Re: foundations of relational theory?

Marshall:

A point to consider is that the goal of most businesses, with respect to application development, is to use some sort of development tool that allows the businessman to write the application. Why? Because the businessman knows his business far better than anyone else.

Some dbms products are far more conducive to this kind of development than others. Give a businessman MS-Access and you get a far different application than if you give them VB and SQL Server. With MS-Access you'll probably get a product that does what it's supposed to do while the nothing at all gets developed with VB & SQL Server. Even .NET is trying to give businessmen the tools to create applications.

Most business environments simply want the database to serve the needs of the business. There is nothing special about this desire. I'd not be interested in a data structure that works better when developed for a "law firm" AND a "warehouse" than with just a "law firm", if I'm working at a law firm. From a "non-business" developers perspective this view seems rather quaint, but from a businessman it should be an essential view.

[stuff snipped]

> > In reality any good application
> > developer can predict the nature of most of the queries that will be
> > asked of the database within a specific business context.
>
> I'm sceptical of this claim.

It is far easier to "predict" the nature of dbms requirements if designed and developed by the business people using the application than if a the application were designed and built by "technical" developers. But that is the nature of the way some dbms products are used.

> > That drasically limits the permutations, and allows the developer to
create
> > only those structures necessary to support the application. When we
> > have needs to extend out of the box it's no problem to reformat data
> > into new tables, or simply create extended definitions that point to
> > existing data.
>
> Wouldn't you agree that it would be better to have a system
> whereby no changes to existing applications are necessary to
> accomodate new applications?

From strictly a business perspective, I would not agree. My competitors are going to change tomorrow and try to figure out a way to get more business. I need to be prepared to respond (or do what they do). Under this dynamic economic model, I think it is better to have "...existing applications change to accommodate new applications."

This is, or course, IMHO. :-)

Bill Received on Sun Oct 26 2003 - 21:39:21 CET