Re: A question for Mr. Celko

On Tue, 20 Jul 2004 21:12:21 -0700, John Jacob wrote:
>
> I think I got more out of the other branch than this one. I'm not at all
> sure I see how something could be in the "type engine" and not in the
> "relational engine." Seems to me the relational engine is the
> implementation of the type engine, the two are inseparably connected,
> and if I could model extraction of RVA's in the type system, certainly I
> would have to provide an implementation for that functionality in the
> relational engine.

Hmmm, I apparently haven't made myself clear yet. Let my try to explain it another way. Bascially what I'm saying is that I would like to stay within FOL plus user defined functions. That means two things.

First, there are the domains (d1, d2, ...) and the functions over these domains (f : d1 -> d2, g : d1 x d3 -> d4 , ...) and these can be anything so the optimizer barely knows more then the signature of this multi-sorted algebra. For simplicity we may assume that equality operators and other boolean tests are included in these functions.

Second, there is the relational algebra with the usual suspects (but flat because we want to stay witin FOL) except we can use the functions over the domains. So the selection becomes SELECT[e](R) where e is an expression built up from the domain functions and the fields of R, and returns a boolean. The projection becomes PROJECT[e1,...,en](R) in a similar fashion.

Now in this model there is nothing that stops you from having lists, sets, or whatever in your fields, but you stay within FOL.

> I'm going to have to think a long time about this one, but I'm still not
> swayed that the complexity isn't worth it. "That sounds hard" never
> stopped us before :)

Actually this has already been researched. An often cited paper is the paper by Latha Colby:

 http://citeseer.ist.psu.edu/context/96554/0

(Let me know if you can't get it. I'll see what I can do.)

For something a little more recent:

 http://citeseer.ist.psu.edu/liu94algebraic.html

• Jan Hidders