Re: What do Oracle professionals think of Fabian Pascal?
Date: Tue, 23 Dec 2014 08:59:18 -0800 (PST)
Message-ID: <2c6b92ab-dd4d-4adb-8a9b-2c6ee35acd2b_at_googlegroups.com>
On Tuesday, December 23, 2014 7:11:30 AM UTC-8, earth..._at_gmail.com wrote:
> On Saturday, August 27, 2005 5:08:35 AM UTC-5, Paul wrote:
> > I have been visiting the site www.dbdebunk.com recently and have found
> > the musings of the author interesting. Having seen some disasters in
> > my time (in terms of adherence to any sort of reasonable database
> > design), I agree with a lot of what he has to say. I'm not sure that I
> > completely understand his constant criticism of SQL (more reading
> > required maybe?).
> >
> >
> > He is constantly railing against vendor extensions (which, from what
> > I've read, he would see as corruptions) of the Relational Model,
> > however my own thougts would be that a competent professional can say
> > to himself "Right, this isn't fully compliant, but it does what I want
> > quickly and easily, so I'll use it until something better comes
> > along".
> >
> >
> > I would be interested in the opinions of other posters in this group,
> > particulary from those who have posted stuff in the past about the
> > need for the data-management industry to get itself some decent
> > independent standards prevalent in other industries (medical, legal,
> > architecture...).
> >
> >
> >
> > Paul...
> >
> >
>
> SQL was developed independent of EF Codd's mathematical work on relational data management. It violates many relational principles by returning non-sensical data to the naive user. A proper query language based on relational principles would not allow non-sensical result sets.
There is no time component of classic relational theory, so it does make some kind of sense that you would respond poorly to a nine year old thread.
I stand by my comments of 9 and 22 years ago.
I stand by my comments of I-don't-recall-which-century that Joe Celko's SQL puzzles are prima facea evidence that there are mapping difficulties between business problems and relational calculus. Part of that problem is due to the potentially large number of hidden variables in those kinds of problems, part is due to dumb analysis by users. In respect to the latter, Fabian is correct (which I've stated in this thread and elsewhere). My major complaint has always been he blames the wrong people, and therefore comes up with the wrong answers for the real world. Patton slapping the private, so to speak.
There's also some aspect of self-promotion for his way of doing things, it doesn't matter what my opinion of that is (nothing wrong with _some_ self-promotion), "the market" seems to judge it correctly. As a giant invisible face-palm.
jg
-- _at_home.com is bogus. http://www.techradar.com/news/internet/cloud-services/oracle-acquires-offline-marketing-specialist-datalogix-1278232Received on Tue Dec 23 2014 - 17:59:18 CET