	The edit/write inode stuff is still very much ALPHA.  So far
I've tested ext2, minix, and xiafs on a 172k partition and they worked
under minimal testing.  I copied the partition to a file, modified one
block pointer, wrote it, then did a "cmp -l" on the partition and the
copy and it reported only one block different (with the correct values
even).  Not bad, but I would appreciate it if someone could verify
something similar on a larger scale, particularly the ext2fs.
