Re: Idempotence and "Replication Insensitivity" are equivalent ?

Aloha Kakuikanu wrote:
> Jan Hidders wrote:
> > You'd split that into computing (minDate, maxDate) and then do some
> > post-processing. For (minDate, maxDate) you get:
> > - E = (null, null)
> > - S(d) = (d,d)
> > - A((d1, d2), (d3, d4)) = (MIN(d1, d3), MAX(d2, d4))
> > with MIN and MAX defined such that MIN(d, null) = MIN(null, d) = MAX(d,
> > null) = MAX(null, d) = d.
>
> This null thingy doesn't make sence:
>
> MIN(MAX(d, null),null) = null (by absorption)
> MIN(MAX(d, null),null) = d (by applying your rule twice)
>

According to ANSI SQL, nulls are ignored in the aggregate functions (except count(*)). It may not make sense to you, but it makes sense to a lot of other folks.

> It is obvious that you are after neutral elements -oo and +oo such that
>
> MIN(d, -oo) = -oo
> MAX(d, -oo) = d
> MIN(d, +oo) = d
> MAX(d, +oo) = +oo

What's +oo in, say, the set of integers ? Received on Fri Sep 22 2006 - 23:37:04 CEST