Re: Hierarchical Relationship

joe_celko_at_my-deja.com wrote:

>Why not follow the Metadata committee standards for naming data
>elememts?

could you give me a good pointer?

>> implementing aggregate functions on the clientside is a rather suboptimal idea.
 

>Agreed, but an aggregate is a set operation. I have a bunch of tricks
>for writing the "fancy ones" with the built-in aggregate operators and
>some simple library function calls.

why should using fancy tricks be any better than using stored functions? i have to admit that some of your sql-tricks i saw so far were really astonishing - but this stuff completely contradicts the goal of maintainability as it's to difficult to understand for the average programmer in a reasonable amount of time. i wrote c programs a couple of years - i know these joys pretty well :->.

>I hope that nobody is still using rule-based optimizers in the 21st
>Century!!

i'm pretty sure oracle apps do exactly this. and to be honest, if you are able to estimate the distribution of data and the underlaying goals of your query well, the queries will always result in reasonable - fixed - plans. tuning a cost-based-optimizer with "hints" and stuff isn't better a dime :-) actually it's even worse as the bad plans show up at the customers site first. otherwise one would have tuned them..

>At least nobody is still making sldie rulers ...

you mean they're no longer implemented? they are.

ciao, jan

http://www.rsp.de/

rs&p-Dossier: Software zur Erstellung technischer Dokumentationen

              und Schriftgutes in Verwaltung und Industrie. Received on Sun Mar 19 2000 - 00:00:00 CET