Stackless goes Limbo

[Donn Cave]

Quoth aahz at pythoncraft.com (Aahz):
| In article <mailman.1019911009.16159.python-list at python.org>,
| Laura Creighton  <lac at strakt.com> wrote:
|>
|> For those of you who can't live without a Limbo introduction written
|> by a real C inventor:
|>
|> http://www.vitanuova.com/inferno/papers/limbo.html
|
| Ugh.  I've gotten so corrupted by Python I can't even read well-written
| C code.

Tsk, that ain't C.  As I read on, I was surprised to find out that the
sys->print notation doesn't imply sys is a pointer, rather that it's
a module.  (Don't know why not sys.print, then.)  In some places it
could remind you of Python - uses "self" more or less the same way,
has garbage collection that apparently supports timely finalization,
tuple unpacking assignment like x, y, z = t.  Also vaguely reminiscent
of ML or something.

The important thing to me is that the system looks like it would work
well for modular function libraries, thanks mostly to the garbage
collection storage model, and who knows, maybe with less function call
overhead.  The thing I miss here is exceptions - nothing like that?
Maybe I skimmed it too fast.

I think I like the channels.  Not sure I like the Unicode strings.
I wouldn't mind trying it out, if it ever escapes from Inferno.

	Donn Cave, donn at drizzle.com