HN user

yard2010

1,887 karma
Posts0
Comments1,240
View on HN
No posts found.

Steve Wozniak is a present that helps me keep this equation on the positive side every time I stumble upon something he said or done. This included. In an alternate universe the society values these kinds of people and not the bullies.

This sends a clear message and it echoes the "you can't solve a societal problem with tech" comment from the other thread - there is a right way and a wrong way of breaking the law. It's not that you have to keep the law, you just need to break it in the way that the consequences can be contained.

I think it's just a matter of time until everyone learns this. And then it will be the end of the slowly dying liberal democracies.

Why not both? In Prepbook[0] (shameless plug) we created a web app, then used it to make a native app. You can use whatever you want. I like both. The native app though gives an objectively better experience - you can set timers on the OS level, open recipes in the app using the OS native share menu, etc.

We worked hard so you don't have to vibe code your way to get the experience you prefer.

[0] https://prepbook.app

I have a coin that can tell you whether a program would halt or not. But it's not always right. I think it's a coin like you have? Still trying to figure out how to use it to prove that p=np.

Honest question - does using claude code and steering it to type the code I want counts as hand-built? Like doing actual PRs, nit picking, basically treating it as a jr/product/typist while I'm the one making ALL the decisions? Reading every line and approving manually if it's a code that I would write?

Ask random questions until you get enough sense of things to ask "the right" questions. Never stop asking questions. Use everything you can to get the answers and never settle for any answer no matter how good it sounds, doubt it and go back to step 1.

I agree with this sentiment so much but before I could figure I turned into it. I'm feeling torn - it's helping me write and ship good code as I couldn't before, but it feels like I don't understand the real price of using it non-stop.

How about leveraging llms to produce deterministic codemods? You can then iterate on this by running the codemods and using other deterministic guardrails, feeding the results back into the llm to improve the codemods?

Superpowers 6 20 days ago

Same here but yesterday I decided to give this a goal. It started a workflow that spawned like a 100 sub agents to research the best o11y product I can use. It burnt through all my max plan (before it could start making any changes), something I could never do simply by using claude code yet. It's not for me.

"You work hard, screw over everybody that you love, hurt, rob, kill indiscriminately and maybe... just maybe, if you're lucky, you become a three bit gangster. It's bullshit. Go to college. Then you can rip people off and get paid for it. It's called capitalism"

Recall that the marginal utility of money diminishes when you have more of it - when you have a lot of money it's easier to turn it into even more money, and vice-verca. It's not linear. So 20$ difference has exponential not linear influence on "being rich".

Not gp, I want to pay the creators. I also want to own what I pay them for. When I have this option (bandcamp, physical media etc.) I'm going for it without even thinking. It's not about the money. It was about the money when I was 13. Now it's simply about owning my stuff. Renting is not for me. If I don't have any other alternative than pirate, what should I do?

Sharing is caring. If I can share my money with creators I go for it every single time, especially when these creators are simply making my life better with their art.

It's more complex than "want stuff for free". Usually I hate free stuff as they cost in time or worse.

It might be the other way around. There are powerful people with money that simply behave. A few assholes turn things into shit for everyone, when they also have money it just becomes worse.

OpenRA 25 days ago

You are awesome, thank you for being such a magical part of my childhood

OpenRA 25 days ago

Instead of driving the agent with an llm, it might work to use the agent to hard code heuristics, and use some kind of a simulation to benchmark its skills? Then feeding the results back to the agent so it can improve the heuristics?