# Re: Date's First Great Blunderhcgjy

Date: Tue, 20 Apr 2004 13:18:47 -0500

Message-ID: <c63pic$dke$1_at_news.netins.net>

"Paul G. Brown" <paul_geoffrey_brown_at_yahoo.com> wrote in message
news:57da7b56.0404200917.742ca28_at_posting.google.com...

*> lauri.pietarinen_at_atbusiness.com (Lauri Pietarinen) wrote in message
*

news:<e9d83568.0404200114.77dd445d_at_posting.google.com>...

*>
*

> [ snip ]

*>
**> > It's like negative integers, rational and real numbers
**> > actually building ontop of the foundation of positive integers instead
**> > of claiming that they are useless.
**>
**> When you see someone apply the following syllogism, you know that
**> they're at the end of their rope.
**>
**> 1. Newtonian mechanics was superceded by relativistic mechanics; the
**> later involving more complicated math.
**>
**> 2. My model is more complicated than yours.
**>
**> 3. Therefore, my model supercedes yours.
**>
**> This misses the truly profound thing about Isaac's contributions. He
*

was

> among the first scientists to apply mathematics with rigor and

consistency

> to answer the question "How does the world work?". Einstein (et al)

used

> Newton's big idea, that the world was amenable to mathematical

descriptions.

*>
*

> Given the originality of Newton's contribution, we ought to be

comfortable

> comparing Codd -- who applied mathematics to the question of how to

manage

> data -- with the former commisioner of the Royal Mint.

*>
**> I mean, compare the way some of the ideas under review are described
**> in this forum with:
**>
**> http://www.win.ua.ac.be/~hidders/pubs/dbpl2003-sat-xpath.pdf
**>
**> This (by our own Jan Hidders) is a first class piece of work.
*

Yes, indeed, Jan does do some very good work in this area.

> What it

*> does is to explain some really hard problems that need to be addressed
*

by

> people who build products based around XQuery/XPath. It is the

embodiment

> of rigor and (to my mind) shows just what happens when you whack

*> loosey-goosey thinking with the big club of formal reasoning.
*

If I didn't have a background in formal reasoning, I wouldn't play so fast and loose with it ;-)

If only the Biblical book of John were presented to us with formal reasoning, then we would all understand the author's intent and agree on all points? Would we all understand each other better if we put language into a mathematical model and then formed our sentences to comply?

Also, in debating whether a particular formal model makes a difference (or where it is useful), one needs to step outside of that model to discuss it.

Nevertheless, it is a good idea to map an approach such as XML to a more rigorous model as it can help us learn more about how to "do it right" even if it did not originate out of any formal reasoning effort. As good as it is to learn from a mathematical model of a particular aspect of our work, it is important not to be a slave to the application of mathematical rigor to every aspect of software development.

It is JUST a model, brother, and you just might have picked a poor model for addressing a paritcular problem. --dawn Received on Tue Apr 20 2004 - 20:18:47 CEST