[simonw]

In this sentence, sure:

> Anthropic is dogfooding this, the machine is still ticking along, and Anthropic employees are directly involved.

The problem is that your article's central claim is that there hasn't been a Bun release since the Rust rewrite - which can be read as implying that the rewrite hasn't been used in production.

But it's been used in production on millions of machines running Claude Code for over a month!

I think failing to acknowledge that hurts the credibility of the article. It's been a heated discussion point in this thread already.

[root_axis]

> which can be read as implying that the rewrite hasn't been used in production.

Well it hasn't. "Production" for a language runtime means being generally available for arbitrary use. The engineers that built the runtime using it to release one closed source binary to the public is, at best, an extremely narrow beta test.

[simonw]

It's available for arbitrary use if you run the canary release.

[philipwhiuk]

In no world does "New iOS released" mean "there's a developer preview" which is what you're proposing.

[tomlockwood]

I link to Jarred's writeup on bun.com, where you'll note he mentions that its being used on Prisma and Claude Code.

I don't agree that using something in a very specific environment is the same as a wide release.

I trust readers will either know the backstory or read the articles I link to. Such is life if they don't.

[simonw]

Why are you resistant to adding a sentence to the article that notes that Claude Code uses the rewrite?

Do you think it would weaken the article?

When I said it would improve the credibility I did mean it. My instinct on reading the article this morning was "this person doesn't know that Claude Code runs on Bun, which weakens their credibility in presenting the argument they are making here."

[aeturnum]

To me this seems like an odd preoccupation for an article that is very little about the Bun rewrite as a running program and very much about the process of the re-write and how it came to be in commits to the repository.

You could argue that the release tag isn't important - but at the top of this comment thread a member of the team confirms they delayed the release to finish some extra testing. So the ambiguity around if the project is complete and pointing to the lack of a release mirrors internal sentiment as well.

[tomlockwood]

Author here: Exactly.

I had a funny thought that Bun might add a release tag while I was still writing the article. In that case I'd still publish it, because the point is we still don't know the cost it took to get to a viable, wide, release. The $165k isn't the full story.

[solid_fuel]

> Why are you resistant to adding a sentence to the article that notes that Claude Code uses the rewrite?

It's not your article, bud. They have already addressed your concerns, and don't need to make changes just because you feel the article isn't "fair".

> My instinct on reading the article this morning was "this person doesn't know that Claude Code runs on Bun, which weakens their credibility in presenting the argument they are making here."

And your instinct was clearly mistaken. You should read the article more carefully before offering criticism.

[simonw]

I stand by what I said. I was offering honest advice: when I read the article I thought exactly that, and so did other people based on the comments in this thread - which I think shows my instinct has value here.

There's an easy fix. If they don't want to take my advice that's completely fine.

[tomlockwood]

Please feel free to write an article that corrects my mistakes.

[simonw]

I try not to publish articles that call out mistakes by other people just for the sake of it. I've written a fair amount about the Bun rewrite already - https://simonwillison.net/search/?q=bun%20rust

[tomlockwood]

You've got my blessing. I think given how mistaken I am, it should be quite an engaging article.