Robert Sanderson writes:
 > >	1:10 -- First ten characters
 > >	0:10 -- *error*
 > >	1 -- Whole string (from first to end)
 > >	2 -- All of string except the first character
 > >	1:-4 -- All of string except last four characters
 > >	2:-1 -- All of string except first and last characters
 > 
 > Like it.
 > 
 > Two questions:
 > 
 > 1.  Is 1:0 an error, or "" ?

I don't see any reason for it to be an error.  It's an odd thing to
ask for, yes, but as always I take the position that we can't know
what use people will put our stuff to, so if as in this case there is
a single meaningful interpretation of an expression, we should let
that interpretation stand.  So <anything>:0 is a zero-length string.

 > 2.  Is it sufficiently useful to also allow negative numbers in the
 >     start position?  eg:  -10:3 -- 3 characters starting at the 10th
 >     from last.

I'd have no objection to that (though again I don't see any use for it
right now).

 _/|_	 ___________________________________________________________________
/o ) \/  Mike Taylor  <[log in to unmask]>  http://www.miketaylor.org.uk
)_v__/\  "I have no idea what a `strong' or a `weak' verb is, and I truly
	 hope I never find out what a `turbo verb' is, but I usually have
	 a pretty good idea what the *right* verb is" -- Jane MacDonald.