Re: Demo: Modelling Cost of Travel Paths Between Towns

From: Hugo Kornelis <hugo_at_pe_NO_rFact.in_SPAM_fo>
Date: Mon, 15 Nov 2004 22:32:48 +0100
Message-ID: <rt7ip01f16m33885jkah7juf48fkmfukkf_at_4ax.com>


On 15 Nov 2004 12:19:40 -0800, Neo wrote:

>> So if state boundaries are changed and the city of Unreal, PA will
>> henceforth be part of Texas, changing the row for Unreal, PA to Unreal, TX
>> would corrupt the db?
>
>Reading thru the earlier posts, it seems I have lost

Yes, Neo - we all know.

(Sorry - couldn't resist the temptation)

>What I failed to catch here was:
>1) Alan was not able to rename town c to a in the original RM solution
>presented by Celko.
>2) The original RM solution requires changes to schema to accomodate
>two towns with the same name.
>
>The above problems resulted because Celko's original one-table schema
>represented cities redundantly. It should have had a separate table
>for cities. In XDb2's solutions, the towns are already normalized.

No, the above "problems" resulted because you didn't provide Joe the full specifications. He gave a good solution for the problem you described, and then you added to the specifications.

Hmmmm, why does this suddenly sound all too familiar?

Best, Hugo

-- 

(Remove _NO_ and _SPAM_ to get my e-mail address)
Received on Mon Nov 15 2004 - 22:32:48 CET

Original text of this message