Re: Date is Incomplete - database application software and database theory

"John Jacob" <jingleheimerschmitt_at_hotmail.com> wrote in message news:72f08f6c.0405231350.51a538a2_at_posting.google.com...
> > > On the contrary, it provides an excellent basis for this management,
> > > as evidenced by your own solution to this perceived problem.
> >
> > The excellence of basis is as a result of running the applications
> > (which is not addressed by the Relational Model, Date, et al)
> > as stored procedures within the RDBMS environment.
>
> You continue to claim that Date has not addressed this problem. On
> the contrary he addresses it very well in What Not How.

The same author reserves one diagram and perhaps a few sentences in regard to the applications layer in his "Intro to Database Systems". I will keep a look out for this other work, but his major work is flawed - for the reasons given.

> > This excellence is due to the fact that all application code
> > external to the RDBMS involves redundancies of redefinition
> > of the database schema (already defined by the model).
> >
> > My solution addresses the interface to the applications
> > environment of which the model says nothing.
>
> I agree that you have recognized a problem with current application
> development, but you have not correctly identified the cause of the
> problem.

This remains to be seen.

> The relational model cannot possibly have anything to say
> about the *implementation* of applications based on it.

This statement is the common response by RM supporters. Date says the same thing. You're just quoting chapter and verse.

The relational model of the data has alot to say about ongoing change management of database systems after implementation because the data schema in most organisations change in the course of time.

Every change management exercise can be viewed as a mini- -implementation exercise and in this, even by your own definitions, the RM is a useful tool to map and consider these changes to the data existent (implemented) structure.

> You are
> blaming the relational model for the inadequacies of current DBMSs.

See above.

> > I believe that we need a relational model and theory of
> > organizational intelligence, and I have summarised the
> > reasons why (process is more than just data!) in the above
> > article.
>
> We already have it. It's called the relational model.

It is a model of data. What needs to be modelled is something slightly in excess of this "data alone", and includes processes and applications, with respect to the organization.

> So I'll ask
> you once again, have you read What Not How?

Date cannot get things right in his major publication concerning an "introduction to database systems". What should I be looking for in these other works?

If you cannot tell the difference between data and intelligence, or perceive that they need to modelled in a different manner, then how am I to assess your assessment?

Pete Brown
Falls Creek
Oz Received on Mon May 24 2004 - 03:49:48 CEST