Path: news.easynews.com!easynews!hub1.nntpserver.com!headwall.stanford.edu!newsfeed.stanford.edu!postnews1.google.com!not-for-mail
From: brian-l-smith@uiowa.edu (Brian Smith)
Newsgroups: comp.databases.theory
Subject: Re: UNIQUE and NULL in SQL
Date: 31 Dec 2001 20:06:09 -0800
Organization: http://groups.google.com/
Lines: 15
Message-ID: <60360d48.0112312006.4a0d51cd@posting.google.com>
References: <60360d48.0112291928.2e7c5818@posting.google.com>
NNTP-Posting-Host: 24.182.69.87
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
X-Trace: posting.google.com 1009857970 15697 127.0.0.1 (1 Jan 2002 04:06:10 GMT)
X-Complaints-To: groups-abuse@google.com
NNTP-Posting-Date: 1 Jan 2002 04:06:10 GMT
Xref: easynews comp.databases.theory:19380
X-Received-Date: Tue, 01 Jan 2002 01:51:11 MST (news.easynews.com)

brian-l-smith@uiowa.edu (Brian Smith) wrote in message news:<60360d48.0112291928.2e7c5818@posting.google.com>...
> What is the rationale for having a UNIQUE constraint treat NULLS as
> equal?

I see I made a big mistake in not specifically refering to SQL
database, and in particular the ANSI SQL standard. I was really just
wanting to know the ANSI commitees rationale for this behavior. My
guess is that the ANSI SQL commitee just standardized on what current
implementations were doing at the time. Is my guess wrong?

Please keep in mind that I am asking about "SQL theory" not
"relational theory".

Thanks,
Brian