ngruhn

Won't there also be a point were you're shipping so fast that that in itself deteriorates UX? What if a page literally changes every time I reload? What if it changes mid-checkout process and the deal I agreed to 5 seconds ago does not exist anymore?!

So as a user I can basically deploy code through bug reports? Sounds like a prompt injection nightmare.

You should call it "pond".

For Dudley Dursley it's the worst getting only 36 presents for his birthday. It's good to maintain a sense of proportionality. If you call everything hell, the word hell looses meaning and it's also an insult to people actually living in hell.

If you live in North Korea or pre-unified East Germany, you know the difference. You don't protest, you don't express your opinion, even in private because who knows who the government spies are or who will snitch on you for slightly better living conditions in the system.

ICE has murdered protestors but each case is a headline. Start worrying when it's not a headline anymore. Either because the media are suppressed or because it becomes so common that it's not newsworthy anymore. Wait until the army indiscriminately opens fire on crowds of people and the government isn't even trying to justify it. And when you hear stories of people just disappearing. Then you will be scared to post comments like this.

ICE arguably has a mandate for aggressive deportation. The Trump campaign did not sugarcoat what their goal was. And people still voted for that. Still, the lack of due process, the excessive use of force, deporting people to random countries, ... that's not normal.

Well it can definitely get worse. You're not yet scared of being picked up tonight for posting this comment.

can't you just track solar panel sales?

This is within hours of the original result. No way this can be reliable, can it?

It's running on 20 watts though. There must be some tricks in there.

I guess I don't understand "enterprise psychology" but why are auditing features worth a 50x premium? I think we're currently blasting through $1000 worth of tokens a month per dev.

Is there a word that combines wholesome and insane?

This is great. Many people online who "verify their software" seem to just tell an LLM "verify dis" and have no idea what comes out. The model could be proving 1+1=2 for all they know.

These guys actually show some Lean theorems they proved. Also interesting approach to compile the existing interpreter into the de-facto spec.

whoops :P

A good outcome of this crisis. These god damn fossil fuels. Never even mind climate change. Being depend on them has turned out to be a huge national security risk. Picking a seller is picking your poison: Russia, the Golf states, and now the US.

First I wasn't sure if technical debt still matters but I think it absolutely does. As long as I'm involved in the planning phase at all, the complexity of the code base still bubbles up to me. Claude will say: ok we can build that but how should we handle legacy code paths A,B,C,D that no one needs and I wasn't even aware existed.

Isn't that overkill? You can check exhaustiveness with a trivial little helper:

    function exhaustive(witness: never): never {
      return witness
    }

https://www.typescriptlang.org/play/?#code/FAFwngDgpgBAKgZQB...

Hell no. Writing formal proofs has huge overhead. Lean is not a productivity tool; it's a reliability tool. It's like asking if airport security made boarding faster.

Mathematicians are also not upset about Lean. They're upset because they feel like they're becoming redundant. And that has everything to do with LLMs and not Lean.

IIRC humans have a threshold at the same point. For up to 4 objects we can determine the exact count at a glance. For more we estimate or have to do deliberate counting. Maybe the crow just can't do the deliberate counting.

haha true

haha :D but consider they were not just on some rooftop nearby. They were quite close. I wouldn't mind per se but the social cost... I think it was obvious to everyone in the streets that these birds were following me. I didn't want to be the bird guy.

I saw some video once about intelligence tests on crows. In one experiment they had a tall but thin glass of water with some treat floating on the surface but it was too low for the crow to reach. So the crow started dropping rocks into the glass, causing the water level to rise, until it could reach the treat.

I was so impressed, I wanted to replicate that. I went to a park around the corner with some nuts to befriend some crows. My plan was to just feed them for a while so they would trust me. But after a few days I had flock of crows following me everywhere through the city. That was a bit too much.

Rationalisation is the default mode of operation for humans. Believe first; find evidence to justify it. That's what everyone is doing. Not that self proclaimed rationalists would be free of that. But pretty unfair characterization for a movement whose key goal is to that as little as possible.

Making rational decisions means picking the most responsible option given imperfect information. It doesn't mean delaying decisions until you have perfect information because that point never comes.

not a red flag at all

nice, welcome to the dark side brother

Isn't that what proof of work does? No idea why this is not common place.

> the more evil stuff these agents do the more capable and impressive it makes them look

This is a silly meme. Who wants to pay for evil models that go rogue in your company infra?

Yes they're skipping safety measures. They sacrifice it for speed in an arms race. That's part of the call for industry-wide slow down.

But also, frontier models are digging up tons of zero days even in "serious software". I'm sure any sandbox you and I can come up with looks like Swiss cheese to Astra 7.

I feel like my role is rather to say "yes". Claude wants to keep all that legacy crap around and I keep saying shut up and purge.