Re: Something new for the New Year (2008).

On 3 jan, 21:47, Rob <rmpsf..._at_gmail.com> wrote:
> On Jan 3, 11:18 am, TroyK <cs_tr..._at_juno.com> wrote:> On Jan 1, 6:01 pm, JOG <j...@cs.nott.ac.uk> wrote:
>
> <<snip>>
>
> > Seconded....
>
> I don't normally respond to ad hominem attacks, but let's clear the
> record here:
>
> > From the paper:
> > "There doesn't seem to be any commonly accepted name for the Figure 1
> > representation. "
>
> > Sure there is. It's called a subset requirement constraint.
>
> I checked the 2005, 2006 and 2007 SIGMOD and PODS Proceedings.

In that case you'd better look for the term "inclusion dependency". But that is just the name of a general type of database constraint, of which foreign keys are usually assumed to be the special case where the destination is a candidate and/or a primary key. What you seem to be talking about here is the use of a foreign key to represent a binary relationship. I'm not sure why you think there should be another name for this "construct" then simply "foreign key" is not clear to me.

• Jan Hidders