Hacker Newsnew | past | comments | ask | show | jobs | submit | VMG's commentslogin

... unless you actually want to edit a change!

well, you can do jj new <revision>, make your edit, and then do jj squash which will add the changes to the prev revision

i do this for example when i want to see a specific edit highlighted in my editor, it's a nice workflow i think


This is exactly how someone explained Git to me 12 years ago or so, and I’ve finally wrapped my head around it. Not changing now.

If I'm understanding the thread correctly, I have a git alias to `git commit --amend --no-edit`, for exactly this workflow. When I'm hacking on something locally and want to just keep amending a commit. I only ever do this if it's HEAD though.

Yes, one way to think about jj in a sort of low-level way is that every jj command does the equivalent of that, every time.

(You can also set up watchman and have that happen on every file change...)


I go back and forth between the two approaches, but because of the whole "accidentally made some temporary changes and now it's a pain to separate/undo them because not all changes were temporary", I also usually do a jj new and then jj squash.

still use new, and then squash your changes in. that way you can actually see what changes you made

then you `new` & `squash` :)

I've had mixed results.

Most models don't have a 100% correct CLI usage and either hallucinate or use some deprecated patterns.

However `jj undo` and the jj architecture generally make it difficult for agents to screw something up in a way that cannot be recovered.


Try using https://github.com/danverbraganza/jujutsu-skill

This is enough of a command reference that with it, agents are able to work with jj pretty well.


I've gone all in on jj with a OSS framework I'm building. With just a little extra context, the agents have been amazingly adapt at slicing and dicing with jj. Gives them a place to play without stomping on normal git processes.


503



Skeptics Guide to the Universe


developers with good taste like Andreas Kling will be able to design entire OSes with coding agents


This comment raises an interesting question: Would Serenity OS have brought Andreas the same kind of serenity had it been developed with AI? Open candid question.


I don't think so because if I remember it correctly, Andreas suffered from alcoholism and serenity prayer helped him to go on the right path and iirc he honored that and created an os named serenityos.

God grant me the serenity

to accept the things I cannot change;

courage to change the things I can;

and wisdom to know the difference.

(courage to change the things I can;):- I think that this line must've given Andreas the strength, the passion to make the project reality.

but if AI made the change. Would the line be changed to courage to prompt an all powerful entity to change the things I asked it to.

Would that give courage? Would that inspire confidence in oneself?

I have personally made many projects with LLM's (honestly I must admit that I am a teenager and so I have been sort of using it from the start)

and personally, I feel like there are some points of curiosity that I can be prideful of in my projects but there is still a sense of emptiness and I think I am not the only one who observes it as such.

I think in the world of AI hype, it takes true courage & passion to write by hand.

Obviously one tries to argue that AI is the next bytecode but that is false because of the non deterministic nature of AI but even that being said, I think I personally feel as if the people who write assembly are definitely likely to be more passionate of their craft than Nodejs (and I would consider myself a nodejs guy and there's still passion but still)

Coding was definitely a form of art/expression/sense-of-meaning for Mr Andreas during a time of struggle. To automate that might strip him of the joy derived from stroking brush on an empty canvas.

Honestly, I really don't know about AI the more I think about it so I will not pretend that I know a thing/two about AI. This message is just my opinion in the moment. Opinions change with time but my opinion right now is that coding by hand definitely is more meaningful than not if the purpose of the project is to derive meaning.


I like the idea that people are either coders or builders. So AI can help fulfill your desire to build, create, bring things into reality. But it can't satisfy you if you like programming for its own sake. SerenityOS was not a practical project, it was clearly done for the enjoyment of programming itself.

The project's use of AI now echoes that - it's not being used to create new features, it's used for practical, boring drudge work of translating between two languages. So still very much on brand.


> design entire OSes with coding agents

They ported an existing project from CPP to Rust using AI because the porting would've been too tedious. I don't think they're planning on vibe coding PRs the way you're imagining.


He already did


Yeah, some weekends ago I tried writing a cross-platform browser without any Rust crates, this weekend I made my own self-hosted compile to Rust Clojure-like lisp, maybe next weekend attempting to create a OS that uses my language to run on bare-metal would actually be a challenge. Thanks for the inspiration :)


Not at all if you consider the internet pre-LLM. That is the standard expectation when you load a website.

The slow word-by-word typing was what we started to get used to with LLMs.

If these techniques get widespread, we may grow accustomed to the "old" speed again where content loads ~instantly.

Imagine a content forest like Wikipedia instantly generated like a Minecraft word...


> Base64-encoded secret in URL Prevented Detected (entropy scan) Logged

Ok so how does this "Entropy scan" work?

Apparently by defining "bits per character"

https://github.com/luckyPipewrench/pipelock/blob/3021f023b0e...

So I guess converting the secret to pure binary will evade the "entropy scanner"?


Or many of the other base encodings?


> Because if it’s worth your time to lie, it’s worth my time to correct it.

https://www.astralcodexten.com/p/if-its-worth-your-time-to-l...


> and I expect within the next ~2 years AI tools will produce a better compiler than gcc

and the "anti" crowd will point to some exotic architecture where it is worse


No, they will point out that the way to make GCC better is not really in the code itself. It's in scientific paper writing and new approaches. Implementation is really not the most work.


Consider applying for YC's Summer 2026 batch! Applications are open till May 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: