Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> comp.databases.theory -> Re: A Normalization Question

Re: A Normalization Question

From: Neo <neo55592_at_hotmail.com>
Date: 7 Jul 2004 19:02:57 -0700
Message-ID: <4b45d3ad.0407071802.77d0098f@posting.google.com>


> It is not LOGICALLY redundant to record that "Car X is brown"
> and "Dog Y is brown", because these are two different facts.

It is true that "Car X is brown" and "Dog Y is brown" are two separate and different facts thus they can't be normalized. But these two facts are built upon other facts (which are stored in the db) one of which is that both the car and the dog are the same color which has the name brown. In your example, which is slightly different than the others examples in this thread, the color named brown is redundant and could be normalized as follows:

"Car X is brown" and "Dog Y is ->brown" or
"Car X is ->brown" and "Dog Y is ->brown" and "brown isa color"

> The relational model is LOGICAL and is not concerned with the physical
> issues.

The above is meaningless in demonstrating that the color brown within the two facts is not redundant.

> If your argument had any merit at all, an RDBMS
> implementation could in fact store every data value only once, and
> then physically point to it from all other records. For all you know,
> maybe that IS what some RDBMSs do (it isn't, but it would be
> impossible to tell using SQL).

The above is meaningless in demonstrating that the color brown within the two facts is not redundant. Received on Wed Jul 07 2004 - 21:02:57 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US