| Date: Sun, 9 Jan 2005 02:31:52 -0800 (PST)
 | From: bear <xxxxxx@sonic.net>
 |
 | On Sun, 9 Jan 2005, Aubrey Jaffer wrote:
 |
 | > | From: xxxxxx@philips.com
 | > | 2. When scanning different libraries of bit-twiddling, I had
 | > | stumbled across an implicit design decision that is worth
 | > | mentioning because it might swiftly break portability's neck:
 | > |
 | > |         "What is the value of (LOGAND)?"
 | > |
 | > | In my application I define (LOGAND) := 0 because the subsets my
 | >
 | >(logand) ==> -1 because (and) ==> #t.
 | >
 | >This is also necessitated because logand is associative:
 | >
 | >(logand a b) == (logand a (logand b) (logand))
 |
 | I do not understand why (logand) with zero arguments
 | ought not signal an error.  Can you enlighten me?

When dealing with variable arity, empty lists can happen.