Re: Splus and year 2000

Bill Venables (wvenable@attunga.stats.adelaide.edu.au)
Fri, 30 Jan 1998 14:57:03 +1030


Doug Moog writes:
> > (Mathsoft should take a look at how MS Excel does this sort
> > of thing. It's much more flexible, and would better serve
> > the customers, in my opinion.)
>
> I agree, with the caveat that the serial (Julian) dates in
> Excel are off by one day because they count February 29, 1900,
> and we all know there's no February 29 in century years not
> divisible by 200.

I don't. End of century years are not leap years unless they are
divisible by 400, as I recall.

> This matters because their origin (day 0) is Dec 31, 1899, and
> because SPlus counts the days correctly (no Feb 29, 1900) so
> converting dates between the two will cause problems till you
> figure this out - at least it did for me. I wrote Excel about
> this bug, but apparently MS has bigger bugs to worry about.

I can well believe that, though!

Bill V.

-- 
Bill Venables, Head, Dept of Statistics,    Tel.: +61 8 8303 5418
University of Adelaide,                     Fax.: +61 8 8303 3696
South AUSTRALIA.     5005.   Email: Bill.Venables@adelaide.edu.au