Re: thinking about UPDATE

From: Mikito Harakiri <mikharakiri_at_iahu.com>
Date: Thu, 22 Jul 2004 17:50:48 -0700
Message-ID: <StZLc.62$X73.387_at_news.oracle.com>


"D Guntermann" <guntermann_at_hotmail.com> wrote in message news:I1A3o9.CHr_at_news.boeing.com...
> "Mikito Harakiri" <mikharakiri_at_iahu.com> wrote in message
> news:QAXLc.59$X73.404_at_news.oracle.com...
> > Perhaps more intuitive reasoning may be the observation that the key
> > property is monotonic:
> > 1. Key containing all the columns of the relation is always valid.
> > 2. Key containing some of the columns of the relation might be valid or
> not.
> > Unlike cases #1, and #3 this is actually the interesting (or nontrivial)
> > one.
> > 3. Key containing none of the columns of the relation is almost never
> valid.
>
> Hi Mikito,
>
> Would you mind clarifying what you mean be "almost never valid" in rule 3,
> above? Is there a way or criterion to discern when it is and when it is
not
> valid?

"Is almost never valid" was a shorthand for "is valid for the relations having no more than one tuple only":-) Received on Fri Jul 23 2004 - 02:50:48 CEST

Original text of this message