Path: dp-news.maxwell.syr.edu!spool.maxwell.syr.edu!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!postnews.google.com!f6g2000cwb.googlegroups.com!not-for-mail
From: "-CELKO-" <jcelko212@earthlink.net>
Newsgroups: comp.databases.theory
Subject: Re: Foreign key problem
Date: 15 Jun 2006 12:43:16 -0700
Organization: http://groups.google.com
Lines: 22
Message-ID: <1150400596.671376.85270@f6g2000cwb.googlegroups.com>
References: <1150175956.098492.15000@i40g2000cwc.googlegroups.com>
   <1150328893.466491.91190@r2g2000cwb.googlegroups.com>
   <AIbkg.3574$oE3.2321@trndny04>
NNTP-Posting-Host: 63.246.191.202
Mime-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
X-Trace: posting.google.com 1150400601 10082 127.0.0.1 (15 Jun 2006 19:43:21 GMT)
X-Complaints-To: groups-abuse@google.com
NNTP-Posting-Date: Thu, 15 Jun 2006 19:43:21 +0000 (UTC)
In-Reply-To: <AIbkg.3574$oE3.2321@trndny04>
User-Agent: G2/0.2
X-HTTP-UserAgent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.0.4) Gecko/20060508 Firefox/1.5.0.4,gzip(gfe),gzip(gfe)
Complaints-To: groups-abuse@google.com
Injection-Info: f6g2000cwb.googlegroups.com; posting-host=63.246.191.202;
   posting-account=EzrcbQ0AAACimcwOd0QEJhKx5W98nxwt
Xref: dp-news.maxwell.syr.edu comp.databases.theory:41489

>> People sometimes build databases whose subject matter is meta-data. <<

Yes, and it is usually an EAV mistake

>>   A data-based data dictionary is an example. Jay Dee may be doing something along these lines. <<

I do not think so.  A data dictionary would have validation ("ZIP codes
are five digits") and verification ("ZIP codes and states match up this
way!") modeled in it.  There would be external data sources ("we buy
ZIP data tapes from the USPS every six months").  There would be syntax
rules ("the following suffixes are allowed on data element names..").
Etc.

I have done a schema for DFD which was a lot of meta-data type stuff
("a process must have at least one inflow, and at least one outflow",
"In and out flows must balance at all levels of abstraction", etc.)
But that is not what we have here.

This looks like a straight attempt at the Schema Information tables,
with some worry about physical implementation details.  What am I
missing here?

