Re: So what's null then if it's not nothing?
Date: 26 Nov 2005 13:36:18 -0800
Jon Heggland wrote:
> In article <1132926804.625596.34020_at_g47g2000cwa.googlegroups.com>,
> boston103_at_hotmail.com says...
> > michael_at_preece.net wrote:
> > [...]
> > >
> > > Any two variables/"fields" (of any type) with "no value at all" are,
> > > indeed, equal.
> > >
> > > Mike.
> > Cute.
> > So, using your valuation for equality, would a query asking for a
> > pairwise list of people living in the same country produce a correct
> > result if some pairs that happen to have nulls for their countries ?
> That, of course, depends on what you mean by a correct result, what it
> means that country is NULL, and how the query is formulated.
> If we assume that it is not generally the case that two people with
> NULLs for their countries (speaking loosely here) live in the same
> country, the query would simply have to be written so as to disregard
> people with NULL country.
So you insist that the valuation of "null=null" as true can make sense in some cases ? If so, what are those cases ? If not so, why do we need to modify our query at all ? The standard SQL does not evaluate null=null to true, so why bother writing the query "so as to disregard people with NULL country" ?
Received on Sat Nov 26 2005 - 22:36:18 CET