Re: Can we solve this -- NFNF and non-1NF at Loggerheads

Roy Hann wrote:
> 1NF does not "mean values are simple or indivisible". It just says
that for
> the purpose of the relational theory, the divisibility (or internal
> structure) of values of a given type is of no interest or use within
the
> theory. The theory does not make use nor reference to the internal
> structure of a value of any type. The theory therefore does not need
to
> define what atomic means. The term "atomic" is a just a narrative
> short-hand that Codd used to say, "That's stuff I don't care about,
so
> forget about it from now on as you read this paper." I can't think
of any
> other theory where a precise definition is demanded for things that
are
> *intended* not to be discussed. Plane geometry isn't required to
define
> colour for the purpose of excluding it from discussion.
>
> There is nothing in RT that *prevents* values from being divisible,
there
> never was, and it would plainly be stupid to want it that way.

That is as good an explanation of 1NF as I have seen, excellent. Now when Date and Darwen introduce the idea of relation-valued attributes (RVAs), they deliberately provide one data type (or any infinity of them if you like) that violates 1NF, in the sense that the values of an RVA are "of interest or use within the theory". Is this what leads you to say (in another post above): "BTW, I suspect it IS worse, for reasons I only dimly suspect at the moment"?

I suppose the justification is that RVAs are the one (class of) data type that it falls within the remit of an RDBMS to operate on: - Numbers: require mathematical operators (not RDBMS's problem) - Character string: require string manipulation operators (not RDBMS's problem)
- Dates: require date manipulation operators (not RDBMS's problem) - Ellipses: require ellipse manipulation operators (not RDBMS's problem)
- Relations: require an RDBMS! Received on Sun Feb 06 2005 - 15:32:22 CET