Re: Why are [Database] Mathematicians Crippled ?

Jan
Norbert
Tegiri

> On Sunday, 1 February 2015 19:58:33 UTC+11, Derek Asirvadem wrote:
>
> Well, the RM can't be "extended". End of story.

In case you do not understand what I mean, let me give you an example, from a totally different subject space. Note I am speaking from the high end of the Relational Database world, the physical universe.

I trust that you know, at least in reasonable detail, enough to understand what it is, and the ramifications, the "work" of Date, Darwen, and Nikos, re "temporal databases", and "extending" the Relational model.

It is all pig slop. Excreta straight from the backside of a sow.

I don't have a problem with TweedleDee, TweedleDum and Nikos eating pig poop.

I do have a problem when they try to trick humans into doing so.

Sure, it is better than the bull poop that Snotgrass wrote. But it is still poop.

If we did what they say, we would have to extend SQL, add a bunch of "temporal" functions, change the data columns that are stored in databases currently, change programs and SQL code, etc. And accept that all that poo is an "extension" to The Relational Model. Sure if you are ignorant, and you don't know The Relational Model (the Codd one, not the theoreticians' 42 ones, which may well require all that work).

In their infantile state, the way the perceive temporal data, is a digital continuum, that /must/ be represented by bits, in order for their pathetic functions to work. When implemented in the database, that translates to trillions of bits. It is a form of hard-coding. Absurd. Breaks standards. They need to change the definition of Atomic, 1NF, in order to get it across the line.

Er, humans can operate on temporal data that is stored as DATETIME, or VERSION, we do not need to represent INSTANT, DURATION, INTERVAL, in the form of trillions of bits, in order to perform time-wise operations on it.

If you do know The Relational Model, Codd defines temporal capability. Tersely, admittedly. But it is total, complete, and final. We can perform all temporal functions using SQL, now. And the better ones, we have been doing that for twenty years.

Again, I repeat, The Relational Model can't be extended. End of story.

(Sure, there my well be 42 or 53 ever-changing "relational models" in the theoretical void that we are blessed with in the RDB space, that can be "extended", but then you are using a well-established title fraudulently, to obtain some credibility where there is none, and it will cause confusion. So please stop, and start acting honestly, scientifically. Use honest terms, and create your own credibility, don't steal Codd's, don't commit fraud.)

Now get back to the thread and start thinking about WHY database theoreticians are CRIPPLED. You really need to watch what you eat.

Cheers
Derek Received on Sun Feb 01 2015 - 10:42:18 CET