Re: So let me get this right: (Was: NFNF vs 1NF ...)

"A paradox about self-referential sets".

Sure - these are the point. And the simplest way to avoid such difficulties is to say, "Hey! You can't reason about domains in the same way you reason about attributes."

"Armstrong's Axioms"

Citation? Where did they say, "Hey! Armstrong's 1974 paper presents a useful way of generating a normalized relation scheme from a set of FDs."

(I'd also argue, though this is kind of a weasel, that all Armstrong & Fagin were doing was exploring the framework that Codd outlined, rather than coming up with some kind of fundamental extension or elaboration (or interpretation). See the Lorentzos note below.) Received on Thu Feb 10 2005 - 20:11:02 CET