Re: Dawn doesn't like 1NF

"Laconic2" <laconic2_at_comcast.net> wrote in message news:pI2dncO9sK7WpvjcRVn-uQ_at_comcast.com... <snip>

> But the limitation to simple attributes has no such counterpart in the
> mathematics of relations. In particular, a domain can be a domain of
> relations in math. Not in the relational data model of 1970.
>
> When the commercial "relational DBMS" systems came out in the 1980s, all
of
> them required that table columns be "simple". Even a blob is "simple"
from
> the point of view of the DBMS. From the DBMS point of view, it's just a
> string bytes with no discernable substructure. Sometime since 1990, both
> the theory and the products have relaxed this rule. But it was definitely
> there when I began to work with relational databases. Dawn can speak for
> Dawn.

Yes -- correct on the history & mathematical background as well as the last sentence. I'm just a tad bit over extended, but definitely plan to jump in on this within the next couple of days.

A mathematical relation may certainly have a relation as an element, as well as a collection, an ordered tuple, etc. The non-mathematical statement that entered into the theory was the statement that since it is simpler to exclude non-simple types, then if we can do everything with relations by excluding such types, then we ought to take this route in order to have simpler mathematics. That is NOT a statement from within mathematics.

As I suspect I've said before -- I can use the "mathematical model" of a point as a metaphor/model for God or I could use a triangle. Each of these metaphors (which is what a mathematical model is) gives some information. The fact that a point is simpler than a triangle does not make it the better mathematical model for God. Simpler mathematics does not mean better model, period.

Additionally, what ought we to simplify -- the mathematics? NO! For reduced errors we want to make data modeling, design, and use easier (than it is today, perhaps?) It might require very complex mathematics to simplify the use of databases by humans. So, we don't want to perpetuate the old 1NF concepts by continued use of SQL-92 and like products that assume no relations or even lists (ordered tuples) within our data models.

I'll try to catch up with the other postings here and pitch in again -- I'd like everyone to agree with me on this one. ;-) Cheers! --dawn Received on Mon Oct 11 2004 - 04:24:32 CEST