Re: A Normalization Question
Date: Thu, 08 Jul 2004 08:41:53 -0700
Message-ID: <40ED6B41.253_at_assist.org>
Neo wrote:
> With respect to dbs, normalization is the process of eliminating or
> replacing duplicate things with a reference to the original thing
> being represented.
Nope. That's just tokenizing data.
If this is normalization, can you tell me what data anomalies you avoid by doing this? I know Jan already asked you this, but apparently I missed your answer.
> Within the context of a db, duplicate references
> are not considered redundant because they are unrelated to the thing
> being represented.
Ah, how convenient for you. You get to cling to definitions
that suit you (like the above) while ignoring ones (like the
definitions for redundancy and normalization) which don't.
How about numbers? Do you tokenize those? Given the
following schema:
create table person_pets (
person_id numeric(5) not null,
number_of_dogs int,
number_of_cats int,
number_of_fish int,
primary key (person_id)
)
Given the tuple (12345, 2, 2, 2) what do you do with those horrific redundant 2's?
Larry Coon
University of California
Received on Thu Jul 08 2004 - 17:41:53 CEST