Re: The Practical Benefits of the Relational Model

        I know you just meant to be provocative, but here there go clarifications for the naïve readers.

Lauri Pietarinen wrote:
>
> Nobody has the exclusive right to whatever
> "relational database management system" means.

        No, but those who do not abide by 1st order predicate logic and n-ary relations according to set theory have no rights whatsoever on the RM.

        So I think it could be said that the set of fundamentally sane people who abide by the theory have rights to the exclusion of those who tries to defy Logic and Math.

> Maybe Oracle, DB2 and SQLServer are decent
> RDBMS's and there is no need to strive for
> something more.

        Huh? Are you stoned? Nothing SQL is neither good, nor relational, nor flexible or powerful enought.

> Maybe Codd is right in his V2-book.

        Maybe, except that it is such a pain to read!

> Maybe Stonebraker got it right...

        Hm. OODBMS? No, thanks.

> Maybe implementing the full SQL1999 standard
> is the holy grail.

        Ouch! A corrupt pseudo-relational data language does not becomes better by incorporating the very bane of SQL, namely pointers.

> Who knows - maybe the OO-guys were right after all...

        If they could produce what is a data model and what are the elements of their data model in particular.

> It is just that to my knowledge
> Dataphor is the only one at the moment.

        Well, actually one could compile an old version of Postgres or Ingres to use QUEL, or bother IBM for the source code to BS-12. But still Dataphor is the only *modern, readily available* RM implementation.

-- 
  _
/ \ Leandro Guimarães Faria Corsetti Dutra        +41 (21) 216 15 93
\ / http://homepage.mac.com./leandrod/        fax +41 (21) 216 19 04
  X  http://tutoriald.sourceforge.net./      Orange Communications CH
/ \ Campanha fita ASCII, contra correio HTML      +41 (21) 644 23 01