Re: [Thought experiment] Bringing swift into an experimental branch?

On Mon, 16 Jun 2025 16:27:04 -0600
Alan Somers <asomers@freebsd.org> wrote:


> Note that said branch does not import the Rust toolchain itself.  It
> relies on an external toolchain.  That made it quite easy to add all
> of the Rust support to the Makefiles.  And there's no cost at all, in
> either runtime, buildtime, or maintenance burden, to people who
> aren't using the Rust stuff.  The same would be true of Swift, I
> assume.  But I think that it's highly premature to consider such a
> thing before Swift even lands in ports.
> 
> As for whether Swift is appropriate for writing base system
> utilities, I don't know enough about it to confidently say.  After
> Swift stabilizes in ports, and after somebody creates a swift-in-base
> branch, then we can have a real holy war ;)

Add...add...add...add... who will support all that crap!?
Every one likes make childrens, but after they born - they requre a lot attention and resources.

Before add something try to see all life cycle of code/language.


PS: I will never belive in "no cost at all" since even C compilator takes time to bootstrap build.

"to people who aren't using the Rust stuff" - and after few years some on will say:
- hey!, lets remove legacy C code and switch to modern crappy rust/swift staff.