PDA

View Full Version : Whatever happened to the leap century?

DirkGntly
11-23-2003, 08:56 PM
I understand that we observe leap year every four years to keep our calendars corrected for our actual solar orbit. I also understood that though we make this correction, an additional correction every 400 years was also necessary to keep things accurate. I recall that the year 2000 was supposed to be one of these "leap centuries" wherein February would actually have 30 days, and yet, no 30-day February in 2000.

What happened?

Ringo
11-23-2003, 09:06 PM
It actually works the other way around; we need to drop a leap year day every so often. From here (http://www.mitre.org/tech/y2k/docs/LEAP.html):

If the year is divisible by 100, it's not a leap year UNLESS it is also divisible by 400.

Derleth
11-23-2003, 09:08 PM
Dirk, I've never heard of anything of the sort. 2000 was a leap year because it was evenly divisible by 400, which is the only time a century year can be a leap year.* 1900 wasn't and 2100 won't be by the rules in place now. There's no such thing as a leap century and February cannot have 30 days by the rules now in place.

*This made it somewhat easier for those who labored to make the Y2K bug fizzle. A naive algorithm for determining if a year is leap is to see if it's evenly divisible by four. Since 2000 was divisible by both four and 400, overly-simple code correctly flagged it as a leap year.

DirkGntly
11-23-2003, 09:09 PM
Ok, so what happened? We still had 29 days in Feb. that year, not 28...

Random
11-23-2003, 09:36 PM
Read the responses again, Dirk. I can't explain it any clearer than others already have.

DirkGntly
11-23-2003, 09:45 PM
I just followed up on Ringo's link...that pretty much summed it up.

Thanks!

Polycarp
11-23-2003, 09:46 PM
The rule is fairly simple: A year is a leap year if it is evenly divisible by four, unless it is evenly divisible by 100 but not by 400.

Hence:
1. Every year that is divisible evenly by four is a leap year, unless
2. It is evenly divisible by 100, in which case it is a common year, except
3. Years that are evenly divisible by 400 remain leap years.

There have only been two century leap years under the Gregorian calendar: 1600 and 2000, and 1600 did not apply in Britain or Russia, which remained under the Julian calendar until much later.

Is there another formula to fine-tune the days-per-year figure even more closely? I seem to remember a note that 4000 AD would not be a leap year in order to bring things even more closely in tune with the 365.2422 days per year -- but I won't swear to that.

yabob
11-23-2003, 10:08 PM
...
Is there another formula to fine-tune the days-per-year figure even more closely? I seem to remember a note that 4000 AD would not be a leap year in order to bring things even more closely in tune with the 365.2422 days per year -- but I won't swear to that.

IIRC, the former Soviet Union proposed an added adjustment involving the divisibility of millenial years to take care of the < 1 day per millenium discrepency in the current rule, but the international standards bodies would not agree to it.

Derleth
11-23-2003, 10:14 PM
yabob: Of course they wouldn't approve of it. In Soviet Russia, year leaps you.

(I'm sorry.)