Re: Transactions: good or bad?

"Todd Bandrowsky" <anakin_at_unitedsoftworks.com> wrote in message news:af3d9224.0306081935.7a35edfa_at_posting.google.com...
> > Pointers? This is not clear yet.
> >
> > Anyway I think you are confusing transactions with data models. Don't.
> >
> > And the state-of-the-art is not transactions, but database assignment.
> > So don't loose your time beating a dead horse.
> >
> > Your are messing relational with SQL. SQL is not relational, and has
> > never been. Relational were QUEL, BS12, and now are Opus, Duro,
Dataphor
> > and perhaps LEAP.
>
> It's relational enough for the marketplace. You know, the people that
> actually pay the bills.

So you hope, but those are the same people he is trying to educate.

> > You are messing testing with proving. Test cases aren't formal proofs.
> > I hope I don't have to use software coded by you for reliability.
>
> And what is a formal proof but test cases based on internal
> consistency?

A proof, an axiomatized derivation of fact, where the derived fact is 100% certainty of correct outcome. Contrast this with a random hodge-podge of inputs for which one has verified outcome.

> Turing PROVED they are all the same.

Really? Would you care to point us to this alleged proof?

> > Most customers don't understand what relational is all about. Most
> > customers can't define what's relational any more than they can define
> > what Mathematics is.
>
> Well then, what good is it?

It is as good as any other strategic weapon of which the general population have profound ignorance.

> > You could be clearer, but since you are confusing relational with SQL it
> > probably doesn't matter anyway.
>
> Feature: better performance bounds for the given environment

One cannot provide better performance than the best performance. You are simply making an unfounded claim.

> Feature: a language easier to use for commodities

Really? It is easier to use something other than mathematics for commodities?!? What attributes of mathematics make it unusable for commodities?

> Look, if whatever there is to relational databases that SQL doesn't
> have was so incredibly useful, then, don't you think IBM, Oracle, or
> Microsoft might have been able to implement or sell them?

They have been able to all along. In face, IBM implemented a couple of them already. However, these companies find it more convenient to make billions off of inferior technology than to educate an ignorant market. In fact, these companies find it more convenient to encourage the ignorance.

> What is one feature a purely relational database has, that a SQL
> Server or an Oracle does not?

Logical identity. Physical independence. Logical independence. A well-founded robust user-defined type system. Oh shit, that's four already! Sorry, I cannot do it. Received on Tue Jun 10 2003 - 02:37:17 CEST