Re: Nested structures
Date: Sun, 17 Sep 2006 13:38:22 GMT
Message-ID: <idcPg.2147$Se.783_at_trndny03>
"paul c" <toledobythesea_at_oohay.ac> wrote in message
news:84UOg.543063$Mn5.530087_at_pd7tw3no...
> Actually, I've been a little distressed lately when thoughtful types
> like JOG and David C talk about flatness or star models. Maybe I'm just
> not catching some in-joke or other. Somebody else, sorry, I forget who,
> mentioned that some of these non-relational techniques can be more
> efficient/elegant from the programmer's point of view for certain
> limited applications whereas a pure RM implementation for such apps
> might be ponderous in terms of its code.
I've been trying (not always successfully) to limit my comments concerning star schema to matters of star schema design, and to skirt questions on the merits or lack of merits of Dimensional Modeling (DM) as such. I don't have enough experience with DM to compare and contrast with RM. And, if I were to attempt such a comparison, it would be between DM and ER modeling.
From a certain point of view, a single star schema differs only trivially from the "one great big table" design. In fact, it's straightforward to create a view that sees a star as one great big table. Whether that's a smart trick or a stupid trick depends on the situation. Once you get into a DW design with multiple stars, things are different. On the question of whether "one great big table" is a good MODEL or not, I think you and I would agree that it isn't.
In any event, I never became so enamoured of RM that I considered it the silver bullet or the holy grail. I have several tricks in my bag. RM is one of them. One of the best. But there's no rule in my book that says that every thing in the bag has to be conformant. Received on Sun Sep 17 2006 - 15:38:22 CEST