Path: dp-news.maxwell.syr.edu!spool.maxwell.syr.edu!drn.maxwell.syr.edu!news.maxwell.syr.edu!postnews.google.com!f14g2000cwb.googlegroups.com!not-for-mail From: "JOG" Newsgroups: comp.databases.theory Subject: Re: What does this NULL mean? Date: 14 Dec 2005 07:05:07 -0800 Organization: http://groups.google.com Lines: 41 Message-ID: <1134572707.151136.207630@f14g2000cwb.googlegroups.com> References: NNTP-Posting-Host: 128.243.220.41 Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" X-Trace: posting.google.com 1134572712 18304 127.0.0.1 (14 Dec 2005 15:05:12 GMT) X-Complaints-To: groups-abuse@google.com NNTP-Posting-Date: Wed, 14 Dec 2005 15:05:12 +0000 (UTC) User-Agent: G2/0.2 X-HTTP-UserAgent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-GB; rv:1.7.12) Gecko/20050919 Firefox/1.0.7,gzip(gfe),gzip(gfe) X-HTTP-Via: 1.1 cache3.nottingham.ac.uk:3128 (squid/2.5.STABLE3) Complaints-To: groups-abuse@google.com Injection-Info: f14g2000cwb.googlegroups.com; posting-host=128.243.220.41; posting-account=WQE_Eg0AAACz0MT8Z7vkSN2k8y0AQ1oA Xref: dp-news.maxwell.syr.edu comp.databases.theory:35123 mountain man wrote: > "David Cressey" wrote in message > news:K2Qnf.2517$0z.2002@trndny02... > > > > "mountain man" wrote in message > > news:ttrnf.19034$ea6.17392@news-server.bigpond.net.au... > > > >> Dependent upon the measure of complexity of your database > >> it may be argued that the design is never complete because it > >> is under constant evolution. > > > > [snip] > > > > Your post is outstanding, perhaps the best I've read in this thread. > > > > I think "change management in databases" deserves a discussion of its own. > > > There has been limited discussion here, and there are articles > about on this issue under the keyphrase of "schema evolution". > > The issue is avoided by Date et al (AFAIK) whereas it is > precisely this issue of schema change that should be subject > to the principles espoused by the relational model. > > The RM guides the design of new databases, and it obviously > also must guide the evolution of old databases, but this aspect > has not yet been thrashed out in any formalised manner. These sort of considerations are the exact reason that boundaries are being pushed in areas like bioinformatics, where knowledge and understanding of the worldview being modelled is constantly improving and shifting - this leads to the schema they work with being in a state of constant flux. All too often developments appear to be performed in an ad hoc manner, but I predict that eventually, it will be from these sort of areas that advancements will emerge (with hopefully a theoretically sound as opposed to purely pragmatic solutions). Now this flux happens in the business world too, but to a lesser extent where it is all too easy to try and paper over the cracks, and I just don't see the impetus for advances stemming from there.