Re: Does Codd's view of a relational database differ from that ofDate&Darwin?[M.Gittens]

In article <qLgye.137278$Nn7.7012386_at_phobos.telenet-ops.be>, jan.hidders_at_REMOVETHIS.pandora.be says...
> > A value has one or more representations.
>
> Yes, but note that that was true in *my* definition of value. In ORM
> terminology the notions of value and representation are usually
> considered synonymous.
>
> > A lexical object is a representation, not a value.
>
> Again, under my definition of value. In ORM's vocabulary this would not
> be true.
>
> > A lexical object type is a set of values.
>
> Yep, although it would probably have been less confusing if I would have
> said it was a set of representations. My apologies for that.

Well, now I am more confused than ever. For some reason you did not comment on my last statement:

> > Thus, a lexical object is not a member of a lexical object type.

---which I presented as the contradictory (or at the very least counterintuitive)  conclusion of the other statements, and now you say that a LOT is a set of representations after all? I give up.

-- 
Jon