RE: Single record access really required (was RE: Views and Indexes)

["Joe Pluta" <joepluta@xxxxxxxxxxxxxxxxx>]

> From: CWilt@xxxxxxxxxxxx
> 
> I'm confident SQL will prove to be better than native when dealing
with
> set
> processing.  The interesting question to answer is how big of a set
does
> it
> take to make a difference?  The other interesting question is the
effect
> of
> added complexity.  If instead of doing 4 chains to the master files
(or
> the
> SQL equivalent) we have to do 5 or 6 or 8.

Yup, this is the issue alright.  And there's no easy way to test it
except to create a suite of test cases.  That's where the website will
come in.  Don't worry about getting done right away.  I'm pretty booked
myself.  I'm going to try to get a basic framework for tests in place
this weekend, and we can move forward from there.

Joe