davidkunz

I'm not an expert. Can't we abuse that LLMs don't need to receive audio as a continuous stream without interruptions? Couldn't we just send data and pipe it into the LLM with deduplication (if resending happens)?

  x...y...y[dedup]...z

I would like if some of that functionality is extracted in CLI tools. Then every coding agent can use it.

I haven't tried Löve, but I somehow enjoyed reading through the README.md, no AI slop, just a natural writing style with tiny indictors showing the authors' enthusiasm in creating software.

We need Net Neutrality for LLMs.

Useful for tests with LLM interactions.

Oh, you should have a look at Pi then.

https://github.com/badlogic/pi-mono/tree/main/packages/codin...

It's still possible to let users already type from the beginning, just delay sending the characters until checks are complete. Hold them in memory until then.

Give it "a week or two"

Just use skills, which allow progressive disclosure of information.

Falling sand games always remind me of the game Clonk. As a kid, I enjoyed digging tunnels, flooding them with water, all physics based. Great times.

On the website[1] it says:

  .opencode/skills

The root cause should be fixed.

Please standardize the folder.

  .claude/skills
  .codex/skills
  .opencode/skills
  .github/skills

It's in Java, but the lessons can be applied in every language.

I can only discourage anyone from applying Java patterns all over the place. One example in JavaScript: There was a functionality that required some parameters with default values. The plain solution would have been:

    function doStuff({ x = 9, y = 10 } = {}) {  ... }

In JavaScript, I love the `async` keyword as it's a good indicator that something goes over the wire.

One step closer to Skynet

What I would love:

- Everything locally stored in the repo: PRs, comments, issues, discussions, boards, ... - CLI first - Offline first (+ syncing) - A website for hosting/presentation

jj describe -m "Good luck, Steve!"

I have huge respect for Mitchell, it's impressive what he achieved.

I agree with all the points of this article and would like to add one: Have a quick feedback loop. For me, it's really motivating to be able to make a change and quickly see the results. Many problems just vanish or become tangible to solve when you playfully modify your source code and observe the effect.

It's all good, man!

there is very little point to any of this to anybody else. Don't expect some great useful guitar pedal experience.

Yeah... He said similar things about Linux.

I tried to run my project with bun - it didn't work so I gave up. Also, there needs to be a compelling reason to switch to a different ecosystem.

Thank you all for these great and detailed explanations, I've learned a lot! I like the approach with an optional pointer, it fits to zig's philosophy quite well. Although there's a bit of a disconnect between the unadorned error and the corresponding data struct. I could imagine it requires care when the data struct is a union, as one needs to know which error corresponds to which variant.

A little bit unrelated, but how do people deal with the abstinence of payloads in zig errors? For example, when parsing a JSON string, the error `UnexpectedToken` is not very helpful. Are libraries typically designed to accept an optional input to store potential errors?

I'm a bit concerned when library authors only test it with blocking Io and the consuming app with a different kind. Wouldn't this potentially lead to bugs?

Thank you! I found ast-grep to be really useful, I hope more people will discover it!

I wish companies would adhere to a consistent naming scheme, like <name>-<params>-<cut-off-month>.

I wouldn't exclude Lua, it's a simple language whose basics you can learn pretty quickly if you know JavaScript or comparable languages. I therefore would recommend Neovim.

Maybe if you can't precisely model your structure with (OpenAI's subset of) JSON schema.

Nothing new, we've been doing it like that for ages here in Germany. But it's a cool Hamburger phone.