Re: Informal Survey #1 -- joins on foreign keys
Date: Mon, 3 Oct 2011 09:56:34 -0700 (PDT)
Message-ID: <6bd05e24-802d-43f4-ae0e-df5bd8bec347_at_y18g2000yqy.googlegroups.com>
On Oct 2, 10:58 am, Rob <rmpsf..._at_gmail.com> wrote:
...
> For example, those people who "will want to join two or more of
> those tables to compare Shipment_Value to Invoice_Amount or
> Receivable_Amount" will likely do so THROUGH the Customer
> table so as to provide a Customer attribute that identifies the rows
> of the
> result. How many would do it without the Customer table?
> ...
Plenty. For one, the preparation of financial statements doesn't require Customer information. That is a simple fact, not a gut reaction.
...
> What I'm really trying to get at is whether a foreignkey-foreignkey
> join makes sense algebraically, ...
It makes sense in the same sense that 2 + 2 makes sense in a numerical algebra, whereas 2 - / 2 doesn't make sense. Looks like you are changing questions, the latest seems quite insensible.
> ... and if so, do operations in higher-level
> abstractions (like E-R, facts) translate to them?
What the heck are the operations of E-R? Now it's even harder to make sense of your theme, can't tell what you mean by 'translate' - mechanical, logical, mystical or what? Looks like you don't even know what your question is, which will make it impossible to assess answers. Received on Mon Oct 03 2011 - 18:56:34 CEST