Re: Temporal database - no end date

On Jan 21, 10:22 am, "DBMS_Plumber" <paul_geoffrey_br..._at_yahoo.com> wrote:
>
> > > As near as I can tell, Bob has already addressed any points your
> > > example brings up.
>
> The answer "0.8 quanta" is not possible in the Date / Darwen /
> Lorentzos system, because (quoting them here - I'm the only one who's
> read the damn book, apparently) "Formally, however, time points are
> indeed points - they are indivisble, and the concept of duration
> strictly does not apply." You can't have "0.8" of a time quantum,
> because as DDL explicitly state, quantum are "indivisible".

Bob already refuted that argument. 1.6 children, remember? You complain a lot about other people not responding to your points, but in fact all of yours have been addressed and I can't recall you addressing any of mine, nor Bob's.

As I've said before, the topic for me is not the Lorentzos book. Haven't read it; have no opinions on it. And what is up with you that you have a hard time with the concept of 4/5th of a day, sampling of continuous functions, using floating point to approximate the reals, etc? People do this every day and it works very well. I gave a number of examples already, such as floating point, MP3, JPEG, etc.

If you had said that there are minor issues with such, aliasing and accumulated rounding errors and so forth, I would have said sure, but instead you're all "it doesn't work" which is ridiculously overblown.

Another example of dividing the undividable:

I can't think of anything more indivisible than the bit. Can't have less than a bit, can you? I mean, what could a third of a bit even mean? The very idea is ridiculous. And yet...

I used to work on a system that had a character encoding that used 5 1/3 bits per character. That's right: five and one third bits for each character. 2^5.333 ~= 40.32; the character set consisted of 40 characters, each consuming about 5 1/3 bits. If you had 3 characters, that was 16 bits. 5 1/3 * 3 = 16. Amazingly, math with fractional numbers works even on quantities that can't be divided!

> Because you can't use the DDL model to solve a wide variety of
> problems I, Joe, most folk who uses math to analyse time, and
> consequently most of the temporal DBMS research community, are obliged
> to reject the DDL model of time because it isn't a practical framework
> to model temporal operations. It's OK if you want to store information
> about time, but that's only one of the many functions a DBMS provides.

Would you also say that you can't use the IEEE floating point standard to solve a wide variety of problems, and so you, Joe, and most other folk who "uses math" to analyze real-valued quantities are obliged to reject it because it isn't practical? If not, please expain why using
discrete approximations works for some problem domains but not time.

> For making this screamingly obvious point, for quoting the authors
> and examining their ideas, for providing examples to illustrate our
> objections (all of which were ignored), those of us outside the cult on
> this point have been vilified.

You're not making a screamingly obvious point, though. You're claiming a technique that people use all the time doesn't work. You're stubbornly clinging to a wrong notion even on the face of valid counterarguments. You complain about others not addressing your points when they've already been addressed, and in turn you fail to address others' arguments. When people don't accede to your opinions, you claim it is brainwashing or a cult or some kind of deficiency on their part. *That* is why you're being vilified.

Marshall Received on Sun Jan 21 2007 - 20:27:01 CET