Re: does a table always need a PK?

"Marshall Spight" <mspight_at_dnai.com> wrote in message news:aUw4b.317792$YN5.217833_at_sccrnsc01...
> "Paul G. Brown" <paul_geoffrey_brown_at_yahoo.com> wrote in message
news:57da7b56.0308311128.3b6d4518_at_posting.google.com...
> > Put it another way: as much as folk would love to be able to say
things
> > like "I don't care about implementation details: we're talking about
the
> > logical model here.", it simply doesn't cut much cheese. In the end
you've
> > got to have a working system or no one (outside a small circle of
friends)
> > cares because what you have is impractical.
>
> I agree that for any given model or design being discussed, it must be
> *implementable* for there to be any point in discussing it. But I
> think you go too far if you're asserting that the implementation
> details need to be all worked out before, or at the same time as,
> the model. Logical models are an important aspect of system design,
> and while it is important to be able to discuss how the parts of
> a system design interact, it is also important to be able to talk
> about those parts in isolation. This is nowhere more true than
> when discussing a logical model, because that is the part the
> human will interact with.

Marshall, this idiot has not established that the relational data model is not implementable. It is more easily implementable than the SQL bag model. Have you seen how inelegant published theories become when they have to account for NULL and duplicates? Yuck!

He ignored the RM when he did his implementation. That doesn't make it unimplementable. Received on Mon Sep 01 2003 - 03:19:09 CEST