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

From: Roy Hann <specially_at_processed.almost.meat>
Date: Sun, 6 Feb 2005 11:39:10 -0000
Message-ID: <rpWdnZ4au7b4n5vfRVn-3w_at_pipex.net>


"Dan" <guntermann_at_verizon.com> wrote in message news:K1kNd.6968$uc.2841_at_trnddc04...

>

> Everyone won't agree with this particular opinion of course. I can accept
> that.

Well I do! There is nothing in Codd's model that precludes RVA's, but there definitely isn't anything in it that allows you to operate on an RVA value as a relation either. Date had to add that, and that means it's not the same model. (Which is not to say Date's is wrong, or that it's worse, but it is definitely not what I mean when I talk about THE relational model. BTW, I suspect it IS worse, for reasons I only dimly suspect at the moment.)

This group suffers from chronic (and possibly terminal) equivocation, and you have pointed out just another example. It is the reason I barely ever bother looking in on it any more.

Roy Received on Sun Feb 06 2005 - 12:39:10 CET

Original text of this message