Re: Objects and Relations
Date: 14 Feb 2007 08:24:44 -0800
On Feb 14, 5:22 pm, "Roy Hann" <specia..._at_processed.almost.meat>
> "Marshall" <marshall.spi..._at_gmail.com> wrote in message
> > On Feb 14, 7:14 am, "Roy Hann" <specia..._at_processed.almost.meat>
> > wrote:
> >> >> As an addendum, once implemented a surrogate key becomes a natural
> >> >> key. I find this fascinating - it seems somehow analagous to "Nature
> >> >> abhoring a vacuum".
> >> > I think it has more to do with the human drive to communicate.
> >> I struggle with that. If, for some reason, I choose to say (just) that,
> >> "I
> >> have a tin of cat food, and its name shall be 1345235", what have I
> >> communicated? No one in the real world could ever point it out and say,
> >> "Hey! There goes 1345235." And nor could I.
> > Let's think of customers and customer ids; it's a very similar case.
> Actually no. I deliberately wrote: 'I choose to say (just) that, "I have a
> tin of cat food, and its name shall be 1345235"', with emphasis on "just".
> Any attempt to distinguish individual cans is probably spurious. Yet people
> do often try to do such things in databases. (Wrongly IMO.)
> > What does this communicate? It communicates customer
> > identity.
> No question, although in the real world it communicates only through
> familiarity over time, and only because customers are not in every way
> The point I am groping towards (I think) is that surrogates are exactly
> that. They stand in place of something else you could have used albeit
> perhaps less conveniently. But if there is nothing else, what does the
> surrogate stand in place of?
In order words surrogate keys are *practical*? Received on Wed Feb 14 2007 - 17:24:44 CET