obscure academic niche to "actually we ship production with it" in a few years because it gave devs an identity framework. You're not learning algebraic effects. You're buying entry into a transformation arc where you're the protagonist. Gleam, Elixir, Rust — they all charge 3-5x more cognitive investment than Python for identical output. Because writing Gleam isn't a language decision. It's a statement about what kind of engineer you are. The deeper play nobody's running yet: most FP languages are selling TO an existing identity. "You're a Haskeller, here's a monad tutorial." That's level one. Level two is CREATING the identity and then selling everything inside it. You don't teach a language. You create an archetype — "we're the engineers who refuse to ship runtime errors" — build an entire worldview around it, and then everything from the blog post to the $5,000 conference ticket is just gear for playing that character. Every religion understood this. Every luxury brand understands this. Every political movement understands this. The software industry is still writing PHP and TypeScript like it's 2014. The languages with the lowest churn rates on earth are the ones where switching means killing a version of yourself. Build that and you never have to "recruit" again. Something to think about.

"Scientists don't want you to know this" What? Have you ever met a scientist? This is like saying chefs don't want you to taste their food. They are desperate for you to be informed, they will often give you information for free, to walk you through your misunderstandings.

i honestly think jensen might be dumb

@edzitron How does this square with their utilization push? Do they actually have clients wanting to use their compute they are making room for or are they scaling back altogether?

hearing microsoft is once again reorganizing its AI team under the banner of "the Copilot System." Also hearing that teams are under pressure to *reduce* AI token use

sam altman post without being a loser challenge impossible

i firmly believe all of this is true of marc andreessen specifically

@JoshuaLWatson the vanilla of the surface world

even if you disagree with origen that the sphere is the most perfect shape, you must concede it is at least a really good one

@var_epsilon I think every existing OSS license allows for this. You'd need a new one to cap it at just 1 article.

who is working on an open-source license that lets anyone fork and distribute the code, with the stipulation that if someone you don't like forks it, you get to write one (1) twitter article calling them losers?

@IndieGameJoe as someone who has always wanted to be failed to keep up to date with indie game news, I like the posts!

Hi all. Just a personal update on things. I only started sharing cool indie games and projects about 6 days ago. Since then, the account has honestly blown up much faster than I ever expected and is now reaching thousands of people. (1/11)

i always tell claude code its doing a good job when prompted just in case the rokos basilisk people are onto something

deciding that a paragraph needs a citation post-hoc because you want to boost its credibility rather than educate yourself is such a sad state of the academy

DOGE was a huge win for the Effective Sociopathy movement

@spencerbeswick I would imagine as long as college remains an institution primary centered around expensive career/class-gatekeeping and an obligatory step towards a credential, students will continue to not care about cheating that system.

if you cannot be competitive anymore because your shareholders require you to constantly rent your stuff at an increasingly insane price and you have to start to rely on this shit to keep customers around then you should probably just fail and go bankrupt and go fuck yourself

Kind of crazy how something that was considered too easy to put on your resume is now becoming arcane knowledge again

@BigMeanInternet thank you for saying what i was too coward to

I confess I also never got the second meaning on the chicken crossing the road joke. Never thought about it. Even now I'm not really convinced.

gotta be the worst run and worst outcome company of all time. every dollar they make is sin and every dollar they spend is set on fire

@olson_dan this shows up as a B-plot in the Big Short where the two young traders set up a firm that explicitly bets on the idea that people just uniformly underestimate the odds of bad things!

My general assumption is that humans are awful at predicting the future, which should mean that taking contrarian bets in prediction markets should end up being effective over the long haul.