Comments on: Why NULL never compares false to anything in SQL http://www.xaprb.com/blog/2006/05/18/why-null-never-compares-false-to-anything-in-sql/ Stay curious! Thu, 09 Feb 2012 20:41:20 +0000 hourly 1 http://wordpress.org/?v=3.2.1 By: Aaron http://www.xaprb.com/blog/2006/05/18/why-null-never-compares-false-to-anything-in-sql/#comment-19862 Aaron Thu, 02 Feb 2012 21:29:02 +0000 http://www.xaprb.com/blog/?p=149#comment-19862 Not quite related, but I guess I would have written a different query to that in the example: select shoeStyle, count(color) as Count, sum(color = 'red') as redCount, sum(color = 'green') as greenCount, sum(color = 'blue') as blueCount from bowlingShoes group by shoeStyle; As an old C programmer, I love abusing types. Not quite related, but I guess I would have written a different query to that in the example:
select shoeStyle,
count(color) as Count,
sum(color = ‘red’) as redCount,
sum(color = ‘green’) as greenCount,
sum(color = ‘blue’) as blueCount
from bowlingShoes
group by shoeStyle;

As an old C programmer, I love abusing types.

]]> By: jflo http://www.xaprb.com/blog/2006/05/18/why-null-never-compares-false-to-anything-in-sql/#comment-19482 jflo Thu, 07 Jul 2011 23:02:12 +0000 http://www.xaprb.com/blog/?p=149#comment-19482 Thanks!! Thanks!!

]]> By: Fred Zarguna http://www.xaprb.com/blog/2006/05/18/why-null-never-compares-false-to-anything-in-sql/#comment-19446 Fred Zarguna Thu, 23 Jun 2011 05:30:24 +0000 http://www.xaprb.com/blog/?p=149#comment-19446 Your two predicates are also incomplete, and do not actually encompass the non-propositional logic of SQL. To wit, there is a third predicate required to describe the state of NULL: 3. No value can ever be unequal to a non-value. Your two predicates are also incomplete, and do not actually encompass the non-propositional logic of SQL. To wit, there is a third predicate required to describe the state of NULL:

3. No value can ever be unequal to a non-value.

]]> By: Fred Zarguna http://www.xaprb.com/blog/2006/05/18/why-null-never-compares-false-to-anything-in-sql/#comment-19445 Fred Zarguna Thu, 23 Jun 2011 05:26:47 +0000 http://www.xaprb.com/blog/?p=149#comment-19445 read NULL 5 as NULL <> 5 in my post. Your posting mechanism treated my > and < as html... read NULL 5 as NULL <> 5 in my post. Your posting mechanism treated my > and < as html…

]]> By: Fred Zarguna http://www.xaprb.com/blog/2006/05/18/why-null-never-compares-false-to-anything-in-sql/#comment-19444 Fred Zarguna Thu, 23 Jun 2011 05:25:11 +0000 http://www.xaprb.com/blog/?p=149#comment-19444 Your "proof" that "NULL 5 is not true" is, in fact, nonsense. This statement cannot be proved or disproved in the context of the propositional calculus, which is what you're attempting to do: in the propositional calculus, an admissible statement is either true or it is false. Your "proof" that NULL 5 produces a contradiction propositionally implies that NOT(NULL 5) is true. This statement is also not true. Therefore, it is not really a propositional statement. Neither the statement NULL 5 nor ~(NULL 5) is true. By the same token, neither statement is false either. They are not propositional statements; they cannot be proven. [The additional restriction in your "proof" that NULL be a finite real is unnecessary. In the domain of extended reals, {x | x 5} can certainly include x = +- infinity. ] Your “proof” that “NULL 5 is not true” is, in fact, nonsense.

This statement cannot be proved or disproved in the context of the propositional calculus, which is what you’re attempting to do: in the propositional calculus, an admissible statement is either true or it is false. Your “proof” that NULL 5 produces a contradiction propositionally implies that NOT(NULL 5) is true. This statement is also not true. Therefore, it is not really a propositional statement. Neither the statement NULL 5 nor ~(NULL 5) is true. By the same token, neither statement is false either. They are not propositional statements; they cannot be proven.

[The additional restriction in your "proof" that NULL be a finite real is unnecessary. In the domain of extended reals, {x | x 5} can certainly include x = +- infinity. ]

]]>