[MLton] upcoming release

[Matthew Fluet]

> Hopefully this mail will get us started down the path toward a
> release, and we can do it in sometime this month.  Please send mail to
> the list about any issues (bugs, features, documentation, ...) that
> affect the release.  Thanks.

One other unresolved issue is whether or not to include an _address
primitive, and what its annotation should be.  See
http://mlton.org/pipermail/mlton/2005-July/027519.html

Wesley has rightly pointed out that the current implementation prohibits 
taking the address of a function from the ML side, except by lying about 
the type of the object.

Also, I still wonder if "define" is the right name for the annotation that 
indicates that the symbol should be (statically) allocated by the ML 
object code.  How about "storage"?

Because MLNLFFI makes extensive use of the MLton FFI primitives, I'll need 
to stabilize the FFI side of things before bringin NLFFI up to date.