[bmacho]

Yeah, sure. It is still a bummer when people that try actively to underpin the whole world tell you 'fuck you'.

Having 'we support proprietary systems until their makers support it' as a hard rule is unnecessary and harsh. It is a simple rule, but I don't think it is good. Why not choose the supported systems on individual merits, e.g. on usage statistics? I think for example the Linux kernel does that.

edit: I just continued the previous thought. If your reply means that you are open or likely to target deprecated systems in the future, that's much better!

[kubkon]

Hi, kubkon here, the author of Zig's MachO linker. I just wanted to explain our reasoning here a little bit. As you have hopefully noticed in the release notes, there are 5 core team members that are paid either full- or part-time for their work on Zig. That's not an awful lot, and as you can see in our latest roadmap for 0.12 release, we have quite a few challenging components to develop including functioning linkers for the most common platforms, and not only macOS/Apple. This makes it really tricky to justify spending a considerable amount of time trying to support an OS that was officially dropped by a vendor such as macOS/x86, or watchOS/arm32 (FYI, as a fun fact, arm32 is considered private as opposed to public by Apple according to comments and conditional includes in Apple's libc headers). That said, after 1.0/stable release of Zig, I would love to spend more time into adding backwards support for old Apple platforms in the MachO linker and Zig as a whole.

[kristoff_it]

> I just continued the previous thought. If your reply means that you are open or likely to target deprecated systems in the future, that's much better!

It would be nice to eventually provide support older systems but first we need to get to a point where doing so doesn't mean taking away resources from more worthy endeavors.