char* vs void* (was: Comparison functions for qsort() and bsearch())

[Andrew Koenig]

In article <277640DD.4A70 at tct.uucp> chip at tct.uucp (Chip Salzenberg) writes:

> This statement is false.  It is entirely possible to do so, thanks to
> ANSI's guarantee that pointer-to-char and pointer-to-void will always
> have identical representations.

How's that again?  Why do you think that char* and void* have to have
identical representations?
-- 
				--Andrew Koenig
				  ark at europa.att.com