home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
ftp.cs.arizona.edu
/
ftp.cs.arizona.edu.tar
/
ftp.cs.arizona.edu
/
tsql
/
doc
/
tsql.mail
/
000076_rts _Thu Apr 8 10:59:44 1993.msg
< prev
next >
Wrap
Internet Message Format
|
1996-01-31
|
1KB
Received: from boojum.CS.Arizona.EDU by optima.cs.arizona.edu (5.65c/15) via SMTP
id AA08626; Thu, 8 Apr 1993 10:59:45 MST
Date: Thu, 8 Apr 1993 10:59:44 MST
From: "Rick Snodgrass" <rts>
Message-Id: <199304081759.AA26401@boojum.cs.arizona.edu>
Received: by boojum.cs.arizona.edu; Thu, 8 Apr 1993 10:59:44 MST
To: tsql@cs.arizona.edu
Subject: Re: more on the benchmark
> From: UZTBB@CUNYVM.BITNET
>
> I still feel strongly that we should not look for 3Nf relations.
> How does TSQL community feel about this?
> Abdullah
>
It depends on whether "3Nf" refers to the model-independent schema,
using the definition as "all snapshots are in 3NF" or it refers to
a particular representation. As the benchmark is to be independent
of representation, the snapshot interpretation should be used.
Given this, it seems that little is lost, and much is gained, by
having the highest feasible normal form. For this reason, I advocate
attempting BCNF, which seems to be possible for the schemas under
consideration.