Re: Example of expression bias?

J M Davitt wrote:
> Cimode wrote:
> > Tony D wrote:
> >
> >>Cimode wrote:
> >>
>
> [snip]
>
> > Data types are not orthogonal to relations (in RM), they are integral
> > part of RM definition.
>
> This statement is completely contrary to conventional thinking and
> requires elaboration. What do you mean when you say this?
I refer to RM thinking as pure applied math which how it has relevance, not conventional thinking (for whatever that may be).

I mean that data type definition are an inherent part of RM necessity for distinguishing domains and data types. They relationship (possible value vs permissible values) is a part of RM. I refuse the term *othogonal* because it implies *total independence* and negates the relationship of *derivability* of domain and data types.

> > If you don't see usefulness into domain derivability and mutualization
> > at logical level, you will probably will never understand what I am
> > speaking of.
>
> These paragraphs were juxtaposed; are you making a distinction between
> data types and domains? If so, your use of domain needs elaboration,
> too.

At logical level, a data type definition allows to define a sub domain with proper specific caracteristics. Than sub domain and its caracteristics can become a new pool of new *possible* values (to insure FK integrity).

> [snip]
Received on Wed Jun 21 2006 - 03:56:42 CDT