Re: Entity and Identity

Nilone wrote:
...
> Yes, that's what I was looking for. Function composition is join, and
> then function application would be selection, right? The domain of
> the function is the propositional condition that restricts the
> selection, while the codomain would be the projection of the resulting
> set of tuples. Recursive queries would substitute the result set into
> the the selection condition (of course they would have to be the same
> type). Recursively defined relations would produce infinite sets.
> Any more references or information in this line?
>
> I looked at some pages on relation algebra, but it's still beyond me,
> I'll keep working at it, thanks for the info.

See TTM Appendix A (pdf is online at ttm site) for concise explanations that reduce composition and selection to combinations of join and project, from a relational algebra viewpoint, it includes very tidy formal definitions. Received on Wed Jul 22 2009 - 23:53:02 CEST