Re: Temporal database - no end date

"DBMS_Plumber" <paul_geoffrey_brown_at_yahoo.com> wrote in news:1169415152.503098.256960_at_11g2000cwr.googlegroups.com:

>
> NENASHI, Tegiri wrote:

>> I only started to read about the chronon.  The articles that I read
>> say that the application defines the granularity of chronon.  Some
>> one asked why is it that you cannot utilize the granularity of
>> smaller value,  one minute but not one day ?  I am interested to know
>> your opinion. 

It is a very bizarre argument. The Planck constant has been measured with some exactness. It is a very simple formula to produce the Planck time from the Planck constant. The Planck time is a known value. What is the problem ? !!

If you talk of how present the Planck constant, one can use units of needed exactness. You cannot say in serious that one cannot present the Planck constant with the exactness of any number of digits that one can demand. The exactness of the Planck constant is a problem of measurement not presentation.

Your argument is also without foundation in practice. It is hard to find the application in practice that needed the granularity of the Planck time. And if the application had needed the exactness of less than the Planck time, one can very simply take the granularity of the Planck_time/1000000 or even more granular. What is the problem ?

>

>> I work on Oracle and I can affirm that the DATE in Oracle is one
>> second of granularity.  You say that 0.8 of the quantum is
>> impossible, 0.8 of the day is impossible but 0.8 of 24*3600 is
>> possible.  1/7 of 1000 is not exact (142) but the error is < 0.1 per
>> cent.  Why one can not select the granularity of chronon to achieve
>> the exactness ?  You can say, go and use floating point,  but
>> floating point is not exact too. 

What implementation ? I said that Oracle has the type DATE that is like the notion of chronon. The notion of chronon is logical. The quantum of time is one second like the illustration of the fact that people use "the chronon of Oracle" with success. Oracle is not a temporal database and it does not have the method of change the quantum of time that is built-in.

>
> First, get the logical model right. Then worry about the
> engineering
> compromises needed by the implementation.

It is the statement that I do not understand. What model ? I said that Oracle has the chronon of one second only like the illustration of the fact that the chronon of one secod is used by people with success for decades in Oracle. In the cases where the quantum of time must be more granular people develop the "chronon" by hand, and not use the automatic type of DATE.

>

>> I read about Snodgrass,  and Snodgrass,  he uses the name of chronon
>> too. I want to know why you think that the chronon is bad in
>> mathematic sense. 

Why it is an inappropriate model ? Please give one example. I think your "mean" example is not good because you choosed a chronon of bad granularity. If you need more exactness you must choose a chronon of more granularity. Take three distances of one, two and two kilometres. The "quantum of space" is one kilometre. The average distance is (1+2+ 2)/3 = 1 kilometre. What is the problem ? The space quantum is not very granular. Take one millimeter and you solve the problem. See ?

If you think your "mean" example is good I do not have anymore to say because the problem is very very simple and you either do a hoax, pull the leg, or do not have simple mathematic knowledge to understand the simple problem.

The model of chronons can have problems but the granularity, it is certainly not a problem.

--
Tegi



>