Getting to an LTS release train of bun is probably a good idea, assuming they are still interested in external adoption rates. They shipped 1.0 a little under 3 years ago and the acquisition has had ~6 months to get settled. Looking at an LTS release would enable a lot of the more slower moving places to look at bun within being worried about getting caught up in the velocity.

I don't know it makes sense to try to make the an LTS version of where the Zig version left off, particularly if they know they are shipping a different solution & codebase to tackle the bulk of their security and memory bugs. Let this settle a few more months with the intent of releasing an LTS of it by the end of the year, if there really is demand for it. All of the benefits of having an LTS version like Java, Node, and .NET without needing to jump to a pre-port version of the codebase which was never targeted to be a good version to LTS in the first place.

A final Zig-based LTS would have made great sense if LTS releases had already been in the picture though.

So not just free work, but you expect Anthropic to pay to maintain a Zig version...?

I don’t follow how it would be in Anthropic’s interests to maintain an LTS version of Bun. It seems unlikely they’d use it, and it would cost them money (in the form of labor).

To my understanding, Java and .NET offer public LTSes because they’re financially buoyed by huge companies using those LTS versions, and Oracle and Microsoft are fundamentally selling that LTS support as a product. I have no idea why Node does, maybe it’s similar. But Bun’s position appears to be different than all three.

Are you asking for an LTS Zig version as a member of Anthropic?