Hi,
Perhaps it'd be nice to tie up all these memory issues with some kind of kernel patch... a way to read/write and generally mess with another process's memory that was broader and more efficient that p-trace.
Something like that would be great for Linux, not sure how it could work with closed unices though. I've been trying to figure out the server and it does't look like anything currently implemented does anything like I wanted to do, so you're right, it probably wouldn't get accepted even if it could be used.
Maybe I'll find something else to dabble with for now :-/
Cheers, Jon