rirze

If your implication is that humans are deterministic, then that's laughable. We like to think highly of our mental processes, but humans work in different ways than AI and it shows. They are good at different things.

I am super interesting in `straylight-nix`. Even a blog post on what you did to improve it would be elucidating.

This mirrors my own experience, except I went with Lean. It was more relevant to the features I wanted to build.

Exciting times ahead.

Try it out! It has its own learning curve, but it's convenient to use in quick and dirty situations.

Was it still in good condition? That price makes me wonder if it was used for crypto mining, which can wear down the hardware.

Interesting, I'll be testing your tool on my repos. You should publish to crates.io!

How did you even afford to use Fable + Ultracode ? I feel like the subscription (even the $200 one) is not enough for this workflow. Are you using API or a company plan?

Seems like white hat work is pretty fruitless nowadays. Disappointing.

If a messaging app ever gets the attention of government regulators, it must succumb to this verification.

I don't know any way to avoid this.

Fundamentally un-American.

That being said, many countries across the world already do this to eliminate burner phones. And many messaging apps require a phone number anyways so this basically locks down anonymous messaging through a phone.

This is amazing.

I love it https://github.com/vorpus/performativeUI/issues/2#issuecomme...

I agree, when agents realize they're using old Zig semantics, they will adapt.

Doesn't help the training sets that Zig changes frequently though.

> the llms write pretty good zig

I doubt this from my personal experience. Every week after a release, I see tweets complaining how AI wrote some depreciated code because Zig is making breaking changes every release. (They are valid in doing so, it's just not AI friendly yet)

There is, it’s being able to tell bullshit from actual cosplayers.

Ladybird’s blog post is arguing that it used to be tolerable to spend the amount of time evaluating this for every PR, now it’s just unmaintainable for their effort-time.

I hope the author is ok with this kind of publicity. He might not have wanted attention if the language is not ready for public opinion.

Forget vacation and overtime pay, I'd expect equity at that point.

Julia only cares about numerical performance, and in that regime, it’s pretty fast.

So not generally fast, no.

It's sad that you voiced an actual question and you got downvoted.

To answer your question, remember that people will only approve a LLM's output if it matches with their perspective and priors. So if you see a slop issue, it reflects on the human user who didn't see an issue in it (thus their prompt framing or refining is wrong).

I was excited for this years ago, but when I tried it about 2 years ago, it felt awkward. Is it different now?

This is not true at all, a year ago it was difficult to force models to use jj consistently, but nowadays, (codex+chatgpt, at the very least) models very nicely obey jj workflows when guided in AGENTS.md . They have been training models to use it.

Edit: my original comment was related to how git worktrees are by default used in the implementation of these agent orchestration tools. I would rather prefer jj workspaces.

Surprise: another day of mechanical engineers not really understanding what open source really means.

I mean this as general advice; never trust software released/written/shared by mechanical/robotics engineers, especially ones from China. There's something deep-seated in their psychology where they ignore coding standards and can't conceptualize and obey software licenses.

I'm confused, I downloaded their repo and ran it and I had the option to continue locally without signing up/in.

I see a bunch of these orchestrators, but is anything out there jj (jujitsu) native? I would really like to avoid hacking onto an existing product.

Jeez, the number of replies taking this as a knock on current political divide is saddening.

If anyone remembers middle school history, the American civil war was ignited by the North refusing to accept Southern secession. Should've made it more obvious, I guess.

That's a super generous plan. Indeed, it's hard to normally saturate the limits of the OpenAI's $200 plan.

Incoming Canadian Civil War?

> high integrity systems with memory allocation constrains

I agree. I also find that a lot of software should have this mindset anyways so Rust is a great fit for that domain. Anything that others depend on should be high integrity and memory efficient.

Not to mention the ecosystem is very wide in Rust. Other ML languages may not have what one needs.

> I'm convinced that most of those folks and their elaborate workflows aren't really for productivity but for bragging rights about how much they use AI.

This is quite the reductive, charged statement. Can I ask what subscription plan you're using?

My personal experience is unlike this at all-- I work on ever-expanding codebases so I can easily burn tokens. Not to mention, structured agentic coding with adverserial reviews & task organization is not token-efficient. Additionally, for the problems I'm working on, only xhigh or high reasoning gives me worthwhile results while saving time. There are definitely configurations where default consumption doesn't work.

For reference, I used 15 billion tokens (most of it cached) last month on my day job's enterprise plan. That doesn't include my personal plans' usage.

You could also do it the menial way; create multiple QR codes that can connect together. Not very practical for everyone except the very-motivated.