[AccessD] Primary Key Best Practices

Michael R Mattys mmattys at rochester.rr.com
Thu Jul 26 10:17:23 CDT 2007


Is this OT yet?

These could possibly be used for:

names of pushpins on a map -
(all have latitude and longitude)

filenames, including Null -
(all have a timestamp based upon the vibration of an atom or crystal)
[Duplicate timestamps occur after several billion years, but who cares?]

Problem:
I wouldn't have a problem as long as records are kept of where or when.
Otherwise, I can't reliably locate the record by it's name in this column.

We are debating this because the records of where or when are not
always kept or communicated, as far as we know.

Michael R. Mattys
MapPoint & Access Dev
www.mattysconsulting.com

----- Original Message ----- 
From: "Bruce Bruen" <bbruen at unwired.com.au>
To: "Access Developers discussion and problem solving" 
<accessd at databaseadvisors.com>
Sent: Thursday, July 26, 2007 10:43 AM
Subject: Re: [AccessD] Primary Key Best Practices


> To get banal, lets consider a dataset that consists of one and only one
> column "NumberOfItemsOrdered". It has, as of this moments snapshot, 234867
> rows:
> 1
> 4
> 87
> 8734
> (null)
> 1
> 34
> six
> 224
> ...
> 1
>
> Of what possible use it this? Where are the problems? Why are we debating
> this?
>
> -- 
> regards
>
> Bruce




More information about the AccessD mailing list