[AccessD] One-to-One relationships

Arthur Fuller fuller.artful at gmail.com
Fri Jul 20 10:21:31 CDT 2007


Wow! I used the term "stripper" because I deferred to the international
character of this list, but you're entirely correct! Applause to you from
here, Charlotte. You are sooo correct.

On 7/20/07, Charlotte Foust <cfoust at infostatsystems.com> wrote:
>
> Actually, she was a fan dancer, which is different from a stripper,
> since they start out without much on and wave the fans around to expose
> bits at a time.  She was quite famous in her time.
>
> Charlotte Foust
>
> -----Original Message-----
> From: accessd-bounces at databaseadvisors.com
> [mailto:accessd-bounces at databaseadvisors.com] On Behalf Of Arthur Fuller
> Sent: Friday, July 20, 2007 6:24 AM
> To: Access Developers discussion and problem solving
> Subject: Re: [AccessD] One-to-One relationships
>
> I would respectfully suggest that you're overlooking something in your
> analysis, Susan, but to observe it you need millions of rows in the
> given table. But suppose that you have a table called Customers, which
> as previously suggested in this thread might include both corporations
> and persons. The first rule of database development is performance,
> above all other considerations. Therefore one ought to identify the
> columns of immediate interest (CustomerID, CustomerName, etc.) and store
> those in a single table, pushing all the other attributes to one or more
> related tables with a 1:1 relationship. This way, I can search a small
> table with multiple indexes very quickly, and not bother with fetching
> the rest of the data until you explicitly request it, at which point it
> would be a lightning-quick sproc that receives a CustomerID and sends
> back the rest of the data. If you really want to push the performance
> button, then you won't return a rowset either. Instead you'll declare as
> many parameters as you have columns of interest, and declare them all
> Output parameters. When you want exactly one record, that's the quickest
> method.
>
> I hope I didn't obscure the point here. The point is what I call the
> Sally Rand principle. (You might have to be older than even I to
> understand the reference -- she was a famous stripper, back when
> stripping meant that you still retained most of your clothes.) Her point
> was, show them as little as possible to still maintain their interest.
> That's my motto in terms of database design. Never open an entire table.
> Show them only enough to pique their curiosity, as it were.
>
> On 7/20/07, Susan Harkins <ssharkins at setel.com> wrote:
> >
> > Yes, if it has a purpose. A one-to-one relationship almost always
> > flows from need rather from the data itself. If you need to force a
> > one-to-one, I'd say do it. However, if there's no business rule
> > saying, "there can be only one..." it might be unnecessary, even if
> > the data is presenting that picture right now. Listen to the data.
> >
> > Susan H.
> >
> > Is there any purpose/advantage in creating a one-to-one relationship
> > in a database (e.g., CustomerId and CustomerName in one table and all
> > the other customer data (e.g., sex, address,  phone, etc) in another
> table?
> >
> >
> > --
> > AccessD mailing list
> > AccessD at databaseadvisors.com
> > http://databaseadvisors.com/mailman/listinfo/accessd
> > Website: http://www.databaseadvisors.com
> >
> --
> AccessD mailing list
> AccessD at databaseadvisors.com
> http://databaseadvisors.com/mailman/listinfo/accessd
> Website: http://www.databaseadvisors.com
>
> --
> AccessD mailing list
> AccessD at databaseadvisors.com
> http://databaseadvisors.com/mailman/listinfo/accessd
> Website: http://www.databaseadvisors.com
>



More information about the AccessD mailing list