Re: A good book

"Cimode" <cimode_at_hotmail.com> wrote in message news:1152270169.948260.305470_at_b28g2000cwb.googlegroups.com...

Cimode wrote:
> The objectives of
> BUSINESS MODELING FOR DATABASE DESIGN
Cimode has failed to point out that this is not a book but a paper, available only from www.dbdebunk.com/page/page/764907.htm.

I am certain it is informative and to be highly recommended, but I doubt it offers the kind of epiphany the OP seems to be looking for. I am certain such a book could be written, but I am also certain it hasn't been. I have been a long-time reader of the relational literature (over 20 years) but the nearest I came to a pulse-racing aha! moment was when I was experimenting with transition constraints, jury-rigged using bits of SQL. It was butt-ugly but I suddenly realized just how much application code could be dispensed with. But then again, many of my students have an aha! moment when I show them the set-oriented solution compared with the row-oriented solution to the same problem. I suspect therefore that what it takes to impress you will depend on where you are starting from. If you don't realize you have a big problem, a solution is unlikely to be amazing.

I am reminded of the story, supposedly from the early part of the 20th century, of an army officer showing a desert sheik an aeroplane doing aerobatics. "Isn't it amazing?", asked the army officer, clearly impressed himself. To which the sheik asked, "Isn't it supposed to do that?"

Roy

PS: I wonder if the best prospect for the OP's kind of book would be an O'Reilly "Dataphor in a Nutshell"! :-) Received on Fri Jul 07 2006 - 13:47:54 CEST