Re: O-O- that backfired

Daniel Morgan <damorgan_at_x.washington.edu> wrote in message news:<1087442683.523716_at_yasure>...
> Serge Rielau wrote:
>
> >> This may well send a few blue suiters into apoplexy but I'll list just
> >> a few:
> >
> > Is that an accidental acknowledgment of IBMers having brains?
> > Don't worry I'm hardy, got to be, living in Canada.
> >
> > German lesson #1: the meaning of "Einen Bock schiessen"
> >
> >> user defined data types with inheritance and methods
> >
> > First DB2 Datajoiner, then merged into DB2 UDB V7 for LUW
> > I did Q.A. for structured types as a student in Almaden 8 years ago
> > - implemented inline SQL PL for efficient methods - loads of fun,
> > learning SQL from Don Chamberlin himself.
> >
> >> object-relational views
> >
> > First in DB2 UDB V5.2 for LUW.
> > My first born, aren't they beautyful :-)
> >
> >> object tables
> >
> > First in DB2 UDB V5.2 for LUW.
> > Oracle 10g still doesn't seem to have subtables. No UNDER clause...
> >
> >> array processing
> >
> > There you go. Oracle invented the array ;-)
> > One point for participation.
> >
> >> Of which I am very thankful for all.
> >
> > You're welcome. I'm glad you like it. IBM put a lot of love into these.
> > http://www.informatik.uni-trier.de/~ley/db/indices/a-tree/r/Rielau:Serge.html
> >
> >
> > Hi Daniel, my name is Serge, I know first aid, can I help you?
> > Dispatch we have an _unresponsive_ UC-Wa (Extension) teacher!
> > Unbelievable!
> >
> > Either way, I have nothing to add to this thread.
>
> This wasn't about who got there first. It was about non-relational
> extensions to the relational engine. If you are proudly proclaiming
> that DB2 is also non-relational ... then it leads to wondering why
> the person that challenged Mark Townsend did so at all doesn't it.

it took the better part of two decades for the COBOL folk to be ripped kicking and screaming out of the land of VSAM, and into the world of application independent data. all courtesy of Dr. Codd and his twelve kids.

now, the java twinks, in particular though not exclusively, are trying *real hard* to put us back to 1964 where data is defined and accessible only through *their* application. locked up tight. don't even attempt to look at it. think, for just a second: is a data object defined in java code any different from a VSAM record defined in a data division?? of course not; in both cases one must have the application to interpret the byte (unicode) stream. etcetera.

this pisses me off. it should piss off application buyers; it probably would, if only they knew up front. it should piss off data wonks, because it bifurcates data concurrency and integrity. there is no good point to it. we should all be screaming at the twinks: you're naked, and so's your emperor.

to the extent that oracle, ibm, m$ et alia add non-relational crap to empower the java twinks, then they ought to be called on the carpet for it. the data folks are supposed to be smarter than the code folks, when it comes to data. lead, don't follow like moronic lemmings.

Dr. Codd conceived his twelve kids because the hierarchical data structure (calling it a model is too effusive) couldn't handle real world needs.

now the java twinks think they have invented something brave and new in XQUERY, XPATH, and the like. all they've done is make an attempt to build relational navigation on top of a hierarchical data structure. no matter how much they kick and scream, won't work.

to the extent that Don Chamberlin is pushing the XML-in-the-DB crap; he too, should be slapped upside the head. and not everyone (see: pascal, date, etc) thinks that his SQL is such a great accomplishment.<G>

there, i feel so much better.

robert Received on Thu Jun 17 2004 - 11:32:32 CDT