Why do "yes/no" values in Microsoft Access map to -1 and 0?

Question

The use of -1 puzzles me. I would have expected the values to be:

Is there an historical reason behind the use of -1 instead of +1?

One of the reasons for this are because booleans within MS Access can also have no value or Null values. As I thought the reason for it being -1 is because of the type of integer data type a boolean is defined as. "This is because the Boolean data type is stored as a 16-bit signed integer." The actual binary value of -1 is 1111111111111111. This question has been asked several times on SO http://stackoverflow.com/questions/8827447/why-is-yes-a-value-of-1-in-ms-access-database — Ramhound, Dec 30 '13 at 17:14
@Ale - This also comes down to base 2 arithmetic. A bitwise NOT of anythng but 11111111 11111111 would not result in 00000000 00000000. The logical bitwise NOT of False should be True and vice versa. — Ramhound, Dec 30 '13 at 17:21
@Ramhound Actually, Yes/No fields in Access cannot be Null, only Yes/True or No/False. If you try to force a Yes/No field to Null it simply becomes No/False. — Gord Thompson, Feb 02 '14 at 13:24

score 5 · Accepted Answer · edited May 23 '17 at 12:41

As discussed on Stack Overflow here, Yes/True values appear in Access as -1 because

Yes/No fields in Access mimic bit fields,
Integer values in Access are signed, two's complement values,
No/False is represented by 0, and
the only other two's complement value of a 1-bit number is -1. To illustrate, two's complement numbers can have the following values:

3-bit:

bits  integer
----  -------
000         0
001         1
010         2
011         3
100        -4
101        -3
110        -2
111        -1

2-bit:

bits  integer
----  -------
00          0
01          1
10         -2
11         -1

1-bit:

bits  integer
----  -------
0           0
1          -1

For the integer representation of a bit field (i.e., 1-bit), if 0 is No/False then the only other value available for Yes/True is -1.

1 Answers1