[imp-hackers] cl-ext:exit

[Nikodemus Siivola]

It's been very quiet here! I'm not sure if that's because we have so
little so say to each other, or because no-one wanted to go first...

Since we are all very out of practice in cross-implementation chatter,
I was thinking to break the ice with CL-EXT:EXIT.

It is downright embarrassing that there is no shared way to do
something as basic as exit a CL program. Every implemention that I
know of provides an operation to do it: some call it QUIT, some call
it EXIT, and the arguments vary wildly -- but it's always there. To my
mind it would make an excellent point to look for a consensus, and
maybe even agree on a package (CL-EXT?) where it could be found.

In practical terms the implementation cost is minimal -- as is the
practical advantage since this is the most trivial implementation
specific feature ever to #+ into submission, but I think the PR value
should not be underestimated: it's been about 15 years since
implementations have last publicly agreed on something by actually
talking to each other...

*If* we could agree on that, I'm sure there are more interesting
things we could look for common ground on -- to everyone's advantage.
I may be an idealist, but I very firmly believe that it is easier to
get a bigger pie than fight over slices.

How does this sound? If the popular opinion is that it is not worth
it, I'm more than happy to talk about potentially more interesting
things like restricting incompatible FTYPE proclamations or the real
meaning "otherwise inaccessible" as it pertains to DX declarations...

Cheers,

 -- Nikodemus Siivola (an SBCL hacker)