Re: What do Oracle professionals think of Fabian Pascal?

fabian pascal wrote:
> Paul,
>
> 1. I say nothing about "vendor extensions" as such. My criticism is of
> any violation or dismissal of relational principles which is completely
> different. If an extension does that, then it's a bad idea.

Why? Because you are a purist or for some reason that supports the business use of the tool?

> 2. Where you are mistaken is that a vast majority of professionals are
> not competent as far as data fundamentals are concerned. They may know
> programming or Oracle, but they have little clue of database and
> relational fundamentals.

No argument here. But the ignorance, or lack thereof, is irrelevant to whether a vendor adds an extension to their product (unless it is a help system).

> Therefore, they cannot assess products or
> features properly and confuse what the product does with the correct
> way of doing it. It is one thing to be aware of the problems, but use
> them because you have no choice (and know what consequences to expect),
> and another to be oblivious to the problems.

Again ... true but irrelevant.

> So be very careful with the replies you receive from Oracle
> professionals. Morgan is an excellent example. For my comments on his
> competence see:
>
> http://www.dbdebunk.com/page/page/1770593.htm
> http://www.dbdebunk.com/page/page/1477865.htm
>
> The relational model is application of science--logic and math--to
> database management. The notion that one is "fanatic" about that is
> tantamount to claiming that civil engineering is fanatical about the
> laws of physics. Such a claim reveals problems more serious than
> ignorance of data fundamentals, and THAT is what deserves disdain.

Your self-aggrandizement goes beyond even that I am capable of and I have never been accused of being shy.

The rules of physics were not created by physicists ... they were created by (take your pick) the underlying principles by which the universe exists or god. Last time I checked relational algebra was man made and both Codd and Date were human.

A database is a tool. When the use of the tool provides value to a business or organization it is a plus: When it does not it is a negative. That the person wielding the hammer is more or less competent is a diversion.

That an object-extension does not fit cleanly within some theoretical framework is irrelevant to anyone except a purist. If use of that object-extension allows the more accurate modeling of a business requirement and interface with an OO front-end it is a plus. If it does not then it should die on the drawing board.

There are plenty of places where, in the real world, relational theory breaks down when it comes to application. Not once have I seen fifth normal form in a production application: Have you?

-- 
Daniel A. Morgan
http://www.psoug.org
damorgan_at_x.washington.edu
(replace x with u to respond)