Re: Relational Databases and Their Guts

> SQL does not define RDBMS. When you insist on idiosynchratic definitions of
> words and terms, you refuse to communicate. I suggest you open a dictionary
> from time to time. If you find books tiresome, try
> http://www.dictionary.com/

I understand your point completely, and it is unfortunately you overlooked my carefully chose phrase: "popularly defined". Of course, SQL and Sybase, etc, are RDBMS's based upon Codd's rules and Date has completely disavowed it. However, when most people refer to an RDBMS, they assume that you mean Oracle or SQL Server. It is extremely unlikely that your protests against products that have billions of dollars in R&D, mindshare, and marketing, will do much good. I think that instead of stubbornly fighting an unwinnable fight to mold public opinion (which you've already said that theorists don't have to do), then, perhaps instead call your "relational" thing something else. So that in the very least you can create a distinction in people's minds that there is actually something better or could be better (if implimented) than SQL Server and Oracle and MYSQL!

You don't have the right to decide the meanings of words if you say your job is not to communicate!

If the job of a theorist is not to communicate, then, isn't it right for the marketers, whose job it is to communicate, to capriciously rename words invented by the theorist? Should not the theorist reasonably expect that if mass communication is not in his field that the public meaning of a phrase is not under his control, and that he should be prepared to adjust his public language to suit the needs of the marketspace?

Possible acronyms for "relational database" in, your, true sense of the phrase:

Set Theory Based Systems (or STS).

Date Codd Derived (or DCD).

>
>
> > Relational Database: A theoretical construct.
>
> A relational database is a very real logical construct. It is theoretical in
> the sense of relating to or being based on theory, but it is not theoretical
> in the sense of being speculative or hypothetical or in the sense of being
> not practical.

I agree. My use of the word "theoretical" was to make the distinction that it defines how something should be made, as opposed to being an implementation itself. Received on Fri Jun 20 2003 - 06:47:27 CEST