Re: database design method

You are not going wrong. The industry and its practitioners went wrong long ago and continue to fumble around in ignorance.

According to the standard definitions:

1. Humans process information while machines process data.
2. The conceptual model captures an information model and need not represent anything for machine consumption. In other words, it captures the concepts used by human beings.
3. The logical model represents some or all of a conceptual model in an abstract form suitable for machine manipulation.
4. A physical model maps the abstract logical model to actual storage structures and locations.

You asked a simple question, and you deserve the simple answer above. If you have access to a library that has them, you can verify the answer from the ISO standard vocabularies (ISO/IEC 2382)

I can think of no better confirmation of Fabian Pascal's point ( see http://www.dbdebunk.com ) regarding fundamentals ignorance ubiquitous in our industry than the example given by this newsgroup where so many practitioners can blow so much wind and completely bypass a simple straightforward answer. It's simple to conclude they are completely ignorant of all fundamentals -- even such simple ones as these.

"ad" <noSpam_at_hotmail.com> wrote in message news:pI_r9.172$OC2.19355_at_wards...
> I'm confused about the methods of database design.
>
> It seems to me that most theoretical texts talk about 'conceptual',
> 'logical' and 'physical' design. These steps produce conceptual, logical
and
> database schema. However in the 'real world' conceptual and logical steps
> seem to be combined into a single step.
>
> Academic texts seem to use the chen notation and the E-R data model to
> identify entity and relationship sets (pure conceptual design independent
of
> any DBMS, no asumption of an RDBMS implementation). Is this ever done in
the
> 'real world'?
>
> It seems to me that professionals simply engage in logical and physical
> design. They produce a normalised logical data design (identifying all
> attributes and primary and foreign keys - possibly also domains). They
also
> use a 'crows foot' notation identifying relations and the relationships
> between them. Thus they are assuming an RDMBS implementation (but no
> particular brand of RDBMS). This 'logical' design activity is sometimes
> referred to as 'conceptual design', even though an RDBMS is being
targeted.
>
> From the logical design a database schema is produced (physical design).
>
> Am i going wrong somewhere?
>
>
>
>
Received on Sun Nov 03 2002 - 19:10:01 CST