Re: Views for denomalizing
Date: Thu, 03 Feb 2005 15:03:26 GMT
Message-ID: <23rMd.8555$S3.3018_at_newsread2.news.atl.earthlink.net>
"Dawn M. Wolthuis" <dwolt_at_tincat-group.comREMOVE> wrote in message
news:cts9lf$9h3$1_at_news.netins.net...
> The top of this posting is intended to be devoid of opinions, so please
> correct if I have any misconceptions prior to the question.
>
> 1) SQL-DBMS's (at least those that conform to SQL92) provide no
constraints
> on the user creating new tables to restrict base tables from being
> denormalized EXCEPT in the case of the first normal form.
On May 7, 2004, mAsterdam (a really good contributor to this newsgroup) provided a correction to your definition of 1NF. In particular, mA mentioned that part of 1NF included the no duplicates allowed rule. At that time, your responses seem to indicate that you accepted and internalized mA's comment.
Now you are starting essentially the same thread you did last year, but
you've once again dropped the no duplicates allowed part of 1NF. Did you
forget what you picked up almost a year ago? Are you going to start the
same conversation again in 2006? Are we going to have to cover this ground
again?
.
Incidentally, it's possible to "get around" the repeating groups part of
the 1NF rule by "faking out" the metadata.
Essentially, if you crosstabulate a relational table, you get data that is
not in 1NF. By suitably disguising the column names, you can make it look
like there are no repeating groups, even though there are.
The reason I mention this is that I've seen more than one SQL based database in the field that was trying to operate on crosstabulated data. As one might expect, they got neither the flexibility and power that one expects from the RDM, nor the performance they needed in the real world.
This is not an argument against crosstabulating data. It's just an argument against storing crosstabulated data in an SQL table, and then pretending that it's relational.
This may be somewhat tangential to the point you are trying to make. But it's important enough so that it needs to be mentioned. Hopefully not again, next year. Received on Thu Feb 03 2005 - 16:03:26 CET