Re: theory and practice: ying and yang

"Alfredo Novoa" <alfredo_novoa_at_hotmail.com> wrote in message news:3n09919j97bfe936nrqm8embca3ch4fg0o_at_4ax.com...
> On Wed, 25 May 2005 12:28:52 GMT, "mountain man"
> <hobbit_at_southern_seaweed.com.op> wrote:
>
>>> Of course, but this is rather contradictorial with many of your
>>> previous posts.
>>
>>Really? Is it not possible that a person can consider
>>multiple points of view, or must one consider just one
>>POV?
>
> A person can consider multiple points of view, but he should not state
> contradictorial points of view.

One should be prepared for anything in this world.

>>> Where do you want to go with this?
>>
>>Consider that:
>>
>>1) Current (RM) theory has not substantially changed in 30 years
>
> Agreed
>
>>2) The practice-environment of DBMS theory has substantially
>> changed in this time, in fact one might be entitled to claim it
>> has changed by a quantum leap
>
> Because there was a gap between the creation of the Relational Model
> and the commercialization of the SQL DBMS.
>
> The practice of DBMS has not substatially changed in 20 years.

Practice has substantially changed because the market place in these last 20 years has seen the emergence of what IBM, Oracle and Microsoft call the RDBMS -- what you call the SQL DBMS.
>> --- in the use of RDBMS
>> software (ie: built in accordance to the principles of the RM).
>
> SQL DBMS software was not built in accordance to the principles of the
> RM.

Clearly what IBM, Oracle and Microsoft term RDBMS software is DBMS software, which has some right to be termed relational.

I have worked with older style DBMS, IBM and Wang in the 80's. They were not relational, but one could make them closer to being relational. This new RDBMS software is certainly partially relational.

>>3) Date openly admits that current (RM) theory is not understood
>> by database professionals.
>>
>>You tell me what the implications are.
>
> The industry is not taking full advantage of the Relational Model.

Vendors IBM, Oracle and Microsoft are taking a large advantage of the RM.

Their systems are far more "relational" than the proprietory DBMS of the 1970's, 1980's, 1990's. Things may not yet be "all the way there" but you must admit they are not "off the track completely".

Other implications include the possibility that Date has made the subject of database systems theory unnecessarily esoteric and pedantic, at the expense of usefulness.

Pete brown
Falls Creek
Oz
www.mountainman.com.au Received on Thu May 26 2005 - 10:48:41 CEST