Re: So what's null then if it's not nothing?

In article <1133288703.505685.293960_at_o13g2000cwo.googlegroups.com>, boston103_at_hotmail.com says...
>
> Jon Heggland wrote:
> > Well, like I said, Codd conflates the truth value "unknown" with NULL.
> > I'm not sure that is a good idea.
>
> He does.

Well, of course he agrees with himself. But who else does? How does he justify it? What are the consequences?

> > I am curious: In Lukasiewicz's system, what do you get when you compare
> > the unknown truth value to itself?
>
> Lukasiewicz's logic as far as I remember deals only with logical
> connectives and "truth" values (0, 1, 2). Its truth table coincides
> with Codd's 3VL, but I believe it has nothing to say about the value1
> <comp> value2.

http://en.wikipedia.org/wiki/Ternary_logic:

"Note that any two statements with the same truth value are equivalent, even if the truth value is unknown."

The truth table for equivalence (EQUALS, <->), 0=unknown, 1=true, 2 =false:

P Q P <-> Q

0 0     1
0 1     0
0 2     0
1 0     0
1 1     1
1 2     2
2 0     0
2 1     2
2 2     1

> Codd just stipulates that the comparison results in
> unknown if either operand is NULL. You are free to redefine ;)

No, he also stipulates that the unknown truth value is the same as NULL, which means we are not able to represent the unknown truth value faithfully. This is a blunder analogous to saying that the empty string is NULL.

Do you know how Codd defines implication in his 3VL?

-- 
Jon