Path: utzoo!attcan!uunet!lll-winken!uwm.edu!zaphod.mps.ohio-state.edu!mips!dimacs.rutgers.edu!rutgers!rochester!pt.cs.cmu.edu!o.gp.cs.cmu.edu!andrew.cmu.edu!dc9l+ From: dc9l+@andrew.cmu.edu (David H. Campbell) Newsgroups: comp.databases Subject: Re: Ingres Question: Duplicate Rows Message-ID: Date: 5 Oct 90 17:45:40 GMT References: <1990Aug8.230913.2897@agate.berkeley.edu> <26867@pasteur.Berkeley.EDU> <12314@blia.BLI.COM>, <1990Oct4.193912@Unify.com> Organization: Administrative Systems, Carnegie Mellon, Pittsburgh, PA Lines: 17 In-Reply-To: <1990Oct4.193912@Unify.com> The issue with duplicates for me (personally, not officially) is that they are often needed in order to model reality. In some cases there may be two exactly duplicate transactions which are BOTH important. It is not proper (in the REAL world, not some relational theory) to ARBITRARILY delete or ignore one of them. This type of behavior makes it virtually impossible to use a relational system for financial applications, and even when workarounds are possible the most powerful functions of SQL are not usable. I don't personally advocate having duplicate transactions in some cases, but as a designer I have to work with the situation as it exists. I don't think it's proper for a relational system to IMPOSE a rigid methodology on the world. Computers are supposed to serve people and organizations, not vice versa. David Campbell P.S. These opinions are my own and not necessarily my employer's.