Re: Database schema for univesal usage

From: David Cressey <david.cressey_at_earthlink.net>
Date: Mon, 30 May 2005 13:24:25 GMT
Message-ID: <duEme.10373$uR4.2008@newsread2.news.atl.earthlink.net>

"Kenneth Downs" <knode.wants.this_at_see.sigblock> wrote in message news:haerm2-nfr.ln1_at_pluto.downsfam.net...

> This is the part of this that confuses me. In this Dec system, is the
data
> dictionary functional, meaning that the computer is making us of it, or is
> it documentational, like a word processing document?

I would have to say that it's both. It's functional, but one of the functions is to present itself like a document.

The compilers, and the SQL DDL executor (my name, not theirs), can all reference the data dictionary to resolve dictionary references in the source code or the DDL. (You'll pardon me, but I view SQL DDL as "source code").

However, the dictionary is typically not needed at runtime. Exception: DEC Datatrieve, which resolves every data name the user enters via the dictionary.

The dictionary can contain "database entries", which point to a database. The entire database metadata can be construed as a "subdictionary" based on this entry, provided you are willing to do some mental gymnastics, and be tolerant of the engineers.

The DEC Data Dictionary was possibly the most misunderstood product within the DEC community. Received on Mon May 30 2005 - 08:24:25 CDT