On Tue, Sep 1, 2026 at 4:37 PM <[email protected]> wrote:
> Greetings.
>
> Zig is, apparently, a fresh new better version of c. For people not using
> plan9, there will be obvious interest in it. Plan9 already has a better
> version of c. Assuming that zig eventually stabilizes, does it offer any
> innovations that, if incorporated into plan9 c would make it better? would
> there be any point in re-writing plan9 in zig, the way others have re-written
> it in rust?
I would argue that Plan 9 _had_ a better C, but that that is no longer
the case. Some things are verifiably misfeatures (anonymous struct
inclusion, for example), and others are gratuitously different from
ISO C with little real-world benefit (designated initializer syntax,
for example, or the integer promotion rules; on balance, I think the
ANSI C folks came down on the wrong side of history here, particularly
given that signed integer overflow is "undefined behavior", but at
this point being different is just a headache).
As pleasant as programming in Plan 9 C because the compilers don't
aren't antagonistic is, porting to other systems occasionally exposes
UB footguns. :-/
- Dan C.
------------------------------------------
9fans: 9fans
Permalink:
https://9fans.topicbox.com/groups/9fans/Td859d3c6cefc922f-Mebb199b52525679bfc5412e8
Delivery options: https://9fans.topicbox.com/groups/9fans/subscription
