Re: c.d.theory glossary - RELATION

x wrote:

> mAsterdam wrote:

> We have some issues:

Yes. I don't pretend I can solve them all, nor am I going to try in one blow. I would
appreciate comments by others, I'll see how I can make something out of it for the glossary.

Just some remarks for now.

> RELATIONs vs. RELATIONSHIPs

Namespaces to make some distance? I mean in this case:

     RM.RELATION vs. ER.RELATIONSHIP

> REPRESENTED vs. DESCRIBED
> RELATION(SHIP)s vs RELATION(SHIP)s SET
> FACT vs. THING (ENTITY).
> First Order Logic vs. Higher Order Logic.
Count me out. Help!

> What is the equivalent of an ENTITY(SET) in the RM ?
I don't think there can be equivalence. There is no possible representation for values in ER-modelling, no room for ENTITIES in RM. I do think a mapping is possible: An ER-graph can be like a roadmap to a relational database.

Traditionally there can be Multivalued ATTRIBUTES in ER, RM has atomic ATTRIBUTES.
So: RM.ATTRIBUTE and ER.ATRRIBUTE ?

> Make sense to talk about ATTRIBUTES of a FACT ?
> How are those different from ATTRIBUTES of an ENTITY ?
> In ER modeling, a RELATIONSHIP is defined over ENTITIES:
> "A relationship is an association between several entities."
> In RM, a RELATIONSHIP is defined over VALUEs.
> What difference is between ENTITIES and VALUEs ?
>
>

>>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?

And that's why they lose the semantics, they are not talking about proper RM.RELATIONs, but about tables. So they are not losing semantics by translating, they are losing by translation incorrectly. Received on Mon May 03 2004 - 16:56:09 CEST