1) Proper constructors, destructors, fallback method searching, and
overloaded operator and assignment for objects
2) Namespaces all finalized
3) Runtime bytecode loading nailed down and documented
4) Library path handling
We're pretty close on #3, so that ought to be an easy one. #2 is
documented, we just need to make it work, and #1, well... I know what
to do, there just wasn't time to do it for the 0.1.0 release. #4 is
just a matter of getting the library routines written so we can wedge
in path searching properly once we've figured out how exactly we want
to do it.
Also, I really, *really* want to get more machines for more different
architectures into the tinderbox. I've got rack & shelf space for the
things even, so if you've got a SPARC, POWER, or Alpha box kicking
around without a spot to plug it in and leave it running full-time,
that's manageable--I have space, power, and connectivity at hand.
--
Dan
--------------------------------------"it's like this"-------------------
Dan Sugalski even samurai
d...@sidhe.org have teddy bears and even
teddy bears get drunk