Re: foundations of relational theory? - some references for the truly starving

"SixFtWabbit" <dragoninbabylon_at_cox.net> wrote in message news:9uElb.66911$La.44306_at_fed1read02...
>
> Pick/MV/Whatever id NOT a relational database as defined
> scientifically/mathematically(all hard science IS Mathematics).

Neither is pick even a database management system. It is a primitive file processor.

> It has no
> published theory of action or any rational document (non-marketing)
> describing its foundation. A few behavioral texts exist. But there is no
> available theoretical treatment.
>
> However, it was devised by mathematicians and computer scientists in the
> pre-relational world to handle a complex, practical, real-world inventory
> control problem (Cheyenne helicopter manufacturing) and its development
was
> financed by the US Government as a contract to TRW.

Are you suggesting the US Government generally produces good things under contract to TRW? Doesn't that combination define waste and inefficiency? Weren't they Proxmire's favourite whipping boy? Or does my memory not serve me well?

> (see GIM - General
> Information Management - work titled GIRLS General Information Retrieval
> Language.) Since it resides in the Public Domain, all related material
> should be available.
>
> Be that as it may, what was created was an integrated data
base/operational
> environment which made it very easy for smart, non-technical businessmen
to
> create sophisticated applications to manage their BUSINESS, a highly
> unscientific world.

When you say "integrated data base/operational environment", if you mean "primitive file processor", then your description is accurate if wordy. If you mean anything else, then the wordy description is not even accurate.

> This goes far beyond accounting into manufacturing,
> MRP, automated dispatch, insurance policy management, medical billing and
> the like.

People have gone to heroic efforts to use primitive file processors in just about every field of endeavour.

> Most of these applications (not small either- many in with
> thousands of online users) were designed, coded and implemented by people
> who understood and worked in these businesses first, and learned how to
> "program" as a means to an end.

In other words, they were coded by people who were generally ignorant of the craft and who had absolutely no objective basis upon which to base their conclusions regarding quality. That fits with empirical evidence.

> The query language (originally marketed as ENGLISH) uses SQL structures
for

Huh? How could it use anything that had not even been invented yet?

> Relationships are based on proximity and virtual links- not relational
> algebra.

You apparently lack good diction with respect to database management. By relationship, do you mean "reference", "pointer" or "relation"?

> Hence, most developers in the Pick world are not Data Base designers
per-se.
> They are application designers. The application has always outweighed
data
> base theory. I am not judging whether this is good or bad

Then you are abdicating your responsibility. It is decidedly bad. It was obvious to data managers more than three decades ago that it was an extremely bad idea to try to manage data within applications. It just doesn't work well.

> I will close with an anecdote.

[anecdote of ignorami substituting emotion for reason snipped]

> Can I here and Amen from the Pickies? (oops- too religious.)

Actually, I would say it is exactly the right level of religiosity in reference to Pick in particular. Received on Thu Oct 23 2003 - 03:09:36 CEST