Re: Can we solve this -- NFNF and non-1NF at Loggerheads

"Alan" <not.me_at_rcn.com> wrote in message news:36mrv7F52lmo2U1_at_individual.net...
>
> "Roy Hann" <specially_at_processed.almost.meat> wrote in message
> news:9LCdna8B3LW2VJjfRVn-rg_at_pipex.net...
> > "Alfredo Novoa" <alfredo_novoa_at_hotmail.com> wrote in message
> > news:h6da01hfpdq3f6fg384g5l0t5dd7dd1std_at_4ax.com...
> > > On Fri, 4 Feb 2005 13:37:17 -0500, "Alan" <alan_at_erols.com> wrote:
> > >
> > > >Nothing has changed.
> > >
> > > It has changed, but it seems that you are not aware of that.
> > >
> > > > 1NF means (and has always meant) that all values are
> > > >atomic (simple, indivisible)
> > >
> > > Atomic is not a precise word. We can not base any precise definition
> > > on that term.
> >
> > Alan is confused, but your comment, erudite though it is, is irrelevant.
> >
> > 1NF does not "mean values are simple or indivisible".
>
> Well, then Elmasri and Navathe are confused, as I quoted directly from the
> book (which is used in many leading universities).

Then Elmasri and Navathe are confused. (And I don't care how many universities use a book. Repetition does not make truth out of error.)

> I am not the one who is confused. You just redefined "atomic" as meaning
> "divisible".

No, I made no attempt to define atomic at all. I said Codd used it as a shorthand for "I don't care if it's divisible or not". Since he certainly never defined it, that is all he could mean.

> It Codd intended 1NF to include divisible attributes, he would
> have used the word "divisible", not "atomic".

No, because he didn't care one way or the other. Saying attributes are atomic was just his way of turning his back on an uninteresting (to him) distraction.

Roy Received on Wed Feb 09 2005 - 12:58:06 CET