Yes, people were really stupid back then… ;-P
2007/3/6, Dan Lewis <[EMAIL PROTECTED]>:
Comment in-line
On Tuesday March 06 2007 6:23 am, Johnny Andersson wrote:
> 2007/3/5, Joe Smith <[EMAIL PROTECTED]>:
> > Johnny Andersson wrote:
> > > ... If I do something like 2007-03-05 - 2007-03-03 the answer
> > > should be 2 (days), right? But if I enter 2007-03-05 in A1,
> > > 2007-03-03 in B1and =A1-B1 in C1, and then format C1 as DD, C1
> > > shows 01, not 02 which I think would be more appropriate.
> >
> > Dates are represented as an integer number of days after some
> > starting date. If you subtract two such dates, you are simply
> > subtracting two integers, so the result is an integer number of
> > days between the dates: 39146-39144 = 2. What is does the result
> > "2" now represent?
> >
> > > So why is 0 days = 1899-12-30? ...
>
> Why not?
>
>
>
> Because it would be very convenient if it was 1899-12-31, or even
> more convenient
> if it was -0001-12-31 (but unfortunately, for some strange reason,
> there was no year 0 as they started with year 1,
> which I find incredibly stupid, on the other hand some people think
> that year 0 actually is the
> year -4 so that the last millennium actually started in 1996… This
> is probably quite another story, though).
Remember that numbers were originally used only to count. And
when we count a group of things, we begin with the number 1 not 0.
The concepts of BC and AD as well as CE (now) were based upon the
Roman calendar. Since 0 did (and does) not exist in the Roman numeral
system, neither can a year be designated by 0. The whole concept of
the number 0 came much later.
Dan
> However, I thought about this for a while yesterday, and I found
> one explanation, which could be very wrong or maybe even right:
> Doesn't Excel start with 1899-12-31? And doesn't Excel have that
> leap year bug? If OpenOffice.org handles 1900 as a non leap year
> (as it should), then the start date have to be 1899-12-30 if
> recent dates are going to have the same numbers in Excel and
> OpenOffice.org Calc. If this is true, all dates, from 1900-03-01
> until eternity, will have the same numbers in Excel and
> OpenOffice.org Calc. Examples:
>
> 1900-01-01:
> OpenOffice.org: 2
> Excel: 1
>
> 1900-02-28:
> OpenOffice.org: 60
> Excel: 59
>
> 1900-02-29:
> OpenOffice.org: Doesn't exist
> Excel: 60
>
> 1900-03-01:
> OpenOffice.org: 61
> Excel: 61
>
> 2007-03-06
> OpenOffice.org: 39147
> Excel: 39147
>
> This is just my own theory, can it possibly be true?
>
> > Or just Document_date - Birth_date + 1, formatted as YY. ...
> >
> > No, for the same reason that "2" formatted as "DD" gives you the
> > wrong answer.
>
> That is what the "+ 1" was for, to prevent that from happening.
>
> Of course (back to the original question), if you're entering the
> birth
>
> > date and the document date by hand already, why not just enter
> > the years rather than the exact date. No more problem, just
> > Doc_year - Birth year.
> >
> > <Joe
>
> Just don't forget about leap years. One year in average is
> approximately 365.2425 days:
> When a year divided by 100 is an integer, such as 1800, 1900, 2000
> etc, it's only a leap year if the year divided by 400 also is an
> integer, so 1900 was not a leap year, 2000 and 1600 was. So in 400
> years we have (100 / 4 - 1) + (100 / 4 - 1) + (100 / 4 - 1) + 100 /
> 4 = 100 / 4 - 3 = 97 leap years -> One year = (400 · 365 + 97) /
> 400 = 365.2425 days in average. Not that it matters THAT much…
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
