Re: By The Dawn's Normal Light

From: Laconic2 <laconic2_at_comcast.net>
Date: Fri, 22 Oct 2004 09:05:38 -0400
Message-ID: <iMOdncX2X68wmOTcRVn-2g_at_comcast.com>


"Paul" <paul_at_test.com> wrote in message news:4178cd36$0$47980$ed2e19e4_at_ptn-nntp-reader04.plus.net...
> To clarify it a bit more, maybe it should be:
>
> "A Relation is in first normal form if and only if none of the domains
> of its attributes permit compound or multivalued values *from the point
> of view of the relational engine*."

As I said to Marshall, I'm prepared to retreat to a standard definition if there is one.

>
> Of course, a string could be viewed as a compound value by the type
> engine, which is why you get type operators like "substring" etc. But
> from the point of view of the relational engine, a string is atomic:
> there is no way for the relational operators to break it down into
> smaller pieces.

The issue of "compound" gets even worse. Every value that is represented by more than one bit is "compound" at some layer of representation/interpretation. The people who like to speak in abstractions like to ignore this fact. Often, that's a useful way to think. Abstraction is, after all, one of the fundamental tools the human mind uses to cope with complexity. Received on Fri Oct 22 2004 - 15:05:38 CEST

Original text of this message