Well folks, I have some good news. I seem to have revived my board
which I thought had something actually wrong with the chip.

After giving up on the board for awhile, I decided to use MPSIM while
I waited for new boards to arrive. It felt a bit silly to replace the
chip and then have new boards the next day. Anyway, it seems that
while writing code I had created a stack overflow. Bad coding on my
part. I fixed that. Then I had a wild idea...why not try it in the
chip again. I hooked the PK3 back up, connected it to the board,
programmed, and voila, it worked fine.

So...I'm not sure what this really tells us. I understand how a stack
overflow might mess up the debug executive programmed into the chip. I
don't know why a stack overflow in my PIC program would cause MPLAB
itself to crash and corrupt the project files. I can't even load
those, let alone get in far enough to fix the error in my code.
Perhaps it's something in the communication between the PK3 and MPLAB
that isn't terminating properly? What's interesting is that this
problem crippled my ICD2 as well.

Anyway, I'd be interested to see if anyone else can duplicate this.
Just beware and don't do it when you've got a deadline looming!

Thanks everyone for the suggestions. Hopefully this won't show up
again, but if it does at least I have a start on where to look...my
coding.

Josh
-- 
A common mistake that people make when trying to design something
completely foolproof is to underestimate the ingenuity of complete
fools.
        -Douglas Adams
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist