Re: Call for an API standard for SQL statements

"Marshall Spight" <mspight_at_dnai.com> wrote in message news:O00fd.2778$R05.1389_at_attbi_s53...
> "Fredrik Bertilsson" <fredrik_bertilsson_at_passagen.se> wrote in message
news:31f7e57d.0410242036.8c3962a_at_posting.google.com...
> > "Marshall Spight" <mspight_at_dnai.com> wrote:
> > > When you say "two dimensional", you mean the result has many
> > > rows and two columns, right?
> > I don't know if you are kidding or if you really don't know the
> > definition of "dimension". A two-dimensional result is a result with
> > many rows and *many* columns.
>
> I was being serious, but mostly I was just being pedantic. It was a bit
> juvenile of me, I admit.
>
> In fact, my definition of dimension is the most technically correct,
> since the dimensionality of a relation is the number of attributes.
> Consider a set of (x,y,z) points: you could record these in a table
> of three attributes, x, y, and z. But you wouldn't call this two
> dimensional data, even if a printout of "select * from points"
> show up as a grid. This data is three dimensional, because each
> dimension of each row/element varies independently.

The whole deal about "star schema" is basically projecting a multidimensional model onto a collection of two dimensional tables. Some people think star schema is juvenile, and some people think it's pedantic. I like it anyway. Received on Mon Oct 25 2004 - 15:09:49 CEST