Re: Data Normalization
- From: R Tanner <tanner.robin@xxxxxxxxx>
- Date: Thu, 16 Oct 2008 17:35:11 -0700 (PDT)
On Oct 16, 4:57 pm, John W. Vinson
<jvinson@xxxxxxxxxxxxxxxxxxxxxxxxxx> wrote:
On Thu, 16 Oct 2008 14:51:30 -0700 (PDT), R Tanner <tanner.ro...@xxxxxxxxx>
wrote:
I understand what the purpose and value of data normalization is, but
I just have one thing I am confused about. If you split up a table,
then it will create two tables. The child table will have a lookup
column to the parent table, which will have whatever value is
applicable that relates that given record to it's parent. What I
don't understand is how that is more efficient. In fact it seems to
duplicate the data if anything...
It duplicates *one field*. Your related table may have many fields!
Just for example, consider a People table with a PersonID, related one to many
to an Address table. Each person may have several addresses; each record of
the address table would have fields like AddressNo, Direction, Street, Suffix
(i.e. St., Ave., Blvd.), Postcode, City, State and Country. These fields don't
need to occur in the parent table and need not be duplicated.
--
John W. Vinson [MVP]
Also in reply to your post about the primary and foreign keys - -
When I use the table analyzer and I split a table, the lookup column
to the parent table will have the actual values that are in the parent
table. Should this be the case, or would it not take up less memory
in the database to have the lookup column look up the primary key in
the parent table, which would be an autonumber...
.
- Follow-Ups:
- Re: Data Normalization
- From: John W . Vinson
- Re: Data Normalization
- From: John Spencer
- Re: Data Normalization
- References:
- Data Normalization
- From: R Tanner
- Re: Data Normalization
- From: John W . Vinson
- Data Normalization
- Prev by Date: Re: Data Normalization
- Next by Date: Open a form to a specific record from a query
- Previous by thread: Re: Data Normalization
- Next by thread: Re: Data Normalization
- Index(es):
Relevant Pages
|