🚀 go-pugleaf

In article <1992Aug29.232155.14657@sq.sq.com> msb@sq.sq.com (Mark Brader) writes:
>>> ... The representation of [decimal] 10000 has to use the bit
>>> sequence 10011100010000.  
>> As I read it, wrong.  Nothing I see prevents use of 2's complement, 
>> 1's complement, or excess-2**(n-1) (the sign bit being reversed from
>> 2's complement).
> These systems are indeed all permissible.  But they also all use the
> bit sequence 10011100010000 in the representation of decimal 10000.

Both wrong. Excess-2**(n-1) is not a permitted system, because the
representation of decimal 10000 is not binary 10011100010000. The
standard requires that the representation of integers be a "pure-binary"
system, except for the sign bit, and that unsigned and signed integers
use the same representation for numbers where their ranges overlap.

-- 
Clive D.W. Feather     | IXI Limited         | If you lie to the compiler,
clive@x.co.uk          | 62-74 Burleigh St.  | it will get its revenge.
Phone: +44 223 462 131 | Cambridge   CB1 1OJ |   - Henry Spencer
Fax:   +44 223 462 132 | United Kingdom      |