[AccessD] On DB Bloat, Bad DB Design, and various

Stuart McLachlan stuart at lexacorp.com.pg
Sat May 22 18:30:17 CDT 2004


On 22 May 2004 at 13:20, Arthur Fuller wrote:

> I respectfully disagree, at least regarding the design of the database.
> As to the programming, I'm agnostic.
> 
I 'd call myself an agnostic as far as database design is concerend, 
but I'm strongly of the "programming as an art" school.

> While at the same time admitting artistic pretensions (published poems,
> screenplays sold and made, short fiction published, tabla-player and
> classical guitar wannabe), I don't see the connection. I used to
> subscribe to your opinion but I do no longer. Long ago, I used to think
> programming was an art. Now I hold it in not much greater regard than
> auto-mechanics. 

"mechanics" is a key word.  There's nothing mechnical about 
programming.  It's more like writing a poetry.  You have a set of 
building tools and rules (language, meter, rhyme in poetry).

With the same tools, you end up with either Robert Burns and Willima 
Topaz "The Great" McGonagall. 

That's why you will often get ten different solutions  when you ask a 
programming question here. Some of these solutions can be very 
McGonagallish (convoluted, long winded, inefficient) but still 
workable.  Then someone will come up a really elegant solution ( 
often as the result of refining some else's initial idea).  

> As I see it now, db design is strictly science: for any
> given database there is one correct design, a collection of
> close-to-correct designs, and a larger collection of incorrect designs.
> These stages correspond to the correct scientific theory, the
> almost-correct theories, and the flat-earth society.
> 
This is where the agnosticism creeps in.
I think that at times, there are valid reasons for not full 
normalising your data, the art comes in knowing where you can safely 
bend the rules.


 
-- 
Lexacorp Ltd
http://www.lexacorp.com.pg
Information Technology Consultancy, Software Development,System 
Support.






More information about the AccessD mailing list