logoalt Hacker News

NewsaHackOtoday at 12:37 AM1 replyview on HN

>Having a project consider a rewrite isn't so big a deal.

I don't agree that them actually doing an entire draft rewrite can just be characterized as them considering a rewrite.

>I don't think the Zig team is concerned at all.

I wonder if that's the mentality that got them in this situation in the first place.


Replies

toshinoriyagitoday at 3:54 AM

>I don't agree that them actually doing an entire draft rewrite can just be characterized as them considering a rewrite.

You're right, a rewrite is in existence, and whether it is good enough to be used or expand upon is what is being considered. I don't think that changes the fact that languages don't live or die by whether or not 1 large project using them continues using them. Especially a language like Zig which has taken plenty time making breaking changes. They know this is par for the course.

>I wonder if that's the mentality that got them in this situation in the first place.

I highly doubt it. To my knowledge, the only "why" Jarred has given is frustration with memory issues. Speculated reasons I see are: 1. Anthropic wants a rewrite to a language with a more favorable AI contribution policy, to avoid bad press by acquiring a framework written in a language that is skeptical of AI code quality. 2. Rust is more stable and a better target for AI-assisted programming or entire vibe coding. 3. Bun is upset Zig does not want to merge their fork into main.

Focusing on the issue Jarred gave as why he started the rewrite, I don't see how Zig got themselves into the situation at all. Zig was always upfront that it aimed to be a modern C: simple language, powerful modern features, and excellent compatibility with all things C. While it certainly has much better behavior concerning memory safety and undefined behavior, it has never aimed for Rust or GC level memory safety.

It's not like Jarred has been begging the Zig devs to implement language changes to make Bun development easier. Zig was always upfront that you will have to manage memory manually, and that allows for operator error. I think Jarred is in this situation because he wants to be, simply. He works for Anthropic, probably has no limit on how many tokens he spends, and may have access to their most powerful internal models like Mythos. I would guess he pointed agents at this problem and let them go, because why not? He has likely has no opportunity cost.