[AccessD] OT: The Great Primary Debate

Scott Marcus marcus at tsstech.com
Tue Jun 15 07:51:29 CDT 2004


I know what you are saying. I even know what a natural key is according to the theory. I tried to say this before, maybe you missed it, a natural key in English might not even be close to a natural key in French or Russian or etc. (so how can it be natural?)

I'm just questioning the whole thought process of 'Natural Key'. To me, the term is an oxymoron because there is nothing natural about the key. I do understand relational theory. Just bare in mind, it is theory.

Scott Marcus
TSS Technologies, Inc.
marcus at tsstech.com
(513) 772-7000

 -----Original Message-----
From: 	accessd-bounces at databaseadvisors.com [mailto:accessd-bounces at databaseadvisors.com]  On Behalf Of Jim Dettman
Sent:	Tuesday, June 15, 2004 8:35 AM
To:	Access Developers discussion and problem solving
Subject:	RE: [AccessD] OT: The Great Primary Debate

Scott,

<< Maybe meaningless to you, but to the system, it is as meaningful as any
other attribute. At some point, the attributes which you call natural, like
house number, were assigned by someone. Hopefully, that someone used a
method to assign a value to that attribute. By definition, what part of that
process is natural(say for instance house number)?>>

  You said "system".  When talking about keys, keep in mind that there is a
significant difference between what a "key" means in relational theory vs
computing systems.

  Relational theory is a logical approach to the organization of
information.  There is no physical component attached to it.  It can be
applied to any store of information; computing systems, a chalk board, pen &
paper, you and I keeping lists in our heads, or what ever.

  The whole point of this is that the word "key" has two very different
meanings between relational theory and computing systems.  They are not the
same thing.  One needs to be aware of this when discussing such things as
the article mentioned, which is what started all this.

  With that, this will be my last post for a bit as work wise things have
really heated up in the past day or two.

Jim
jimdettman at earthlink.net

-----Original Message-----
From: accessd-bounces at databaseadvisors.com
[mailto:accessd-bounces at databaseadvisors.com]On Behalf Of Scott Marcus
Sent: Tuesday, June 15, 2004 7:48 AM
To: Access Developers discussion and problem solving
Subject: RE: [AccessD] OT: The Great Primary Debate


<<An autonumber cannot be a natural key by definition since it is
meaningless.

It is not meaningless. It isn't even arbitrary in its assignment(there is a
method to which it is assigned) after which, it has fundamental meaning.
Maybe meaningless to you, but to the system, it is as meaningful as any
other attribute. At some point, the attributes which you call natural, like
house number, were assigned by someone. Hopefully, that someone used a
method to assign a value to that attribute. By definition, what part of that
process is natural(say for instance house number)?

I know it's splitting hairs, but to a balding man(which I'm not) every hair
is important.

Scott Marcus
TSS Technologies, Inc.
marcus at tsstech.com
(513) 772-7000

 -----Original Message-----
From: 	accessd-bounces at databaseadvisors.com
[mailto:accessd-bounces at databaseadvisors.com]  On Behalf Of Jim Dettman
Sent:	Monday, June 14, 2004 3:59 PM
To:	Access Developers discussion and problem solving
Subject:	RE: [AccessD] OT: The Great Primary Debate

Scott,

<<What exactly is the point?>>

  That there is a difference between a surrogate key (something like an
autonumber) and a natural key.  An autonumber cannot be a natural key by
definition since it is meaningless.  Look back at the response I gave to
Lambert with the house number example.

<<We are getting close to what someone else said(I think Gustav). The only
natural key for an object, is the object.>>

  It was Lambert and it was in regards to talking about the attributes of a
person.

Jim
(315) 699-3443
jimdettman at earthlink.net

-----Original Message-----
From: accessd-bounces at databaseadvisors.com
[mailto:accessd-bounces at databaseadvisors.com]On Behalf Of Scott Marcus
Sent: Monday, June 14, 2004 3:37 PM
To: Access Developers discussion and problem solving
Subject: RE: [AccessD] OT: The Great Primary Debate


<< Nope sorry, but your missing the point.

What exactly is the point?

We are getting close to what someone else said(I think Gustav). The only
natural key for an object, is the object.

Scott Marcus
TSS Technologies, Inc.
marcus at tsstech.com
(513) 772-7000

 -----Original Message-----
From: 	accessd-bounces at databaseadvisors.com
[mailto:accessd-bounces at databaseadvisors.com]  On Behalf Of Jim Dettman
Sent:	Monday, June 14, 2004 3:32 PM
To:	Access Developers discussion and problem solving
Subject:	RE: [AccessD] OT: The Great Primary Debate

Francis,

 <<Perversely, I would state the situation as the inverse of your
statement, as long as you don't change the autonumber, you can change
any attribute without changing the instance.>>

 Nope sorry, but your missing the point.

Jim
(315) 699-3443
jimdettman at earthlink.net

-----Original Message-----
From: accessd-bounces at databaseadvisors.com
[mailto:accessd-bounces at databaseadvisors.com]On Behalf Of Francis Harvey
Sent: Monday, June 14, 2004 1:23 PM
To: 'Access Developers discussion and problem solving'
Subject: RE: [AccessD] OT: The Great Primary Debate


Jim,

Perversely, I would state the situation as the inverse of your
statement, as long as you don't change the autonumber, you can change
any attribute without changing the instance. By using this approach,
you provide a solution to fix mistakes in the attributes that make up
the so-called "natural" keys due to miskeying, incomplete information,
or a change in value.

In fact, if you used "natural" keys as your primary key, in order to
fix such mistakes you would have to allow for the very same thing that
you are claiming prevents autonumbers from being natural keys,
changing an attribute without changing the instance.

Francis R Harvey III
WB 303, (301)294-3952
harveyf1 at westat.com


> -----Original Message-----
> From: accessd-bounces at databaseadvisors.com
> [mailto:accessd-bounces at databaseadvisors.com] On Behalf Of Jim Dettman
> Sent: Monday, June 14, 2004 8:33 AM
> To: Access Developers discussion and problem solving
> Subject: RE: [AccessD] OT: The Great Primary Debate
>
>
> Scott,
>
<snip>
>
>   No because the serial number would be associated with the
> instance, so it
> becomes an attribute even though it was assigned.  An
> autonumber is not.  I
> can change an autonumber at any time with no affect at all.
> Looking at any
> given row, if I change the autonumber, nothing happens.  If I
> changed the
> serial number, I'd no longer be referring to the same instance.  One
> meaningless, the other derived from the attributes of what
> I'm referring to.
>
> Jim Dettman
> (315) 699-3443
> jimdettman at earthlink.net
<snip>
--
_______________________________________________
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


--
_______________________________________________
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