Re: c.d.theory glossary - RELATION

From: mAsterdam <mAsterdam_at_vrijdag.org>
Date: Fri, 30 Apr 2004 20:34:53 +0200
Message-ID: <40929c58$0$36169$e4fe514c@news.xs4all.nl>

x wrote:

> mAsterdam wrote:


>>>>"John likes Mary" would be in the body of a relation.

>>>>It's predicate would be (short) "<Person> likes <Person>".

>>>
>>>Why we don't directly represent John in the database ?
>>
>>Because of the context we were discussing this in.
>>We were discussing RM.RELATION vs ER.RELATION.
>>
>>AFAIK this (see previous post) is the way
>>RM goes about representing facts, ER has no way
>>of showing the facts in the body of a RM.RELATION,
>>it only represents what would be in the heading.

> 
> Sorry. The text at:
>  http://www.cs.sfu.ca/CC/354/zaiane/material/notes/Chapter2/node1.html
> made a distinction between:
> - ENTITY and ENTITY SETs
> - RELATIONSHIP and RELATIONSHIP SETs

A rephrase in their terminology would be: "ER-modelling has no way of representing an individual ENTITY, it only represents ENTITY SETs." But now I'am back where we started, because this would be a statement *within* ER-modeling.

At the same site there is
http://www.cs.sfu.ca/CC/354/zaiane/material/notes/Chapter2/node10.html, on how to get from ER to RM. They call it 'reducing' an ER-model to a relational model, which is BTW qualified as a 'record oriented' approach (I guess they needed something as opposed to object oriented). I don't see how this 'reduces'. More like 'expands', no? Received on Fri Apr 30 2004 - 13:34:53 CDT