- From: Charles McCathieNevile <chaals@opera.com>
- Date: Thu, 30 Aug 2007 03:19:03 +0200
On Wed, 29 Aug 2007 19:16:10 +0200, timeless <timeless at gmail.com> wrote: > On 8/29/07, WeBMartians <webmartians at verizon.net> wrote: >> There is a disagreement between astronomers and historians about how to >> count the years preceding year one; astronomers count the BC >> years astronomically..... For example, in the historical >> practice of counting, the rule of divisibility by 4 revealing the >> Julian leap-years is no longer valid; these years are, 1 BC, 5 BC, >> 9 BC, 13 BC... In the astronomical sequence, however, these leap-years >> are called 0, -4, -8, -12..., and the rule of divisibility by >> four subsists. In this system we can speak, for instance, of the solar >> eclipse of -1203-08-28 (twenty-eighth day of August in the >> year 1204 BC) > I must have missed something. But the calendar changed somewhere > around 1500, which means the leap year calculations can't simply be > done in any way like this. The Gregorian reform that corrected things a bit more, by making every hundredth year that was not also a four-hundredth stop being a leap year. So 2000 was a leap year, but 1900 wasn't. It also shifted the date by acouple of weeks one day (which upset people no end, and during the centuries between then and standardisation introduced lots of interoperability errors). There is not much point trying to calculate dates with a simple algorithm before the gregorian reforms anyway. You need to know which calendar was in use for anywhere you are talking about. In China they switched in 1949, in Russia after the 1917 "October" revolution (whose name is based on using the right calendar), in England and her American colonies in the 1750s, and I forget if the Orthodox churches accepted the shift, and if so when. Anyway leap years and a regular Julian calendar were instituted by Julius Caesar about 2050 years ago, after the 434-day year which came about because the Pontifex Maximus could just decide how many days a year should have to make it fit. The summary is that a simple Gregorian date algorithm will work for a number of cases where people use a western calendar now, and probably in the future, but falls apart in historical use until the adoption of the Gregorian reforms (depends where you are talking about). There are also issues with dating things in other calendars in current use, like the Japanese (not so hard) and the Hejri (which is particularly tricky). The use cases where it works are probably enough to justify doing it for them, but it should just be admitted that the complexity of dating in general means that outside of this set of cases there is no point trying to use the mechanisms being developed here, and developing further mechanisms that would work universally is complex. (As an historian of the middle ages in Europe (which I used to be), I would explicitly avoid using a defined date syntax to give semantics to a date unless I was confident that I could at least name the applicable calendar - otherwise you have a situation where you can apparently make calculations, but they are based on erroneous assumptions that are likely to lead to errors). >> Also, the (politically?) correct way to specify BC is "Before Common >> Era", while AD (Anno Domini) is now CE (Common Era). > > Yes, I would hope not to see AD/BC standardized somewhere. Personally I don't really care much, although I agree with the sentiment behind the name change. The Common Era is less common than it seems though. Japan still uses an imperial system where the year is given by the name of the emperor and the year of his reign, in all kinds of common documents. Islam also has a calendar that is different and is important to practising muslims (even those whose practise is pretty loose), and Judaism has its own calendar (about which I know relatively little). cheers Chaals -- Charles McCathieNevile, Opera Software: Standards Group hablo espa?ol - je parle fran?ais - jeg l?rer norsk chaals at opera.com Catch up: Speed Dial http://opera.com
Received on Wednesday, 29 August 2007 18:19:03 UTC