Re: A good book

J M Davitt <jdavitt_at_aeneas.net> wrote:
> Well, there are databases and there are databases. I suppose
> a collection of business cards could be considered a database.
> But in the world of c.d.t, relational theory is the only
> foundation of on which any database should be built.

Okay, after taking a day or so to step back and look at this...

The problem is that I'm looking for information on the connection between relational databases and the theory of relations. I am interested in examples of how the mathematical properties of relations are actually used to solve real-world software problems in databases. I am quite familiar with first-order predicate calculus, and with relations and their properties, as the words are used by mathematicians (or, at least, the several different ways; as the definitions differ from author to author)... I am also familiar with uses of SQL, with rules of thumb for good database designs, and with normalization of databases in a practical sense. I am looking for a source that actually goes through developing the theory that connects the relational model (database) side of this to the practical side.

It turns out there was a little bit of that in Codd's paper (in the section on redundancy and consistency), but not nearly so much as I hoped. That appears to be intentional given the publication -- CACM is more of an overview journal than a really technical -- and the general tone. Perhaps Codd did that work elsewhere. In any case, I'm sure I'll find it eventually.

-- 
Chris Smith - Lead Software Developer / Technical Trainer
MindIQ Corporation