[AccessD] question on normalization

Okay, I just wonder what happened to the original intent of Zip codes - 
wasn't that to uniquely identify delivery locations?  That certainly is 
what I remember - unique codes to make it possible for automatic and 
correct mail sorting by machines.  Hmmmmmmmmmmmm <sits scratching her head>
Tina

Charlotte Foust wrote:

>Unfortunately, while zip codes are supposed to be unique, they may be
>shared among several small towns or a town may have multiple zip codes.
>Some buildings even have unique zip codes, but generally zip codes
>belong to post offices.  I would say they were not a good candidate for
>a primary key in dealing with addresses.  I've worked with postal
>databases and seen some of the "duplicate" zips, where several small
>towns share a post office and a zip code.  A further complication is the
>+four extension.  
>
>Charlotte Foust
>
>-----Original Message-----
>From: John Clark [mailto:John.Clark at niagaracounty.com] 
>Sent: Friday, March 12, 2004 7:03 AM
>To: accessd at databaseadvisors.com
>Subject: RE: [AccessD] question on normalization
>
>
>But, in my case, it would be OK then? I will not have an issue with them
>being unique. 
> 
>JW Clark
>
>  
>
>>>>ssharkins at bellsouth.net 3/12/2004 9:55:22 AM >>>
>>>>        
>>>>
>
>
>As a sidebar--sort of--why can't the zip itself be a primary key?
>
>========We're told that ZIP codes aren't unique -- most are, but there
>those few... I've not encountered it myself, but that's what we're told.
>;) 
>
>Susan H. 
>
>  
>