Re: c.d.theory glossary - RELATION

From: mAsterdam <mAsterdam_at_vrijdag.org>
Date: Fri, 30 Apr 2004 17:18:43 +0200
Message-ID: <40926e5d$0$562$e4fe514c@news.xs4all.nl>

x wrote:

> mAsterdam wrote:

>>This is where a difference come in:
>>
>>In ER-modelling the parts of the model are
>>abstractions of (real or fantasized) world _things_
>>and associatons between them, vs.
>>the relation values in a relational model represent _facts_
>>(or beliefs) about a (real or fantasized) world.
>>
>>In short
>>RM.REPRESENT yes, *facts*.
>>ER.REPRESENT yes, *things*.

> 
> Do  some of the _facts_  about a (real or fantasized) world
> tell something about _things_  in a (real or fantasized) world ?

I really don't know. Thinking out loud:
Say we have a history of axioms, and we would create a database for searching purposes - we'ld probably have no use for it if it didn't somehow represent persons or documents.

An empty database - would it qualify as a proper database? A database without any facts about things - would it qualify?

> If yes, those _things_ are explicitly or implicitly represented > in relational model ? And by what ? Why not otherwise.

Assuming yes about the first question:
'represent' would be 'explicit', imho.
By what? By the propositions (by means of the predicates).

As I said: just thinking out loud. Received on Fri Apr 30 2004 - 10:18:43 CDT