Re: What is that "more" that makes E-R model truly independent ?

beginner16 schreef:

>>> But as I stated before, E-R model diagram would be the same no matter
>>> if logical  DB implementation is hierarchical or relational. For that
>>> reason I assume that  ER modeling is more than just using symbols for
>>> entity, attributes, relationships

>> It is not.

> What do you mean by "it is not"?
> That E-R modeling is nothing more than just a bunch of symbols or that
> the structure of E-R model does depend on whether underlying logical
> structure is relational or hierarchical ( well isn't the whole point
> of E-R model to be independent, at least for the most part, of
> underlying logical structure ) ?

AFAIK Alfredo does not post riddles, so indeed.

Maybe it helps you to make the distinction between graph and model.

E-R is a group of (mostly kind of sloppy, some with overprecision on unimportant details) modeling languages with a partly graphical syntax.

The graph can only highlight a part of the whole model - that is to say there is no /underlying/ anything, there is just visible and unvisible parts of the model.

>>> But what is that "more" that makes E-R model truly independent of logical DB
>>> implementation ?
>> It is more vague and more imprecise. Being even more vague we could be
>> even more independent, and saying nothing we achieve complete
>> independence :-)
>>

Mention your sources for them and we might be able to compare. Received on Tue Jul 24 2007 - 18:17:20 CEST